Search
1000 results for “Flip_Switch”
-
Flip the coin, I'm a gay switch 🔄
#TeamVers #nsfw #superman #TomOfFinland #LeatherSling #Humanpup #pupBoner #BLUF #gear365 #recon #scruff #growlr #gaymastodon #gaybdsm #gayrubber #fetlife #OnlyFans #jff #LeatherGayMan #MuirCap #GayPorn #Batman #GayBDSM #CumSlut #GayIrish #HumpOfTheWeek #GayChastity #Gay #NoFearInGear #WeAreFetish #headspace #dominate #submissive #RACK #dainese #gaybiker
-
Flip the coin, I'm a gay switch 🔄
#TeamVers #nsfw #superman #TomOfFinland #LeatherSling #Humanpup #pupBoner #BLUF #gear365 #recon #scruff #growlr #gaymastodon #gaybdsm #gayrubber #fetlife #OnlyFans #jff #LeatherGayMan #MuirCap #GayPorn #Batman #GayBDSM #CumSlut #GayIrish #HumpOfTheWeek #GayChastity #Gay #NoFearInGear #WeAreFetish #headspace #dominate #submissive #RACK #dainese #gaybiker
-
Mini Listening Post
One of the most fascinating pieces of electronic artwork I’ve seen is Mark Hansen and Ben Rubin’s Listening Post when it was on display at the Science Museum, London.
Designed in the early 2000s it was on display from around 2008 for a number of years, but unfortunately it has been packed away in storage amongst the museum’s collections now for some time.
https://www.youtube.com/watch?v=Rzfnndd9fCk
You can read about it on the following links:
- https://www.earstudio.com/projects/project-page/listening-post
- https://archive.org/details/markhansenbenrub1624hans
- https://en.wikipedia.org/wiki/Listening_Post_(artwork)
- https://collection.sciencemuseumgroup.org.uk/objects/co8091687/listening-post
The original has 3 PCs, a Mac and 231 (11 rows of 21) embedded modules each with a four-line text display. As well as the visuals there is a complex surround audio experience of sound and synthesized speech coming out of 10 or more speakers. Each embedded module also has a clicking relay which can create the impression of an older flip display working.
The videos that can be found online really don’t do it justice. There is nothing quite like sitting in front of it with the sound coming at you from all sides.
By the way, it is called Listening Post as it taps into live forums and IRC and pulls out text for the display. It is literally showing a snapshot of the global conversations as they happen – swearing and all. In fact, I recall a “parental controls” warning prior to going into the display, to that effect.
I’ve often wondered if a small part of it could be reproduced in a simpler form, so have started a series of blog posts listing some experiments working towards that idea.
The original is pretty amazing and a serious implementation, with a huge does of imagination and creativity. It would take an awful lot of effort to get anywhere close to something similar, so I’m not aiming for anything even close to that. But I’m going to keep fiddling around on and off to see if I can get something a lot more limited going at a smaller scale that might at least be considered to be in the spirit of the original somehow.
Warning! I strongly recommend using old or second hand equipment for your experiments. I am not responsible for any damage to expensive instruments!
If you are new to microcontrollers, see the Getting Started pages.
The Display Modules
My starting point is my Waveshare Zero Multi Display PCB Design and Expander Board.
The original has 231 small embedded modules, each with a Vacuum Fluorescent Display (VFD) that displays the text. There are two modes for the text: a four-line smaller display, and a large character scrolling display. Each of these options seems to be a function of the type of display in use, which I believe to be a VFD 420.
There is a data sheet and programmers guide for the VFD 420 here:
The VFD 420’s normal display mode looks like this:
The programmers guide describes the “big character” mode as follows:
“VFD-420’s “Big-character” mode can generate full-screen-height numbers and letters (uppercase only)—up to four letters/digits per screen. Big-mode symbols—consisting of dot/decimal point (.), hyphen/minus sign (-), colon (:) and space—take up one-half the width of other “big” characters.”
It does seem possible to buy similar displays today, but what I have seen so far are very, very expensive! I did temporarily wonder about the 4-line version of the HD44780 20×2 character displays. I might still come back to that at some point.
But for now, the trick for me is to attempt to reproduce the look and feel of the VFD big character display with my ST7789 SPI displays used with my Waveshare Zero Multi Display.
I spent a bit of time studying various videos of the Listening Post in action and started to sketch out the main characters that I’d require to rebuild the big character font.
It looks like the originals have a 4×4 matrix of small characters per big character. And each small character is a 8×5 grid and of course there are 20 of these for each of the four lines of the display.
I decided I could probably get away with 10 characters in total.
The Code
I’ve opted for a class that implements the “large text LCD” that effectively drives the TFT display. There are functions for each of the 10 character patterns shown above. Each of these uses the primitives from the Adafruit GFX library to construct the 10 glyphs using filled rectangles or sequences of lines. For example:
// "bottom left triangle"
void CLCD2004::botLTri (uint16_t x, uint16_t y, uint16_t colour) {
uint16_t xlen = 0;
for (int i=1; i<CHAR_H; i++) {
m_pdisplay->drawFastHLine(1+x*CHAR_W, i+y*CHAR_H, xlen/10, colour);
xlen = xlen + (CHAR_W*10-1)/CHAR_H;
}
}From there, I’ve built a set of functions that look up which characters are required for each bt character and gets them on the display one row at a time.
void CLCD2004::printbigchar (char c, uint16_t x, uint16_t colour) {
c = c - LCD2004FONT_START;
printglyph (lcd2004font[c][0], x*BIGCHAR_X, 0, colour);
printglyph (lcd2004font[c][1], x*BIGCHAR_X, 1, colour);
printglyph (lcd2004font[c][2], x*BIGCHAR_X, 2, colour);
printglyph (lcd2004font[c][3], x*BIGCHAR_X, 3, colour);
}
void CLCD2004::printglyph (uint16_t g, uint16_t x, uint16_t y, uint16_t colour) {
if (g & 0xF000) {
printchar('0' + (g>>12 & 0xF), x, y, colour);
}
if (g & 0x0F00) {
printchar('0' + (g>>8 & 0xF), x+1, y, colour);
}
if (g & 0x00F0) {
printchar('0' + (g>>4 & 0xF), x+2, y, colour);
}
if (g & 0x000F) {
printchar('0' + (g & 0xF), x+3, y, colour);
}
}
void CLCD2004::printchar (char c, uint16_t x, uint16_t y, uint16_t colour) {
switch (c) {
... case '0' to '5' ...
case '6':
botLTri(x, y, colour);
break;
... case '7' to '9' ...
}
}There is a font definition that has each block of four characters stored as a 32-bit value for all ASCII characters from 0 through to Z.
Everywhere a 6 occurs in the table, that will trigger the above botLTri() function to draw a “bottom, left, triangle”.
const uint16_t lcd2004font[LCD2004FONT_SIZE][4] = {
{0x7446,0x1001,0x1001,0x9558}, // 0
{0x0720,0x0320,0x0320,0x0110}, // 1
{0x7446,0x0078,0x0780,0x7111}, // 2
...
{0x0760,0x7896,0x1441,0x1001}, // A
{0x1446,0x1558,0x1446,0x1558}, // B
{0x7446,0x1000,0x1000,0x9558}, // C
...
{0x4441,0x0078,0x7800,0x1555}, // Z
};Finally I can now wrap this all up in a general print routine.
void CLCD2004::print (char *c) {
uint16_t xc = 0;
while ((c[xc] != 0) && (xc < MAX_BIGCHAR_X)) {
printbigchar (' ', xc, m_bg);
printbigchar (c[xc], xc, m_fg);
xc++;
}
}Using this class is now a case of setting up the TFT display and LCD objects for each screen.
SPIClass MySPI(FSPI);
Adafruit_ST7735 tft1 = Adafruit_ST7735(&MySPI, TFT_CS1, TFT_DC, TFT_RST);
CLCD2004 lcd((Adafruit_ST77xx*)&tft1, ST_WHITE, ST_BLACK);
void setup() {
MySPI.begin(SPI_SCLK, SPI_MISO, SPI_MOSI, SPI_SS);
tft1.initR(TFT_TYPE);
tft1.setSPISpeed(SPI_SPEED);
tft1.setRotation(3);
tft1.fillScreen(ST_BLACK);
}
void loop() {
lcd.setColour(ST_CYAN, ST_BLACK);
lcd.printclr();
lcd.print("ABCD");
}To create the scrolling is a bit more complex. To do that I decided to add the following:
- A “scrolled” parameter to the display.
- Use a GFX canvas to maintain a pixel bitmap in memory and write to that.
- Add an update() function to draw the bitmap over to the display.
- Expand the print functions to accept a longer string that is stored in a buffer ready to be printed to the canvas.
- Add a scroll function that determines where in the string the updating/printing starts.
There were a few things I had to sort out to get this working.
- The writing of an RGB bitmap to the display was really slow. This is because the provided function has additional SPI writes per pixel that could actually be done just one. This resulting in my writing my own more optimised version of drawRGBBitmap.
- It turns out that setting the SPI speed has to be performed after the initialisation of the display object (i.e. call setSPISpeed() after calling initR()).
- Getting the correct “windowing” to allow the scrolling effect was quite tricky.
- As the display shows four big characters, I always print five on the canvas and let the scroll work, via the windowing, for a whole character. Then a new set of five characters gets printed, and the scroll position resets again.
- The fastest supported SPI speed for the ESP32S3 is 80MHz and it often works at that speed. But occasionally one of the displays will be a bit glitchy, so a slower speed is advisable. I’ve been using 60MHz. I don’t know if that translates to a legitimate SPI speed or not, so the driver may be rounding that to something more sensible – I don’t know!
The basic operation is now:
Set up SPI, TFT and the LCD.
Print a character string to the LCD.
Calling update() will put the first part of the string on the display.
Calling scroll() will shift it as required.The scroll() function has an option to wrap back to the first character in the string if required.
The Results so far
It has taken a bit of tweaking but I believe I can get useful performance out of a single ESP32S3 driving eight SPI displays.
The original has many display modes, so I went with starting to model one of the simpler modes – it searches for the phrase “I am” and then displays what follows. This mode can be seen in action in the video at the end of this post.
For now, to show the display as potential, I’m just printing the two words “I AM”. This video shows the hard-coded “I AM” display in action with a randomised start time and scroll speed for each display.
https://makertube.net/w/hnxY9deqZm1CWCufjS6A4Z
Closing Thoughts
As I say, the original has many display modes, but there is something quite hypnotic about the “I AM” mode that I like. It would be nice to find a way to hook this up to Mastodon searching for uses of the phrase.
So this is starting small, but getting the font in an acceptable format is quite a key part of this project doing anything useful for me. I think this is a really useful start.
Kevin
-
Mini Listening Post
One of the most fascinating pieces of electronic artwork I’ve seen is Mark Hansen and Ben Rubin’s Listening Post when it was on display at the Science Museum, London.
Designed in the early 2000s it was on display from around 2008 for a number of years, but unfortunately it has been packed away in storage amongst the museum’s collections now for some time.
https://www.youtube.com/watch?v=Rzfnndd9fCk
You can read about it on the following links:
- https://www.earstudio.com/projects/project-page/listening-post
- https://archive.org/details/markhansenbenrub1624hans
- https://en.wikipedia.org/wiki/Listening_Post_(artwork)
- https://collection.sciencemuseumgroup.org.uk/objects/co8091687/listening-post
The original has 3 PCs, a Mac and 231 (11 rows of 21) embedded modules each with a four-line text display. As well as the visuals there is a complex surround audio experience of sound and synthesized speech coming out of 10 or more speakers. Each embedded module also has a clicking relay which can create the impression of an older flip display working.
The videos that can be found online really don’t do it justice. There is nothing quite like sitting in front of it with the sound coming at you from all sides.
By the way, it is called Listening Post as it taps into live forums and IRC and pulls out text for the display. It is literally showing a snapshot of the global conversations as they happen – swearing and all. In fact, I recall a “parental controls” warning prior to going into the display, to that effect.
I’ve often wondered if a small part of it could be reproduced in a simpler form, so have started a series of blog posts listing some experiments working towards that idea.
The original is pretty amazing and a serious implementation, with a huge does of imagination and creativity. It would take an awful lot of effort to get anywhere close to something similar, so I’m not aiming for anything even close to that. But I’m going to keep fiddling around on and off to see if I can get something a lot more limited going at a smaller scale that might at least be considered to be in the spirit of the original somehow.
Warning! I strongly recommend using old or second hand equipment for your experiments. I am not responsible for any damage to expensive instruments!
If you are new to microcontrollers, see the Getting Started pages.
The Display Modules
My starting point is my Waveshare Zero Multi Display PCB Design and Expander Board.
The original has 231 small embedded modules, each with a Vacuum Fluorescent Display (VFD) that displays the text. There are two modes for the text: a four-line smaller display, and a large character scrolling display. Each of these options seems to be a function of the type of display in use, which I believe to be a VFD 420.
There is a data sheet and programmers guide for the VFD 420 here:
The VFD 420’s normal display mode looks like this:
The programmers guide describes the “big character” mode as follows:
“VFD-420’s “Big-character” mode can generate full-screen-height numbers and letters (uppercase only)—up to four letters/digits per screen. Big-mode symbols—consisting of dot/decimal point (.), hyphen/minus sign (-), colon (:) and space—take up one-half the width of other “big” characters.”
It does seem possible to buy similar displays today, but what I have seen so far are very, very expensive! I did temporarily wonder about the 4-line version of the HD44780 20×2 character displays. I might still come back to that at some point.
But for now, the trick for me is to attempt to reproduce the look and feel of the VFD big character display with my ST7789 SPI displays used with my Waveshare Zero Multi Display.
I spent a bit of time studying various videos of the Listening Post in action and started to sketch out the main characters that I’d require to rebuild the big character font.
It looks like the originals have a 4×4 matrix of small characters per big character. And each small character is a 8×5 grid and of course there are 20 of these for each of the four lines of the display.
I decided I could probably get away with 10 characters in total.
The Code
I’ve opted for a class that implements the “large text LCD” that effectively drives the TFT display. There are functions for each of the 10 character patterns shown above. Each of these uses the primitives from the Adafruit GFX library to construct the 10 glyphs using filled rectangles or sequences of lines. For example:
// "bottom left triangle"
void CLCD2004::botLTri (uint16_t x, uint16_t y, uint16_t colour) {
uint16_t xlen = 0;
for (int i=1; i<CHAR_H; i++) {
m_pdisplay->drawFastHLine(1+x*CHAR_W, i+y*CHAR_H, xlen/10, colour);
xlen = xlen + (CHAR_W*10-1)/CHAR_H;
}
}From there, I’ve built a set of functions that look up which characters are required for each bt character and gets them on the display one row at a time.
void CLCD2004::printbigchar (char c, uint16_t x, uint16_t colour) {
c = c - LCD2004FONT_START;
printglyph (lcd2004font[c][0], x*BIGCHAR_X, 0, colour);
printglyph (lcd2004font[c][1], x*BIGCHAR_X, 1, colour);
printglyph (lcd2004font[c][2], x*BIGCHAR_X, 2, colour);
printglyph (lcd2004font[c][3], x*BIGCHAR_X, 3, colour);
}
void CLCD2004::printglyph (uint16_t g, uint16_t x, uint16_t y, uint16_t colour) {
if (g & 0xF000) {
printchar('0' + (g>>12 & 0xF), x, y, colour);
}
if (g & 0x0F00) {
printchar('0' + (g>>8 & 0xF), x+1, y, colour);
}
if (g & 0x00F0) {
printchar('0' + (g>>4 & 0xF), x+2, y, colour);
}
if (g & 0x000F) {
printchar('0' + (g & 0xF), x+3, y, colour);
}
}
void CLCD2004::printchar (char c, uint16_t x, uint16_t y, uint16_t colour) {
switch (c) {
... case '0' to '5' ...
case '6':
botLTri(x, y, colour);
break;
... case '7' to '9' ...
}
}There is a font definition that has each block of four characters stored as a 32-bit value for all ASCII characters from 0 through to Z.
Everywhere a 6 occurs in the table, that will trigger the above botLTri() function to draw a “bottom, left, triangle”.
const uint16_t lcd2004font[LCD2004FONT_SIZE][4] = {
{0x7446,0x1001,0x1001,0x9558}, // 0
{0x0720,0x0320,0x0320,0x0110}, // 1
{0x7446,0x0078,0x0780,0x7111}, // 2
...
{0x0760,0x7896,0x1441,0x1001}, // A
{0x1446,0x1558,0x1446,0x1558}, // B
{0x7446,0x1000,0x1000,0x9558}, // C
...
{0x4441,0x0078,0x7800,0x1555}, // Z
};Finally I can now wrap this all up in a general print routine.
void CLCD2004::print (char *c) {
uint16_t xc = 0;
while ((c[xc] != 0) && (xc < MAX_BIGCHAR_X)) {
printbigchar (' ', xc, m_bg);
printbigchar (c[xc], xc, m_fg);
xc++;
}
}Using this class is now a case of setting up the TFT display and LCD objects for each screen.
SPIClass MySPI(FSPI);
Adafruit_ST7735 tft1 = Adafruit_ST7735(&MySPI, TFT_CS1, TFT_DC, TFT_RST);
CLCD2004 lcd((Adafruit_ST77xx*)&tft1, ST_WHITE, ST_BLACK);
void setup() {
MySPI.begin(SPI_SCLK, SPI_MISO, SPI_MOSI, SPI_SS);
tft1.initR(TFT_TYPE);
tft1.setSPISpeed(SPI_SPEED);
tft1.setRotation(3);
tft1.fillScreen(ST_BLACK);
}
void loop() {
lcd.setColour(ST_CYAN, ST_BLACK);
lcd.printclr();
lcd.print("ABCD");
}To create the scrolling is a bit more complex. To do that I decided to add the following:
- A “scrolled” parameter to the display.
- Use a GFX canvas to maintain a pixel bitmap in memory and write to that.
- Add an update() function to draw the bitmap over to the display.
- Expand the print functions to accept a longer string that is stored in a buffer ready to be printed to the canvas.
- Add a scroll function that determines where in the string the updating/printing starts.
There were a few things I had to sort out to get this working.
- The writing of an RGB bitmap to the display was really slow. This is because the provided function has additional SPI writes per pixel that could actually be done just one. This resulting in my writing my own more optimised version of drawRGBBitmap.
- It turns out that setting the SPI speed has to be performed after the initialisation of the display object (i.e. call setSPISpeed() after calling initR()).
- Getting the correct “windowing” to allow the scrolling effect was quite tricky.
- As the display shows four big characters, I always print five on the canvas and let the scroll work, via the windowing, for a whole character. Then a new set of five characters gets printed, and the scroll position resets again.
- The fastest supported SPI speed for the ESP32S3 is 80MHz and it often works at that speed. But occasionally one of the displays will be a bit glitchy, so a slower speed is advisable. I’ve been using 60MHz. I don’t know if that translates to a legitimate SPI speed or not, so the driver may be rounding that to something more sensible – I don’t know!
The basic operation is now:
Set up SPI, TFT and the LCD.
Print a character string to the LCD.
Calling update() will put the first part of the string on the display.
Calling scroll() will shift it as required.The scroll() function has an option to wrap back to the first character in the string if required.
The Results so far
It has taken a bit of tweaking but I believe I can get useful performance out of a single ESP32S3 driving eight SPI displays.
The original has many display modes, so I went with starting to model one of the simpler modes – it searches for the phrase “I am” and then displays what follows. This mode can be seen in action in the video at the end of this post.
For now, to show the display as potential, I’m just printing the two words “I AM”. This video shows the hard-coded “I AM” display in action with a randomised start time and scroll speed for each display.
https://makertube.net/w/hnxY9deqZm1CWCufjS6A4Z
Closing Thoughts
As I say, the original has many display modes, but there is something quite hypnotic about the “I AM” mode that I like. It would be nice to find a way to hook this up to Mastodon searching for uses of the phrase.
So this is starting small, but getting the font in an acceptable format is quite a key part of this project doing anything useful for me. I think this is a really useful start.
Kevin
-
Watch #gaypornstar #Baxxx #flipfuck #musclehunk #JakeSwitch until they blow their loads
More nude photos -> https://mensparkle.com/blog/2025/12/chaosmen-baxxx-flip-fucks-inked-muscle-hunk-jake-switch/
-
Watch #gaypornstar #Baxxx #flipfuck #musclehunk #JakeSwitch until they blow their loads
More nude photos -> https://mensparkle.com/blog/2025/12/chaosmen-baxxx-flip-fucks-inked-muscle-hunk-jake-switch/
-
Scientists at Monash University say they have found a way to permanently "switch off" genes that help cancer cells thrive. This could help development of a new style of treatment that might be more effective and less difficult for patients. “We have potentially identified a new way to exploit cancer’s weaknesses,” said Dr. Omer Gilan, who worked on the research. Here's more from SciTech Daily.
#Science #Medicine #Cancer #CancerTreatment #Genetics #Epigenetics
-
My wishlist for #apple #wwdc 2025:
SYNC. WITH. ICLOUD. NOW.
I came home from a race. What did I want to do? Sit down for 5 minutes, flip through my photos, and upload them to share with my run club.
What did I do? Sat down at my insanely overpowered (for my uses) Mac Studio. "Photos synced on Friday." (It's Sunday now). I check my phone, and of course the 100+ photos from today have been uploaded since roughly the moment I walked into my garage because I have WiFi 7 APs on a 2.5gbit uplink to my 10gb switches to my 10gb symmetric fiber connection. Not a tough chore, uploading a couple hours' worth of iPhone snaps.
Back to the Mac. Twiddle with buttons, settings, force quit some things, relaunch some things. Nada. Do some Google searches, no luck. Turn iCloud Photos syncing off. Turn it back on, get prompted for my password, type password, setting stays off. Repeat 3 or 4 times until finally System Settings hangs for 60 seconds.
"Optimizing system performance". My system ain't doing shit. Stop optimizing.
"Sync now" button appears. Click it.
"You don't have enough space to store your originals". Yes I do. Or at least, I _did_. Maybe if you've decided to download an entire second copy of my photos library I won't. Who knows!
What's it doing now? God only knows. Piddling along at single digit megabits a second up and down. "syncing". No new photos have appeared in ages. I guess I'll just let it do its thing... for a few hours? Days? God only knows.
Ironically, my unplugged MacBook Air already had the photos when I opened it to write this.
-
We have another new feature available for people to use from the AI Horde. This is the capacity to use Layer Diffuse to generate images with a transparent background directly (as opposed to stripping the image background with a post-processor).
As someone who’s dabbled into video game development in the past (which was in fact the reason I started the AI Horde) being able to generate sprites, icons and other assets can be quite useful, so once I saw this breakthrough, it immediately became something I wanted to support.
To use this feature, you simply need to flip on the transparent switch if your UI supports it, and the Horde will do the rest. If you’re an integrator, simply send “transparent: true” in your payload.
Take note that the images generated by this feature will not match the image you get with the same seed when transparency is not used! Don’t expect to take an image you like and remove the background this way. For that you need to use the post-processor approach.
Also keep in mind, not every prompt will work well for a transparent image generation. Experiment and find what works for you.
As part of making this update work, me and Tazlin also developed, discovered and fixed a number of other issues and bugs.
What would be most interesting for you is a slight change on how hires-fix works. I discovered that the implementation we were using was using the same amount of steps for the upscaled denoising which was completely unnecessary and wasting compute. So we now use a smart system which dynamically determines how many steps to use for the hires-fix based on the denoising strength you used for hires-fix and the steps for the main generation, and we also exposed a new key on the API where you can directly pass a hires-fix denoising strength.
The second fix is allowing hires-fix on SDXL models, so now you can try to generate larger SDXL images at the optimal resolution.
Finally there were a lot of other minor tweaks and fixes, primarily in the horde-engine. You can read further for more development details on this feature.
This update required a significant amount of work as it required that we onboard a new comfyUI node. Normally this isn’t difficult, but it turns out this node was automatically downloading its own LoRa models on startup, and those were not handled properly for either storage or memory. Due to the efficiency of the AI Horde worker, we do a lot of model preloading along with some fancy footwork in regards to RAM/VRAM usage.
So to make the new nodes work as expected, I had to reach in and modify the methods which were downloading models so that they use our internal mechanisms such as the model manager. Sadly the model manager wasn’t aware of strange models like layer diffuse, so it required me adding a new catch-all class of the model manager for all future utility models like these.
While waiting for Tazlin to be happy with the stability of the code, we discovered another major problem: The face-fixer post-processors we were using until now had started malfunctioning, and generating faces with a weird gray sheen. After some significant troubleshooting and investigation, we discovered that ComfyUI itself on the latest version had switched to a different internal library which didn’t play well with the custom nodes doing the face-fixing.
First I decided to update the code of the face-fixer nodes we were using, which is harder than it sounds, as it also downloads models automatically on startup, which again needs to be handled properly. Updating the custom nodes fixed the codeformer face-fixer, but gfpgan remained broken and the comfyUI devs mentioned that someone would have to fix it. Unfortunately those nodes didn’t seem to be actively maintained anymore so there was little hope to just wait for a quick fix.
Fortunately another custom node developer had run into the same problems, and created a bespoke solution for gfpgan licensed liberally, which I could copy. I love FOSS!
In the meantime, through our usual beta testing process, we discovered that there were still some funkiness in the new hires-fix approach, and Tazlin along with some power users of the community were able to tweak things so that they could work more optimally.
All in all, quite a bit of effort in the past month for this feature, but now we provide something which along with the embedded QR Code generation, I’ve seen very few other GenAI services provide, if at all.
Will you use the new transparent image generation? If so, let us know how! And remember if you have a decent GPU, you can help other generate images by adding your PC onto the horde!
-
Something big is happening in AI — and most people will be blindsided – Fortune
Matt Shumer, co-founder and CEO of OthersideAI. courtesy of OthersideAISomething big is happening in AI — and most people will be blindsided
By Matt Shumer, February 11, 2026, 9:22 AM ET
Matt Shumer is the co-founder and CEO of OthersideAI, an applied AI company building the most advanced autocomplete tools in the world, powered by large-scale AI systems like GPT-3. OthersideAI is the company behind HyperWrite, the leading AI autocomplete Chrome extension for consumers. Previously, while in high school, Matt founded Visos, a startup developing next-generation Virtual Reality software designed for medical use, and FURI, a company aiming to disrupt the sporting goods industry by creating high-performance products and selling them for fair prices.
Think back to February 2020.
A few people were talking about a virus spreading overseas. If someone told you they were stockpiling toilet paper you would have thought they’d been spending too much time on a weird corner of the internet. Then, over the course of about three weeks, the entire world changed.
I think we’re in the “this seems overblown” phase of something much, much bigger than Covid.
I’ve spent six years building an AI startup and investing in the space. I live in this world. And I’m writing this for the people in my life who don’t. I keep giving them the polite, cocktail-party version. Because the honest version sounds like I’ve lost my mind. But the gap between what I’ve been saying and what is actually happening has gotten far too big. The people I care about deserve to hear what is coming, even if it sounds crazy.
I should be clear about something up front: even though I work in AI, I have almost no influence over what’s about to happen, and neither does the vast majority of the industry. The future is being shaped by a remarkably small number of people: a few hundred researchers at a handful of companies… OpenAI, Anthropic, Google, DeepMind, and a few others.
Most of us who work in AI are building on top of foundations we didn’t lay. We’re watching this unfold the same as you… we just happen to be close enough to feel the ground shake first.
But it’s time now. Not in an “eventually we should talk about this” way. In a “this is happening right now and I need you to understand it” way.
I know this is real because it happened to me first
Here’s the thing nobody outside of tech quite understands yet: we’re not making predictions. We’re telling you what already occurred in our own jobs, and warning you that you’re next.
For years, AI had been improving steadily. Then in 2025, new techniques for building these models unlocked a much faster pace of progress. This year, something clicked. Not like a light switch… more like the moment you realize the water has been rising around you and is now at your chest.
I am no longer needed for the actual technical work of my job. I describe what I want built, in plain English, and it just… appears. Not a rough draft I need to fix. The finished thing. I tell the AI what I want, walk away from my computer for four hours, and come back to find the work done. Done well, done better than I would have done it myself, with no corrections needed. A couple of months ago, I was going back and forth with the AI, guiding it, making edits. Now I just describe the outcome and leave.
Let me give you an example so you can understand what this actually looks like in practice. I’ll tell the AI: “I want to build this app. Here’s what it should do, here’s roughly what it should look like. Figure out the user flow, the design, all of it.” And it does. It writes tens of thousands of lines of code. Then, and this is the part that would have been unthinkable a year ago, it opens the app itself. It clicks through the buttons. It tests the features. It uses the app the way a person would. If it doesn’t like how something looks or feels, it goes back and changes it, on its own. It iterates, like a developer would, fixing and refining until it’s satisfied. Only once it has decided the app meets its own standards does it come back to me and say: “It’s ready for you to test.” And when I test it, it’s usually perfect.
I’m not exaggerating. That is what my Monday looked like this week.
I’ve always been early to adopt AI tools. But the last few months have shocked me. These new AI models aren’t incremental improvements. This is a different thing entirely.
The experience that tech workers have had over the past year, of watching AI go from “helpful tool” to “does my job better than I do”, is the experience everyone else is about to have. Law, finance, medicine, accounting, consulting, writing, design, analysis, customer service. Not in 10 years. The people building these systems say one to five years. Some say less. The market was spooked enough this month that it wiped out $1 trillion worth of software value in just a week. And given what I’ve seen in just the last couple of months, I see more disruption, and soon.
“But I tried AI and it wasn’t that good”
If you tried ChatGPT in 2023 or early 2024 and thought “this makes stuff up” or “this isn’t that impressive”, you were right. Those early versions were genuinely limited. They hallucinated. They confidently said things that were nonsense.
The models available today are unrecognizable from what existed even six months ago. The debate about whether AI is “really getting better” or “hitting a wall” — which has been going on for over a year — is over. It’s done. Anyone still making that argument either hasn’t used the current models, has an incentive to downplay what’s happening, or is evaluating based on an experience from 2024 that is no longer relevant. I don’t say that to be dismissive. I say it because the gap between public perception and current reality is now enormous, and that gap is dangerous… because it’s preventing people from preparing.
Part of the problem is that most people are using the free version of AI tools. The free version is over a year behind what paying users have access to. Judging AI based on free-tier ChatGPT is like evaluating the state of smartphones by using a flip phone. The people paying for the best tools, and actually using them daily for real work, know what’s coming.
I think of my friend, who’s a lawyer. I keep telling him to try using AI at his firm, and he keeps finding reasons it won’t work. And I get it. But I’ve had partners at major law firms reach out to me for advice, because they’ve tried the current versions and they see where this is going. One of them, the managing partner at a large firm, spends hours every day using AI. He told me it’s like having a team of associates available instantly. He’s not using it because it’s a toy. He’s using it because it works. And he told me something that stuck with me: every couple of months, it gets significantly more capable for his work. He said if it stays on this trajectory, he expects it’ll be able to do most of what he does before long… and he’s a managing partner with decades of experience. He’s not panicking. But he’s paying very close attention.
Think about what that means for your work.
What this means for your job
I’m going to be direct with you because I think you deserve honesty more than comfort.
Dario Amodei, who is probably the most safety-focused CEO in the AI industry, has publicly predicted that AI will eliminate 50% of entry-level white-collar jobs within one to five years. And many people in the industry think he’s being conservative. Given what the latest models can do, the capability for massive disruption could be here by the end of this year. It’ll take some time to ripple through the economy, but the underlying ability is arriving now.
This is different from every previous wave of automation, and I need you to understand why. AI isn’t replacing one specific skill. It’s a general substitute for cognitive work. It gets better at everything simultaneously. When factories automated, a displaced worker could retrain as an office worker. When the internet disrupted retail, workers moved into logistics or services. But AI doesn’t leave a convenient gap to move into. Whatever you retrain for, it’s improving at that too.
I think the honest answer is that nothing that can be done on a computer is safe in the medium term. If your job happens on a screen (if the core of what you do is reading, writing, analyzing, deciding, communicating through a keyboard) then AI is coming for significant parts of it. The timeline isn’t “someday.” It’s already started.
Eventually, robots will handle physical work too. They’re not quite there yet. But “not quite there yet” in AI terms has a way of becoming “here” faster than anyone expects.
What you should actually do
I’m not writing this to make you feel helpless. I’m writing this because I think the single biggest advantage you can have right now is simply being early. Early to understand it. Early to use it. Early to adapt.
Editor’s Note: Read more online. This is a general AI-supportive post by Matt. And there are many others similar “heads up” posts about AI right now. Keep an eye out for further posts on the AI changes coming. Whatever your views, I recommend keeping up with AI over your own work, or job, society changes, regulations, world impacts, and more. –DrWeb
Continue/Read Original Article Here: Something big is happening in AI — and most people will be blindsided | Fortune
Tags: AI, AI Future, artificial intelligence, Big Changes, Fortune, Heads Up, Jobs Changing, Matt Shumer, Newer Models, OthersideAI, Rapid Developments in AI, Social Impacts, Technology
#AI #AIFuture #artificialIntelligence #BigChanges #Fortune #HeadsUp #JobsChanging #MattShumer #NewerModels #OthersideAI #RapidDevelopmentsInAI #SocialImpacts #Technology -
Something big is happening in AI — and most people will be blindsided – Fortune
Matt Shumer, co-founder and CEO of OthersideAI. courtesy of OthersideAISomething big is happening in AI — and most people will be blindsided
By Matt Shumer, February 11, 2026, 9:22 AM ET
Matt Shumer is the co-founder and CEO of OthersideAI, an applied AI company building the most advanced autocomplete tools in the world, powered by large-scale AI systems like GPT-3. OthersideAI is the company behind HyperWrite, the leading AI autocomplete Chrome extension for consumers. Previously, while in high school, Matt founded Visos, a startup developing next-generation Virtual Reality software designed for medical use, and FURI, a company aiming to disrupt the sporting goods industry by creating high-performance products and selling them for fair prices.
Think back to February 2020.
A few people were talking about a virus spreading overseas. If someone told you they were stockpiling toilet paper you would have thought they’d been spending too much time on a weird corner of the internet. Then, over the course of about three weeks, the entire world changed.
I think we’re in the “this seems overblown” phase of something much, much bigger than Covid.
I’ve spent six years building an AI startup and investing in the space. I live in this world. And I’m writing this for the people in my life who don’t. I keep giving them the polite, cocktail-party version. Because the honest version sounds like I’ve lost my mind. But the gap between what I’ve been saying and what is actually happening has gotten far too big. The people I care about deserve to hear what is coming, even if it sounds crazy.
I should be clear about something up front: even though I work in AI, I have almost no influence over what’s about to happen, and neither does the vast majority of the industry. The future is being shaped by a remarkably small number of people: a few hundred researchers at a handful of companies… OpenAI, Anthropic, Google, DeepMind, and a few others.
Most of us who work in AI are building on top of foundations we didn’t lay. We’re watching this unfold the same as you… we just happen to be close enough to feel the ground shake first.
But it’s time now. Not in an “eventually we should talk about this” way. In a “this is happening right now and I need you to understand it” way.
I know this is real because it happened to me first
Here’s the thing nobody outside of tech quite understands yet: we’re not making predictions. We’re telling you what already occurred in our own jobs, and warning you that you’re next.
For years, AI had been improving steadily. Then in 2025, new techniques for building these models unlocked a much faster pace of progress. This year, something clicked. Not like a light switch… more like the moment you realize the water has been rising around you and is now at your chest.
I am no longer needed for the actual technical work of my job. I describe what I want built, in plain English, and it just… appears. Not a rough draft I need to fix. The finished thing. I tell the AI what I want, walk away from my computer for four hours, and come back to find the work done. Done well, done better than I would have done it myself, with no corrections needed. A couple of months ago, I was going back and forth with the AI, guiding it, making edits. Now I just describe the outcome and leave.
Let me give you an example so you can understand what this actually looks like in practice. I’ll tell the AI: “I want to build this app. Here’s what it should do, here’s roughly what it should look like. Figure out the user flow, the design, all of it.” And it does. It writes tens of thousands of lines of code. Then, and this is the part that would have been unthinkable a year ago, it opens the app itself. It clicks through the buttons. It tests the features. It uses the app the way a person would. If it doesn’t like how something looks or feels, it goes back and changes it, on its own. It iterates, like a developer would, fixing and refining until it’s satisfied. Only once it has decided the app meets its own standards does it come back to me and say: “It’s ready for you to test.” And when I test it, it’s usually perfect.
I’m not exaggerating. That is what my Monday looked like this week.
I’ve always been early to adopt AI tools. But the last few months have shocked me. These new AI models aren’t incremental improvements. This is a different thing entirely.
The experience that tech workers have had over the past year, of watching AI go from “helpful tool” to “does my job better than I do”, is the experience everyone else is about to have. Law, finance, medicine, accounting, consulting, writing, design, analysis, customer service. Not in 10 years. The people building these systems say one to five years. Some say less. The market was spooked enough this month that it wiped out $1 trillion worth of software value in just a week. And given what I’ve seen in just the last couple of months, I see more disruption, and soon.
“But I tried AI and it wasn’t that good”
If you tried ChatGPT in 2023 or early 2024 and thought “this makes stuff up” or “this isn’t that impressive”, you were right. Those early versions were genuinely limited. They hallucinated. They confidently said things that were nonsense.
The models available today are unrecognizable from what existed even six months ago. The debate about whether AI is “really getting better” or “hitting a wall” — which has been going on for over a year — is over. It’s done. Anyone still making that argument either hasn’t used the current models, has an incentive to downplay what’s happening, or is evaluating based on an experience from 2024 that is no longer relevant. I don’t say that to be dismissive. I say it because the gap between public perception and current reality is now enormous, and that gap is dangerous… because it’s preventing people from preparing.
Part of the problem is that most people are using the free version of AI tools. The free version is over a year behind what paying users have access to. Judging AI based on free-tier ChatGPT is like evaluating the state of smartphones by using a flip phone. The people paying for the best tools, and actually using them daily for real work, know what’s coming.
I think of my friend, who’s a lawyer. I keep telling him to try using AI at his firm, and he keeps finding reasons it won’t work. And I get it. But I’ve had partners at major law firms reach out to me for advice, because they’ve tried the current versions and they see where this is going. One of them, the managing partner at a large firm, spends hours every day using AI. He told me it’s like having a team of associates available instantly. He’s not using it because it’s a toy. He’s using it because it works. And he told me something that stuck with me: every couple of months, it gets significantly more capable for his work. He said if it stays on this trajectory, he expects it’ll be able to do most of what he does before long… and he’s a managing partner with decades of experience. He’s not panicking. But he’s paying very close attention.
Think about what that means for your work.
What this means for your job
I’m going to be direct with you because I think you deserve honesty more than comfort.
Dario Amodei, who is probably the most safety-focused CEO in the AI industry, has publicly predicted that AI will eliminate 50% of entry-level white-collar jobs within one to five years. And many people in the industry think he’s being conservative. Given what the latest models can do, the capability for massive disruption could be here by the end of this year. It’ll take some time to ripple through the economy, but the underlying ability is arriving now.
This is different from every previous wave of automation, and I need you to understand why. AI isn’t replacing one specific skill. It’s a general substitute for cognitive work. It gets better at everything simultaneously. When factories automated, a displaced worker could retrain as an office worker. When the internet disrupted retail, workers moved into logistics or services. But AI doesn’t leave a convenient gap to move into. Whatever you retrain for, it’s improving at that too.
I think the honest answer is that nothing that can be done on a computer is safe in the medium term. If your job happens on a screen (if the core of what you do is reading, writing, analyzing, deciding, communicating through a keyboard) then AI is coming for significant parts of it. The timeline isn’t “someday.” It’s already started.
Eventually, robots will handle physical work too. They’re not quite there yet. But “not quite there yet” in AI terms has a way of becoming “here” faster than anyone expects.
What you should actually do
I’m not writing this to make you feel helpless. I’m writing this because I think the single biggest advantage you can have right now is simply being early. Early to understand it. Early to use it. Early to adapt.
Editor’s Note: Read more online. This is a general AI-supportive post by Matt. And there are many others similar “heads up” posts about AI right now. Keep an eye out for further posts on the AI changes coming. Whatever your views, I recommend keeping up with AI over your own work, or job, society changes, regulations, world impacts, and more. –DrWeb
Continue/Read Original Article Here: Something big is happening in AI — and most people will be blindsided | Fortune
Tags: AI, AI Future, artificial intelligence, Big Changes, Fortune, Heads Up, Jobs Changing, Matt Shumer, Newer Models, OthersideAI, Rapid Developments in AI, Social Impacts, Technology
#AI #AIFuture #artificialIntelligence #BigChanges #Fortune #HeadsUp #JobsChanging #MattShumer #NewerModels #OthersideAI #RapidDevelopmentsInAI #SocialImpacts #Technology -
Something big is happening in AI — and most people will be blindsided – Fortune
Matt Shumer, co-founder and CEO of OthersideAI. courtesy of OthersideAISomething big is happening in AI — and most people will be blindsided
By Matt Shumer, February 11, 2026, 9:22 AM ET
Matt Shumer is the co-founder and CEO of OthersideAI, an applied AI company building the most advanced autocomplete tools in the world, powered by large-scale AI systems like GPT-3. OthersideAI is the company behind HyperWrite, the leading AI autocomplete Chrome extension for consumers. Previously, while in high school, Matt founded Visos, a startup developing next-generation Virtual Reality software designed for medical use, and FURI, a company aiming to disrupt the sporting goods industry by creating high-performance products and selling them for fair prices.
Think back to February 2020.
A few people were talking about a virus spreading overseas. If someone told you they were stockpiling toilet paper you would have thought they’d been spending too much time on a weird corner of the internet. Then, over the course of about three weeks, the entire world changed.
I think we’re in the “this seems overblown” phase of something much, much bigger than Covid.
I’ve spent six years building an AI startup and investing in the space. I live in this world. And I’m writing this for the people in my life who don’t. I keep giving them the polite, cocktail-party version. Because the honest version sounds like I’ve lost my mind. But the gap between what I’ve been saying and what is actually happening has gotten far too big. The people I care about deserve to hear what is coming, even if it sounds crazy.
I should be clear about something up front: even though I work in AI, I have almost no influence over what’s about to happen, and neither does the vast majority of the industry. The future is being shaped by a remarkably small number of people: a few hundred researchers at a handful of companies… OpenAI, Anthropic, Google, DeepMind, and a few others.
Most of us who work in AI are building on top of foundations we didn’t lay. We’re watching this unfold the same as you… we just happen to be close enough to feel the ground shake first.
But it’s time now. Not in an “eventually we should talk about this” way. In a “this is happening right now and I need you to understand it” way.
I know this is real because it happened to me first
Here’s the thing nobody outside of tech quite understands yet: we’re not making predictions. We’re telling you what already occurred in our own jobs, and warning you that you’re next.
For years, AI had been improving steadily. Then in 2025, new techniques for building these models unlocked a much faster pace of progress. This year, something clicked. Not like a light switch… more like the moment you realize the water has been rising around you and is now at your chest.
I am no longer needed for the actual technical work of my job. I describe what I want built, in plain English, and it just… appears. Not a rough draft I need to fix. The finished thing. I tell the AI what I want, walk away from my computer for four hours, and come back to find the work done. Done well, done better than I would have done it myself, with no corrections needed. A couple of months ago, I was going back and forth with the AI, guiding it, making edits. Now I just describe the outcome and leave.
Let me give you an example so you can understand what this actually looks like in practice. I’ll tell the AI: “I want to build this app. Here’s what it should do, here’s roughly what it should look like. Figure out the user flow, the design, all of it.” And it does. It writes tens of thousands of lines of code. Then, and this is the part that would have been unthinkable a year ago, it opens the app itself. It clicks through the buttons. It tests the features. It uses the app the way a person would. If it doesn’t like how something looks or feels, it goes back and changes it, on its own. It iterates, like a developer would, fixing and refining until it’s satisfied. Only once it has decided the app meets its own standards does it come back to me and say: “It’s ready for you to test.” And when I test it, it’s usually perfect.
I’m not exaggerating. That is what my Monday looked like this week.
I’ve always been early to adopt AI tools. But the last few months have shocked me. These new AI models aren’t incremental improvements. This is a different thing entirely.
The experience that tech workers have had over the past year, of watching AI go from “helpful tool” to “does my job better than I do”, is the experience everyone else is about to have. Law, finance, medicine, accounting, consulting, writing, design, analysis, customer service. Not in 10 years. The people building these systems say one to five years. Some say less. The market was spooked enough this month that it wiped out $1 trillion worth of software value in just a week. And given what I’ve seen in just the last couple of months, I see more disruption, and soon.
“But I tried AI and it wasn’t that good”
If you tried ChatGPT in 2023 or early 2024 and thought “this makes stuff up” or “this isn’t that impressive”, you were right. Those early versions were genuinely limited. They hallucinated. They confidently said things that were nonsense.
The models available today are unrecognizable from what existed even six months ago. The debate about whether AI is “really getting better” or “hitting a wall” — which has been going on for over a year — is over. It’s done. Anyone still making that argument either hasn’t used the current models, has an incentive to downplay what’s happening, or is evaluating based on an experience from 2024 that is no longer relevant. I don’t say that to be dismissive. I say it because the gap between public perception and current reality is now enormous, and that gap is dangerous… because it’s preventing people from preparing.
Part of the problem is that most people are using the free version of AI tools. The free version is over a year behind what paying users have access to. Judging AI based on free-tier ChatGPT is like evaluating the state of smartphones by using a flip phone. The people paying for the best tools, and actually using them daily for real work, know what’s coming.
I think of my friend, who’s a lawyer. I keep telling him to try using AI at his firm, and he keeps finding reasons it won’t work. And I get it. But I’ve had partners at major law firms reach out to me for advice, because they’ve tried the current versions and they see where this is going. One of them, the managing partner at a large firm, spends hours every day using AI. He told me it’s like having a team of associates available instantly. He’s not using it because it’s a toy. He’s using it because it works. And he told me something that stuck with me: every couple of months, it gets significantly more capable for his work. He said if it stays on this trajectory, he expects it’ll be able to do most of what he does before long… and he’s a managing partner with decades of experience. He’s not panicking. But he’s paying very close attention.
Think about what that means for your work.
What this means for your job
I’m going to be direct with you because I think you deserve honesty more than comfort.
Dario Amodei, who is probably the most safety-focused CEO in the AI industry, has publicly predicted that AI will eliminate 50% of entry-level white-collar jobs within one to five years. And many people in the industry think he’s being conservative. Given what the latest models can do, the capability for massive disruption could be here by the end of this year. It’ll take some time to ripple through the economy, but the underlying ability is arriving now.
This is different from every previous wave of automation, and I need you to understand why. AI isn’t replacing one specific skill. It’s a general substitute for cognitive work. It gets better at everything simultaneously. When factories automated, a displaced worker could retrain as an office worker. When the internet disrupted retail, workers moved into logistics or services. But AI doesn’t leave a convenient gap to move into. Whatever you retrain for, it’s improving at that too.
I think the honest answer is that nothing that can be done on a computer is safe in the medium term. If your job happens on a screen (if the core of what you do is reading, writing, analyzing, deciding, communicating through a keyboard) then AI is coming for significant parts of it. The timeline isn’t “someday.” It’s already started.
Eventually, robots will handle physical work too. They’re not quite there yet. But “not quite there yet” in AI terms has a way of becoming “here” faster than anyone expects.
What you should actually do
I’m not writing this to make you feel helpless. I’m writing this because I think the single biggest advantage you can have right now is simply being early. Early to understand it. Early to use it. Early to adapt.
Editor’s Note: Read more online. This is a general AI-supportive post by Matt. And there are many others similar “heads up” posts about AI right now. Keep an eye out for further posts on the AI changes coming. Whatever your views, I recommend keeping up with AI over your own work, or job, society changes, regulations, world impacts, and more. –DrWeb
Continue/Read Original Article Here: Something big is happening in AI — and most people will be blindsided | Fortune
Tags: AI, AI Future, artificial intelligence, Big Changes, Fortune, Heads Up, Jobs Changing, Matt Shumer, Newer Models, OthersideAI, Rapid Developments in AI, Social Impacts, Technology
#AI #AIFuture #artificialIntelligence #BigChanges #Fortune #HeadsUp #JobsChanging #MattShumer #NewerModels #OthersideAI #RapidDevelopmentsInAI #SocialImpacts #Technology -
Something big is happening in AI — and most people will be blindsided – Fortune
Matt Shumer, co-founder and CEO of OthersideAI. courtesy of OthersideAISomething big is happening in AI — and most people will be blindsided
By Matt Shumer, February 11, 2026, 9:22 AM ET
Matt Shumer is the co-founder and CEO of OthersideAI, an applied AI company building the most advanced autocomplete tools in the world, powered by large-scale AI systems like GPT-3. OthersideAI is the company behind HyperWrite, the leading AI autocomplete Chrome extension for consumers. Previously, while in high school, Matt founded Visos, a startup developing next-generation Virtual Reality software designed for medical use, and FURI, a company aiming to disrupt the sporting goods industry by creating high-performance products and selling them for fair prices.
Think back to February 2020.
A few people were talking about a virus spreading overseas. If someone told you they were stockpiling toilet paper you would have thought they’d been spending too much time on a weird corner of the internet. Then, over the course of about three weeks, the entire world changed.
I think we’re in the “this seems overblown” phase of something much, much bigger than Covid.
I’ve spent six years building an AI startup and investing in the space. I live in this world. And I’m writing this for the people in my life who don’t. I keep giving them the polite, cocktail-party version. Because the honest version sounds like I’ve lost my mind. But the gap between what I’ve been saying and what is actually happening has gotten far too big. The people I care about deserve to hear what is coming, even if it sounds crazy.
I should be clear about something up front: even though I work in AI, I have almost no influence over what’s about to happen, and neither does the vast majority of the industry. The future is being shaped by a remarkably small number of people: a few hundred researchers at a handful of companies… OpenAI, Anthropic, Google, DeepMind, and a few others.
Most of us who work in AI are building on top of foundations we didn’t lay. We’re watching this unfold the same as you… we just happen to be close enough to feel the ground shake first.
But it’s time now. Not in an “eventually we should talk about this” way. In a “this is happening right now and I need you to understand it” way.
I know this is real because it happened to me first
Here’s the thing nobody outside of tech quite understands yet: we’re not making predictions. We’re telling you what already occurred in our own jobs, and warning you that you’re next.
For years, AI had been improving steadily. Then in 2025, new techniques for building these models unlocked a much faster pace of progress. This year, something clicked. Not like a light switch… more like the moment you realize the water has been rising around you and is now at your chest.
I am no longer needed for the actual technical work of my job. I describe what I want built, in plain English, and it just… appears. Not a rough draft I need to fix. The finished thing. I tell the AI what I want, walk away from my computer for four hours, and come back to find the work done. Done well, done better than I would have done it myself, with no corrections needed. A couple of months ago, I was going back and forth with the AI, guiding it, making edits. Now I just describe the outcome and leave.
Let me give you an example so you can understand what this actually looks like in practice. I’ll tell the AI: “I want to build this app. Here’s what it should do, here’s roughly what it should look like. Figure out the user flow, the design, all of it.” And it does. It writes tens of thousands of lines of code. Then, and this is the part that would have been unthinkable a year ago, it opens the app itself. It clicks through the buttons. It tests the features. It uses the app the way a person would. If it doesn’t like how something looks or feels, it goes back and changes it, on its own. It iterates, like a developer would, fixing and refining until it’s satisfied. Only once it has decided the app meets its own standards does it come back to me and say: “It’s ready for you to test.” And when I test it, it’s usually perfect.
I’m not exaggerating. That is what my Monday looked like this week.
I’ve always been early to adopt AI tools. But the last few months have shocked me. These new AI models aren’t incremental improvements. This is a different thing entirely.
The experience that tech workers have had over the past year, of watching AI go from “helpful tool” to “does my job better than I do”, is the experience everyone else is about to have. Law, finance, medicine, accounting, consulting, writing, design, analysis, customer service. Not in 10 years. The people building these systems say one to five years. Some say less. The market was spooked enough this month that it wiped out $1 trillion worth of software value in just a week. And given what I’ve seen in just the last couple of months, I see more disruption, and soon.
“But I tried AI and it wasn’t that good”
If you tried ChatGPT in 2023 or early 2024 and thought “this makes stuff up” or “this isn’t that impressive”, you were right. Those early versions were genuinely limited. They hallucinated. They confidently said things that were nonsense.
The models available today are unrecognizable from what existed even six months ago. The debate about whether AI is “really getting better” or “hitting a wall” — which has been going on for over a year — is over. It’s done. Anyone still making that argument either hasn’t used the current models, has an incentive to downplay what’s happening, or is evaluating based on an experience from 2024 that is no longer relevant. I don’t say that to be dismissive. I say it because the gap between public perception and current reality is now enormous, and that gap is dangerous… because it’s preventing people from preparing.
Part of the problem is that most people are using the free version of AI tools. The free version is over a year behind what paying users have access to. Judging AI based on free-tier ChatGPT is like evaluating the state of smartphones by using a flip phone. The people paying for the best tools, and actually using them daily for real work, know what’s coming.
I think of my friend, who’s a lawyer. I keep telling him to try using AI at his firm, and he keeps finding reasons it won’t work. And I get it. But I’ve had partners at major law firms reach out to me for advice, because they’ve tried the current versions and they see where this is going. One of them, the managing partner at a large firm, spends hours every day using AI. He told me it’s like having a team of associates available instantly. He’s not using it because it’s a toy. He’s using it because it works. And he told me something that stuck with me: every couple of months, it gets significantly more capable for his work. He said if it stays on this trajectory, he expects it’ll be able to do most of what he does before long… and he’s a managing partner with decades of experience. He’s not panicking. But he’s paying very close attention.
Think about what that means for your work.
What this means for your job
I’m going to be direct with you because I think you deserve honesty more than comfort.
Dario Amodei, who is probably the most safety-focused CEO in the AI industry, has publicly predicted that AI will eliminate 50% of entry-level white-collar jobs within one to five years. And many people in the industry think he’s being conservative. Given what the latest models can do, the capability for massive disruption could be here by the end of this year. It’ll take some time to ripple through the economy, but the underlying ability is arriving now.
This is different from every previous wave of automation, and I need you to understand why. AI isn’t replacing one specific skill. It’s a general substitute for cognitive work. It gets better at everything simultaneously. When factories automated, a displaced worker could retrain as an office worker. When the internet disrupted retail, workers moved into logistics or services. But AI doesn’t leave a convenient gap to move into. Whatever you retrain for, it’s improving at that too.
I think the honest answer is that nothing that can be done on a computer is safe in the medium term. If your job happens on a screen (if the core of what you do is reading, writing, analyzing, deciding, communicating through a keyboard) then AI is coming for significant parts of it. The timeline isn’t “someday.” It’s already started.
Eventually, robots will handle physical work too. They’re not quite there yet. But “not quite there yet” in AI terms has a way of becoming “here” faster than anyone expects.
What you should actually do
I’m not writing this to make you feel helpless. I’m writing this because I think the single biggest advantage you can have right now is simply being early. Early to understand it. Early to use it. Early to adapt.
Editor’s Note: Read more online. This is a general AI-supportive post by Matt. And there are many others similar “heads up” posts about AI right now. Keep an eye out for further posts on the AI changes coming. Whatever your views, I recommend keeping up with AI over your own work, or job, society changes, regulations, world impacts, and more. –DrWeb
Continue/Read Original Article Here: Something big is happening in AI — and most people will be blindsided | Fortune
Tags: AI, AI Future, artificial intelligence, Big Changes, Fortune, Heads Up, Jobs Changing, Matt Shumer, Newer Models, OthersideAI, Rapid Developments in AI, Social Impacts, Technology
#AI #AIFuture #artificialIntelligence #BigChanges #Fortune #HeadsUp #JobsChanging #MattShumer #NewerModels #OthersideAI #RapidDevelopmentsInAI #SocialImpacts #Technology -
Gallery Assistant Showcase
This is an article meant to showcase the Gallery Assistant beta that was released for the Korean Galaxy devices before the APK was revealed to the public. Since then, we’ve managed to get our hands on the Gallery Assistant application that we will showcase.
You’ll have to turn off the Auto Blocker feature on your device before you’ll be able to install the APK to your device. Once done, you can turn the feature back on when needed. Please note that Gallery Assistant doesn’t support landscape display rotation yet on tablets, but you can open it in a pop up then add it to the multi window.
Gallery Assistant runs on One UI 6 and higher, and this means that you are required to have a phone or a tablet that’s running at least One UI 6 in order to be able to use this application.
When you open the landing page for the Gallery Assistant, this is what you first see:
You’ll have the following options:
- Reduce size
- Rotate and flip
- Image clipper
- Add watermark
- Compare images
- Save as PDF
Some of those features already exist in the built-in Gallery app, but the Gallery Assistant offers further options.
Of course, you’ll have to select images and/or videos before every operation, such as reducing the image/video size. You can do this by pressing the plus sign at the top of the screen.
Fitst, let’s introduce you to Reduce Size. It allows you to shrink selected images and videos in bulk. You can also select specific pictures to reduce their size, and you can adjust the slider up to ten levels:
Of course, the size that the slider tells you is estimated, and the actual size differs. When you save the picture, the size has been reduced as in below:
Let’s try again with the significant reduction in size, with the slider adjusted to 20%, which means 0.42 MB estimated.
You can notice that the resolution has dropped as part of the resize process. Not only that, but the sharpness of the image has been reduced. You can notice this by the texture of the hair. The original resolution of the picture was in 3456×2304 (2.1 MB), but it has been reduced to 3382×2255 (1.57 MB) and 1590×1060 (402 KB), respectively.
Here are the two comparisons where we showcase the detail loss:
Reduced to 1.57 MB Reduced to 402 KBSecond, the Rotate and Flip feature allows you to rotate and flip images in bulk. You can, for example, straighten, rotate and flip more than one image at once.
The Straighten button automatically detects the subject and tries to straighten the shot, if necessary.
The rotate and flip buttons are trigger switches, and the images that you select are affected when you save them.
Let’s compare the images, one by one:
NormalRotatedFlippedFlipped and RotatedAs for the image clipper, the result is best when you choose a picture that has a clear subject, such as a picture of an object or a person.
In this case, the subject (an old vintage laptop running Windows 95) has been chosen here, and you can either copy the clipped image or save it. The result is:
Fourth, you can add a watermark to an image using either text or a smaller image, such as your name or your logo. You can also repeat the watermark across the whole image.
You can edit the watermark settings in the Info to show screen. You can select either an image or a text. You can customize the font, the color, the size, and the transparency if you want to show text. Else, you can resize the image and change its transparency to give it a more branded look.
In the Repeat screen, you can manipulate with the watermark by either letting it repeat across the whole image, or by adjusting its direction if it’s not repeated.
Once done, you can save the image as a copy or over the original. After that, you can see the new image with the watermark applied.
The fifth tool that Gallery Assistant provides is the Compare Images feature, where you can compare between two, three, or four images. To achieve the best results, you can choose pictures that look identical, but differ in small or large details, as long as both of them are the same subject.
Sixth, you can save images to a single PDF file.
In the hamburger menu, you can change the page size (A3, A4, or Letter), the layout (Portrait or Landscape), and the image quality (Original, Medium, or Low). You can either save the PDF or share it to someone. The only issue is that the banding in the original images that we’ve used is visible.
Finally, you can print the images. Not only that, but you can also print video thumbnails!
After that, the resultant PDF is saved at the folder that you choose.
Again, the banding issues are here.
While we are very excited to have this app handy, this version of the application has room for improvements. Once the global rollout of the final version starts, we’ll inform you.
Image by drobotdean on Freepik
#Android #Android14 #Android15 #Android16 #AndroidB #AndroidBaklava #AndroidU #AndroidV #news #oneUi #OneUI6 #OneUI60 #OneUI61 #OneUI611 #OneUI70 #OneUI80 #Samsung #smartphone #Tech #Technology #update
-
Gallery Assistant Showcase
This is an article meant to showcase the Gallery Assistant beta that was released for the Korean Galaxy devices before the APK was revealed to the public. Since then, we’ve managed to get our hands on the Gallery Assistant application that we will showcase.
You’ll have to turn off the Auto Blocker feature on your device before you’ll be able to install the APK to your device. Once done, you can turn the feature back on when needed. Please note that Gallery Assistant doesn’t support landscape display rotation yet on tablets, but you can open it in a pop up then add it to the multi window.
Gallery Assistant runs on One UI 6 and higher, and this means that you are required to have a phone or a tablet that’s running at least One UI 6 in order to be able to use this application.
When you open the landing page for the Gallery Assistant, this is what you first see:
You’ll have the following options:
- Reduce size
- Rotate and flip
- Image clipper
- Add watermark
- Compare images
- Save as PDF
Some of those features already exist in the built-in Gallery app, but the Gallery Assistant offers further options.
Of course, you’ll have to select images and/or videos before every operation, such as reducing the image/video size. You can do this by pressing the plus sign at the top of the screen.
Fitst, let’s introduce you to Reduce Size. It allows you to shrink selected images and videos in bulk. You can also select specific pictures to reduce their size, and you can adjust the slider up to ten levels:
Of course, the size that the slider tells you is estimated, and the actual size differs. When you save the picture, the size has been reduced as in below:
Let’s try again with the significant reduction in size, with the slider adjusted to 20%, which means 0.42 MB estimated.
You can notice that the resolution has dropped as part of the resize process. Not only that, but the sharpness of the image has been reduced. You can notice this by the texture of the hair. The original resolution of the picture was in 3456×2304 (2.1 MB), but it has been reduced to 3382×2255 (1.57 MB) and 1590×1060 (402 KB), respectively.
Here are the two comparisons where we showcase the detail loss:
Reduced to 1.57 MB Reduced to 402 KBSecond, the Rotate and Flip feature allows you to rotate and flip images in bulk. You can, for example, straighten, rotate and flip more than one image at once.
The Straighten button automatically detects the subject and tries to straighten the shot, if necessary.
The rotate and flip buttons are trigger switches, and the images that you select are affected when you save them.
Let’s compare the images, one by one:
NormalRotatedFlippedFlipped and RotatedAs for the image clipper, the result is best when you choose a picture that has a clear subject, such as a picture of an object or a person.
In this case, the subject (an old vintage laptop running Windows 95) has been chosen here, and you can either copy the clipped image or save it. The result is:
Fourth, you can add a watermark to an image using either text or a smaller image, such as your name or your logo. You can also repeat the watermark across the whole image.
You can edit the watermark settings in the Info to show screen. You can select either an image or a text. You can customize the font, the color, the size, and the transparency if you want to show text. Else, you can resize the image and change its transparency to give it a more branded look.
In the Repeat screen, you can manipulate with the watermark by either letting it repeat across the whole image, or by adjusting its direction if it’s not repeated.
Once done, you can save the image as a copy or over the original. After that, you can see the new image with the watermark applied.
The fifth tool that Gallery Assistant provides is the Compare Images feature, where you can compare between two, three, or four images. To achieve the best results, you can choose pictures that look identical, but differ in small or large details, as long as both of them are the same subject.
Sixth, you can save images to a single PDF file.
In the hamburger menu, you can change the page size (A3, A4, or Letter), the layout (Portrait or Landscape), and the image quality (Original, Medium, or Low). You can either save the PDF or share it to someone. The only issue is that the banding in the original images that we’ve used is visible.
Finally, you can print the images. Not only that, but you can also print video thumbnails!
After that, the resultant PDF is saved at the folder that you choose.
Again, the banding issues are here.
While we are very excited to have this app handy, this version of the application has room for improvements. Once the global rollout of the final version starts, we’ll inform you.
Image by drobotdean on Freepik
#Android #Android14 #Android15 #Android16 #AndroidB #AndroidBaklava #AndroidU #AndroidV #news #oneUi #OneUI6 #OneUI60 #OneUI61 #OneUI611 #OneUI70 #OneUI80 #Samsung #smartphone #Tech #Technology #update
-
Gallery Assistant Showcase
This is an article meant to showcase the Gallery Assistant beta that was released for the Korean Galaxy devices before the APK was revealed to the public. Since then, we’ve managed to get our hands on the Gallery Assistant application that we will showcase.
You’ll have to turn off the Auto Blocker feature on your device before you’ll be able to install the APK to your device. Once done, you can turn the feature back on when needed. Please note that Gallery Assistant doesn’t support landscape display rotation yet on tablets, but you can open it in a pop up then add it to the multi window.
Gallery Assistant runs on One UI 6 and higher, and this means that you are required to have a phone or a tablet that’s running at least One UI 6 in order to be able to use this application.
When you open the landing page for the Gallery Assistant, this is what you first see:
You’ll have the following options:
- Reduce size
- Rotate and flip
- Image clipper
- Add watermark
- Compare images
- Save as PDF
Some of those features already exist in the built-in Gallery app, but the Gallery Assistant offers further options.
Of course, you’ll have to select images and/or videos before every operation, such as reducing the image/video size. You can do this by pressing the plus sign at the top of the screen.
Fitst, let’s introduce you to Reduce Size. It allows you to shrink selected images and videos in bulk. You can also select specific pictures to reduce their size, and you can adjust the slider up to ten levels:
Of course, the size that the slider tells you is estimated, and the actual size differs. When you save the picture, the size has been reduced as in below:
Let’s try again with the significant reduction in size, with the slider adjusted to 20%, which means 0.42 MB estimated.
You can notice that the resolution has dropped as part of the resize process. Not only that, but the sharpness of the image has been reduced. You can notice this by the texture of the hair. The original resolution of the picture was in 3456×2304 (2.1 MB), but it has been reduced to 3382×2255 (1.57 MB) and 1590×1060 (402 KB), respectively.
Here are the two comparisons where we showcase the detail loss:
Reduced to 1.57 MB Reduced to 402 KBSecond, the Rotate and Flip feature allows you to rotate and flip images in bulk. You can, for example, straighten, rotate and flip more than one image at once.
The Straighten button automatically detects the subject and tries to straighten the shot, if necessary.
The rotate and flip buttons are trigger switches, and the images that you select are affected when you save them.
Let’s compare the images, one by one:
NormalRotatedFlippedFlipped and RotatedAs for the image clipper, the result is best when you choose a picture that has a clear subject, such as a picture of an object or a person.
In this case, the subject (an old vintage laptop running Windows 95) has been chosen here, and you can either copy the clipped image or save it. The result is:
Fourth, you can add a watermark to an image using either text or a smaller image, such as your name or your logo. You can also repeat the watermark across the whole image.
You can edit the watermark settings in the Info to show screen. You can select either an image or a text. You can customize the font, the color, the size, and the transparency if you want to show text. Else, you can resize the image and change its transparency to give it a more branded look.
In the Repeat screen, you can manipulate with the watermark by either letting it repeat across the whole image, or by adjusting its direction if it’s not repeated.
Once done, you can save the image as a copy or over the original. After that, you can see the new image with the watermark applied.
The fifth tool that Gallery Assistant provides is the Compare Images feature, where you can compare between two, three, or four images. To achieve the best results, you can choose pictures that look identical, but differ in small or large details, as long as both of them are the same subject.
Sixth, you can save images to a single PDF file.
In the hamburger menu, you can change the page size (A3, A4, or Letter), the layout (Portrait or Landscape), and the image quality (Original, Medium, or Low). You can either save the PDF or share it to someone. The only issue is that the banding in the original images that we’ve used is visible.
Finally, you can print the images. Not only that, but you can also print video thumbnails!
After that, the resultant PDF is saved at the folder that you choose.
Again, the banding issues are here.
While we are very excited to have this app handy, this version of the application has room for improvements. Once the global rollout of the final version starts, we’ll inform you.
Image by drobotdean on Freepik
#Android #Android14 #Android15 #Android16 #AndroidB #AndroidBaklava #AndroidU #AndroidV #news #oneUi #OneUI6 #OneUI60 #OneUI61 #OneUI611 #OneUI70 #OneUI80 #Samsung #smartphone #Tech #Technology #update
-
Gallery Assistant Showcase
This is an article meant to showcase the Gallery Assistant beta that was released for the Korean Galaxy devices before the APK was revealed to the public. Since then, we’ve managed to get our hands on the Gallery Assistant application that we will showcase.
You’ll have to turn off the Auto Blocker feature on your device before you’ll be able to install the APK to your device. Once done, you can turn the feature back on when needed. Please note that Gallery Assistant doesn’t support landscape display rotation yet on tablets, but you can open it in a pop up then add it to the multi window.
Gallery Assistant runs on One UI 6 and higher, and this means that you are required to have a phone or a tablet that’s running at least One UI 6 in order to be able to use this application.
When you open the landing page for the Gallery Assistant, this is what you first see:
You’ll have the following options:
- Reduce size
- Rotate and flip
- Image clipper
- Add watermark
- Compare images
- Save as PDF
Some of those features already exist in the built-in Gallery app, but the Gallery Assistant offers further options.
Of course, you’ll have to select images and/or videos before every operation, such as reducing the image/video size. You can do this by pressing the plus sign at the top of the screen.
Fitst, let’s introduce you to Reduce Size. It allows you to shrink selected images and videos in bulk. You can also select specific pictures to reduce their size, and you can adjust the slider up to ten levels:
Of course, the size that the slider tells you is estimated, and the actual size differs. When you save the picture, the size has been reduced as in below:
Let’s try again with the significant reduction in size, with the slider adjusted to 20%, which means 0.42 MB estimated.
You can notice that the resolution has dropped as part of the resize process. Not only that, but the sharpness of the image has been reduced. You can notice this by the texture of the hair. The original resolution of the picture was in 3456×2304 (2.1 MB), but it has been reduced to 3382×2255 (1.57 MB) and 1590×1060 (402 KB), respectively.
Here are the two comparisons where we showcase the detail loss:
Reduced to 1.57 MB Reduced to 402 KBSecond, the Rotate and Flip feature allows you to rotate and flip images in bulk. You can, for example, straighten, rotate and flip more than one image at once.
The Straighten button automatically detects the subject and tries to straighten the shot, if necessary.
The rotate and flip buttons are trigger switches, and the images that you select are affected when you save them.
Let’s compare the images, one by one:
NormalRotatedFlippedFlipped and RotatedAs for the image clipper, the result is best when you choose a picture that has a clear subject, such as a picture of an object or a person.
In this case, the subject (an old vintage laptop running Windows 95) has been chosen here, and you can either copy the clipped image or save it. The result is:
Fourth, you can add a watermark to an image using either text or a smaller image, such as your name or your logo. You can also repeat the watermark across the whole image.
You can edit the watermark settings in the Info to show screen. You can select either an image or a text. You can customize the font, the color, the size, and the transparency if you want to show text. Else, you can resize the image and change its transparency to give it a more branded look.
In the Repeat screen, you can manipulate with the watermark by either letting it repeat across the whole image, or by adjusting its direction if it’s not repeated.
Once done, you can save the image as a copy or over the original. After that, you can see the new image with the watermark applied.
The fifth tool that Gallery Assistant provides is the Compare Images feature, where you can compare between two, three, or four images. To achieve the best results, you can choose pictures that look identical, but differ in small or large details, as long as both of them are the same subject.
Sixth, you can save images to a single PDF file.
In the hamburger menu, you can change the page size (A3, A4, or Letter), the layout (Portrait or Landscape), and the image quality (Original, Medium, or Low). You can either save the PDF or share it to someone. The only issue is that the banding in the original images that we’ve used is visible.
Finally, you can print the images. Not only that, but you can also print video thumbnails!
After that, the resultant PDF is saved at the folder that you choose.
Again, the banding issues are here.
While we are very excited to have this app handy, this version of the application has room for improvements. Once the global rollout of the final version starts, we’ll inform you.
Image by drobotdean on Freepik
#Android #Android14 #Android15 #Android16 #AndroidB #AndroidBaklava #AndroidU #AndroidV #news #oneUi #OneUI6 #OneUI60 #OneUI61 #OneUI611 #OneUI70 #OneUI80 #Samsung #smartphone #Tech #Technology #update
-
Umm so uh.... has every search engine that DDG aggregates all flip on the "only show AI content farm bs" switch??
Past couple days... if you attempt to look up even basic facts or questions... 90% of the results are all garbage.
Apparently #uBlacklist supports syncing in lists... anyone know of one already for blocking this kinda crap? Or dare I spin up my own and share it? Blocked like 20 in just a couple days.
-
CW: AI slop
Had some fun asking an LLM to translate Sor Juana's Hombres Necios into a hip-hop diss track. I kinda like it.
-----
You loud-ass fools talk shit on women,
like you ain’t the cause from the beginning,
set the play, then call it wrong,
then act confused like “what went on?”You chase, you beg, you won’t take no,
apply that pressure, put on a show,
then when she folds, you flip that tone—
now she a “bitch,” a “hoe”? That’s your own.Yeah, you the one throwin’ words like that,
“slut,” “thot,” like it’s fact,
but you wrote the script, you pushed that scene,
now you mad how it all turned out to be?You hype her up, then drag her name,
switchin’ rules mid-fuckin’ game,
say you want “good,” say you want “real,”
then shame her soon as she catch a feel.What you want, huh? Make it make sense—
a saint at home, in public intense?
You want her quiet, but sexy too,
then call her “easy” when she rocks with you.If she curves you, she “stuck-up,” “ice,”
if she don’t, she “ain’t that nice,”
either way you talk your mess,
can’t let a woman just exist.You build that pressure, stack that heat,
then blame her when she hits the street,
like you ain’t the one who lit that spark,
now you barkin’ loud in the dark.You want respect? Then give that first,
not just when you tryna quench your thirst,
quit playin’ judge, quit actin’ clean,
when your own hands dirty as hell, unseen.Real shit—look in the damn mirror,
all that blame? Yeah, bring it nearer,
‘cause half that dirt you throw around
started with you, then circled back down. -
Switching to a Flip Phone Helped Me Cut Down on My Smartphone Addiction - Was it inconvenient? Yes. Did T9 texting drive me crazy? Definitely. Was it worth doing? ... - https://www.nytimes.com/2024/01/06/technology/smartphone-addiction-flip-phone.html #centerforhumanetechnology #addiction(psychology) #lanelogan(activist) #mobileapplications #textmessaging #smartphones #socialmedia #googleinc #appleinc #boredom #iphone #brain #maps
-
The thread about the Edinburgh Police Box; architectural Time Traveller but no TARDIS!
The year 2024 was celebrated as the 900th anniversary of the “foundation” of the city of Edinburgh and 2025 is also an important local commemoration; the centenary of the appointment of the wonderfully named Ebenezer James (“E.J.“) Macrae as City Architect. His twenty years of service was a time of great change in our built environment and his office was directly responsible for much of that, not without good cause has he been dubbed “the man who shaped modern Edinburgh“. His tenure is characterised by both the volume of public buildings and housing that was erected and also their distinctive style; at once both modern in form and function but also very sympathetic to tradition. A splendid example of that contrast is the Edinburgh Police Box; a mix of anachronistic classical styling and what was then the cutting edge of modern policing.
Former police box at corner of Waverley Bridge and Market Street. CC-by-NC-SA 2.0, Ian T. Edwards via Flickr.The first police boxes with telephones were established in Chicago back in 1881, just 5 years after the unveiling of the telephone itself by a son of Edinburgh. In 1923, Chief Constable Frederick Crawley of Newcastle City Police instituted what would become known as the Police Box System to Sunderland and in doing so revolutionised British policing. He was looking to increase the efficiency of his his force and focused on trying to reduce time spent by officers walking to and from their beats; he estimated up to a quarter of each man’s time on shift was wasted in this manner. His solution was decentralisation. By placing many small, telephone-equipped police boxes at strategic points throughout the city, officers had shorter distances to walk and could devote more time to duty. Crawley recognised this would place the police more centrally within the communities they were expected to serve, creating a ready point of contact for the public – thus increasing the efficiency of reporting emergencies and also making it far easier for the police to contact and coordinate their own officers. Boxes could also be used as temporary lock ups for prisoners while transport was summoned, avoiding the long and often dangerous walks with them back to a police station. A final and significant attraction was that the increased efficiency also allowed the closure of most district police stations and therefore afforded a significant cost saving.
Wooden police box of the type instituted by Crawley for Newcastle City Police. Note the public-facing telephone and first aid boxes mounted to the left of the door. From The Police Journal, vol.1, No.1, January 1928Police boxes soon spread across the country but Edinburgh, as is often the case, was rather slow to catch on. It was not until May 1928 that a deputation was sent by Chief Constable Roderick Ross from the Edinburgh City Police to inspect the system in Newcastle. This was at the insistence of the Scottish Office who refused to sanction an increase in headcount for Ross and instead wanted efficiencies. He submitted a strongly favourable report to the Town Council, which approved a box system for the city in 1929. Ross served as Chief Constable for the exceptional term of 34 years and it was towards the end of his long watch that his force would be wholly and rapidly modernised.
Roderick Ross, when Chief Constable of Ramsgate Borough Police c. 1898The Edinburgh Evening News threw its editorial weight behind the scheme but also amplified significant local concerns that the appearance of boxes would have a detrimental effect on the city. As the system spread, there had been a plethora of different design styles before a utilitarian, standardised version was developed for the Metropolitan Police by the architect George Mackenzie Trench. Trench’s design is instantly familiar to generations of Dr. Who fans as the TARDIS. But “Cheapness has been obtained in England” wrote the News’ editor “by mass production, but Edinburgh has an architectural standard of its own, which the Cockburn Association endeavours to maintain.” The gauntlet was thus thrown down to the City Architect’s office that something altogether different and better was needed.
George Mackenzie Trench standard police box at the National Tramway Museum, Crich. Note the light on the roof, which would flash to indicate an officer was required to attend the box. CC-by-SA 3.0 Dan Sellers via WikimediaE.J. Macrae, along with his assistants Andrew Rollo and James A. Tweedie are credited with the design of the Edinburgh Box, with the signature of their colleague Robert Somerville Ellis on some of the drawings. The initial inspiration may have been taken from the barrel-topped box used in Sheffield which was used as an illustrative example by the Evening News. Two alternative designs were prepared and plans and models were put to the Lord Provost’s Committee in December 1929. The preferred option was then “submitted for the consideration of the Fine Art Commission“. After that a full-size wooden mock-up was erected on the corner of George Street and Frederick Street in October 1930 to test the practicalities of installing boxes and also to familiarise both the police and the public with the design.
Sheffield City Police box, as used as an illustrative example by the Evening News. September 13th 1938The approved box was, dare I say it, an iconic piece of British street furniture design, unique to the city and instantly at home in its environment. It is described in architectural terms thusly:
Rectangular cast-iron police box with classical details, 6ft by 4ft on plan, 2-bay pilastered long elevations, one of which contains door bearing City Arms. Painted blue. Single bay short elevations surmounted by open pediments containing ribboned wreath paterae. Saltire patterned glazing to all elevations. Low-pitched roof.
Official description of the Edinburgh Police Box from Historic Scotland listing
Each box was constructed of prefabricated cast iron panels produced by the Carron Company in Stirlingshire and tipped the scales at over two tons. The understated classical styling was decorated only with a small cast iron castle motif from the city’s coat of arms on the door and on each gable a wreath; symbolising power or triumph. Inside they were equipped with a desk, flip-down seat, telephone, sink and a small wall-mounted electric heater. There was shelving, pigeon holes and notice boards on the walls to accommodate items such as logbooks and forms and hooks were provided for hanging coats, helmets and capes. Hooks were provided for “beat keys”, premises officers on duty were expected to visit and check, or need access to, during their duties. An unofficial but entirely necessary function of the sink was an ersatz urinal; 8 hours in a district with few or no public toilets was a long time for a beat officer to spend without spending any pennies! (This was apparently best achieved by balancing on the stool and taking careful aim. Each box was provisioned with a supply of bleach to keep things as sanitary as possible.) All of this came at a price however; £58 per box (before foundations and services were laid), far more than the wooden hut type which had cost £13 each in Newcastle or £43 for a reinforced concrete standard box as used by the Metropolitan Police.
Sketch design of the Edinburgh City Police Box, redrawn by self from a copy of the original in the Edinburgh City Archives. The original is signed RSE (Robert Somerville Ellis), 6th September 1928. From Dean of Guild Court of Edinburgh, Edinburgh Police Boxes, Lord Provost, Magistrates and Council of the City of Edinburgh, 26th August 1932.On the outside of the box were small doors that gave members of the public access to a Speakerphone that would connect them to police headquarters and another containing a first aid kit. The Speakerphone was a hands-free system activated upon opening of its door. It was felt at the time that the general public were not familiar enough with the use of telephones to provide a handset, and it was also harder to accidentally damage or vandalise.
A police officer demonstrating the use of the Speakerphone unit. Opening the box door automatically connected the phone to the headquarters switchboard. Photo via Lothian & Borders Police WordPress.Despite the best efforts of Macrae’s office to produce a design that was sympathetic to Edinburgh’s built environment, not everyone was pleased. “W.M.H.” wrote to the Evening News that the box at the foot of Drummond Street by the old City Wall was a “case of outrageous vandalism and should be prohibited.” They questioned who in the authorities was responsible for such “outrages” and challenged the city’s heritage watchdog – the Cockburn Association – to “get busy!“. In Portobello, the Communist party had a particularly niche objection; it charged that the boxes were “designed for use in a rebellion” and that “the master class knew that they were driving the workers to desperation, and they were preparing in advance to deal with rebellion“.
The police box at Drummond Street, immediately in front of the Flodden Wall. The photo dates from 1951 and the box still sports its white stripes applied during WW2 to make it more visible during blackout conditions. Records of RCAHMS, SC1164082. © Crown Copyright: HESBoxes were installed throughout 1932 and a considerable public relations exercise was undertaken to get the public to understand how to use them. The Evening News maintained a regular stream of editorials on the subject, Chief Constable Ross gave numerous lectures, model boxes were taken around schools to show children how to use them and Boy Scouts were encouraged to learn the location of as many boxes as possible as part of their Pathfinding badge. In the final run-up to commissioning, public demonstrations of the boxes in use were staged and the press cameras invited.
Photograph showing a staged accident to demonstrate the use of the public call facility on the new police boxes, along with an operator of the switchboard at police headquarters on the High Street that received the calls. Scotsman, May 26th 1933.The box system and “a new era in the history of Edinburgh City Police” was inaugurated in its entirety on Sunday May 28th 1933 at 6AM. This was a year later than intended, a delay that the Lord Provost blamed on the General Post Office which had been slow to install the necessary telephony infrastructure (500 miles of underground and 23 miles of overhead wire).
Bailie Rutherford Fortune places the first call on from a police box with Chief Constable Ross (dark coat and light hat, with moustache) and Mr F. J. Milne (light coat and dark hat, with umbrella) Secretary of the Post Office in Edinburgh.The boxes were only one part of a greater overall system; policing of the city was entirely restructured at this time. The boxes were allocated to four divisions, each with its own headquarters – A at Braid Place, B at Gayfield Square, C at Torphicen Place and D at Leith – and were numbered sequentially and by division. A map of the all their locations as installed in 1933 can be seen here. Each division had a dedicated pool of motor vehicles for response and prisoner transport and was supported by a non-territorial traffic and mounted division (E) based in the Cowgate. At the same moment that the boxes were first unlocked for duty, the doors of nine district police stations (at the Pleasance, West Port, Abbeyhill, Piershill, Stockbridge, Waverley Market, Morningside, Gorgie and Newhaven) and eighteen smaller sub-stations closed for the last time. Most of these sites were disposed of, leaving only the four divisional stations, a sub-divisional station for Portobello plus city police HQ on the High Street.
The Leith Police. Relaxing on break time with tea and “pieces” at Leith Police station in 1930. Photograph by Photo Press Agency, CC-by-NC-SA via ThelmaThe reduction in manpower required by the box system saw fifteen open vacancies for constables written off, three inspectors and five sergeants made redundant and a further five sergeants demoted to constables. Overall the changes reduced the running cost of the force by £5,800 annually.
Six or seven constables might be based out of a single box and would serve their entire 8-hour shift from it, returning after every half hour or hour long “turn ” of their beat to check in with base by phone, write up their logbook and take breaks. Check in calls were performed according to a strict timetable and if any officer missed one his absence would be noted and a colleague sent to investigate. Men on duty could expect a visit by a section sergeant once every shift. The boxes were accessed by a universal key, which each officer kept on his chain with his whistle. A blue light on the roof of the box would flash to let him know that there was a call waiting for him. Sometimes these lights had to be mounted on an extension pole to be better seen from a distance and in the case of the box outside the Tron Kirk on the High Street, it was a high-mounted “sky lantern” on the building on the corner with North Bridge.
The High Street “sky lantern” is still in place on the corner with North Bridge, appropriately mounted next to a symbol of modern police surveillance, the CCTV camera.Commencing in 1938, air raid sirens began to be installed on top of the roofs of many of Edinburgh’s boxes as part of the city’s ARP (Air Raid Precautions) measures. By April 1939, thirty two sirens had been installed, all controlled from master switches at HQ on the High Street and tests of the system were under way, helping to familiarise the public with the sound. In May 1940, a writer to the Evening News’ letters page using the pseudonym Tenement Warden and Old Contemptible suggested that police boxes be used to store “machine guns, hand grenades, ammunition and rifles” to deal with enemy paratroopers and “Hitler’s Fifth Column and Fascists all over Britain“. I cannot see that this idea was ever taken seriously!
Photograph of the type of air raid siren installed on the roof of Edinburgh police boxes. Evening News, 30th November 1938In 1939, the annual Estimates of Expenditure of the Town Council reported that there were now 143 police boxes in the city backed up by 40 telephone pillars. Running costs were £3,350, not including £250 for maintenance, £800 for electricity and £3,350 to the Post Office for telephony. The authorised strength of the force was reported as 871, comprising 688 constables, 91 sergeants, 30 inspectors and one each “woman sergeant” and “woman constable“.
In practice the boxes proved to be stiflingly hot in the summer and bitterly cold in the winter; the issued heater was much to small and badly located, so boxes often sourced their own additional heaters to make them more habitable. On account of the metal structure they “sweated considerably” in damp weather as a result of condensation. The roof interior would eventually be insulated in 1956 to try and tackle this particular issue. All boxes were to have been provided with both electricity and a water supply but in the end economies meant only 86 of the 140 boxes were plumbed in. It was some time before enamel mugs, at 6d per unit, were issued from which the water could be drunk and it took until 1947 for the Town Council to approve an expenditure of £781 to equip each box with an electric kettle for making tea.
“For Bobby’s Cup of Tea”, Evening News, 5th June 1947Uncomfortable they may have been, but the boxes proved to be immensely strong. This was demonstrated in November 1945 when PC John Anderson – on what was his last day of service of a thirty years police career – escaped with a fractured leg when a fire engine crashed into his box at the foot of the Canongate. In 1954, PC Donald Budge walked away from his box at Balgreen with only cuts and bruises after a two ton lamp standard, being installed nearby fell onto the roof of the box he was sitting in. The damage to the box was restricted to a cracked roof, a broken window and cracked sink. Also that year, two constables in the box at Murrayfield Avenue survived it being struck by lightning, although the interior lights, radiator and telephones were put out of service and the air raid siren on the roof activated itself.
It took the public some time to get used to the new system. In 1936, three years after its institution, Chief Contable W. B. R. Morren lamented that there was a general ignorance, particularly on the part of grown ups, as to the location and facilities offered by boxes. Boxes were always subject to interference and vandalism throughout their working lives. The authorities were keen to make an example of anyone caught in such an act and the first prosecution came in November 1933 when 19 year old Colin Gosschalk was caught breaking into the first aid compartment of the box on Prestonfield Avenue. His defence that a friend had dared him to do it was not accepted and he was fined 10s (the maximum being £2).
The system was not without its critics as evidenced in the columns of and letters to the Evening News – a particular but unfounded complaint was that constables were either never in the boxes when needed, or spent too much time sheltering within them rather than “on the beat” – a classic of the Schrödinger’s box genre! In an interview with the ‘News in 1946, Chief Constable Morren said that boxes “fulfilled and continues to fulfil a very useful purpose, but… did not develop that contact between the police and the public which was so desirable, and it had been proved that the system had not been the success in that direction that was anticipated”. Brigadier-General Dudgeon, HM Inspector of Constabulary for Scotland said that the box system had “proved to be of value to both the police and the public” but “the beat constable is the eyes and ears of the police, and be careful that the police box system is not overdone.”
Post-war, policing would begin to change again, with smaller district police stations re-established for the new suburbs. As was the case after its 1920 expansion, it was found once again that the city had “more or less outgrown the numbered strength of the police force“. This was particularly felt in the extensive housing schemes been built since the boxes were introduced and where petty crime and antisocial behaviour were an increasing problem. After the initial roll-out of boxes, too few had been added. For instance, in 1946 just one was approved for the West Pilton housing scheme at the junction of Ferry Road Drive and West Pilton Avenue. The peripheral estates were harder to police on foot as they had a much lower housing density than the inner city, so officers had a far greater distance to cover.
New council housing at the Inch, 1955, Dinmont Drive. Photograph by A. G. Ingram, © Edinburgh City LibrariesThese issues saw a move in the 1950s away from the “box and beat” approach to policing the suburbs to more mechanisation (cars) and technology (walkie talkies). They continued in use for the centre of the city however, but the last box installed in Edinburgh may have been that erected in Davidson’s Mains in 1958.
It is all very nice to see policemen going their rounds, but in these days of radio telegraphy the greatly increased use of telephones and the system of 999 calls it is quite reasonable to expect that there should be some saving in the actual pedestrian work
Bailie Matt A. Murray, Chairman of the Progressive Group of Edinburgh Town Council
The air raid warning system was renewed and expanded in 1952 with 56 sirens refurbished, ten additional ones installed and the remote control system replaced. The signalling was replaced again in the 1960s and the sirens were replaced in the early 1970s. Just before 1pm on Thursday 5th June 1969, the air raid sirens sounded across Edinburgh as an engineer working at the city Police headquarters on the High Street accidentally activated the system. A similar incident occurred on August 1st 1986 when all sirens in the Lothian & Borders Police areas were accidentally activated at 7:30 in the morning due to a fault in the telephone system.
Just as Edinburgh had been slow to catch on to adopting police boxes, it was also slow to let them go. While the Metropolitan police started removing boxes in 1969 and demolished its last in 1981, those in Edinburgh were still nominally in active service into the 1990s. After 1984 however the Chief Constable wanted all officers to have a daily briefing at a station before they came on duty and so after then they were more rarely used and many that were found themselves relegated to providing shelter and storage for traffic wardens. In 1993 the air raid sirens were deactivated by the Scottish Office and in 1995 the Lothian & Borders Police Board deemed thirty five of the eighty six remaining boxes were surplus to requirements and put them up for auction, seeking to save the £500 per annum per box maintenance costs of the increasingly dilapidated estate.
Newspaper advert, Scotsman, June 13th 1995, advertising the sale of 35 surplus police boxesThese were the first boxes maid available on the open market and generated much interest; a variety of proposals from public toilets to newspaper kiosks to air quality monitoring stations to removing the boxes entirely to install them as curios in pubs or people’s gardens were proposed. In 1990, the predecessor of Historic Environment Scotland listed thirteen boxes as Category B to protect them (there are now a total of seventeen) and the city’s Planning Convenor would issue guidelines requiring any changes to the boxes or their interiors needing planning permission.
Former lawyer Gordon Thomson purchased eight boxes and, as American-style coffee drinking swept across the nation, established a small chain of bijou “cappuccino kiosks” called the California Coffee Company. Thomson may not have realised it, but his innovation was very close to recreating a street scene once common in 18th century Edinburgh. A 2000 attempt by Feyzullah Marasli to emulate this success by converting a box on Princes Street into a coffee kiosk came to nothing when it was discovered that despite him refurbishing the box, changing the locks on it, paying £400 to have an electricity supply installed and applying for the necessary Street Trader’s Licence, he neither owned nor leased the box in question and it was still in operational use by the Police!
‘A street coffee house Edinburgh’. Paul Sandby, 1750s, Royal Collection Trust RCIN 914503Lothian & Borders Police attempted to rehabilitate some boxes in the late 1990s by installing touch screen public information points with a video-link to a police station within them. The first such box was unveiled to the press on Princes Street in 1998 at a cost of £10,000. It had 61,000 “hits” during its first year of operation and was judged to have been a success, with two further such boxes converted, however funding never followed through and the innovation was allowed to lapse.
Eleven more boxes were auctioned in 2001, advertised as “an exciting and unique opportunity to obtain a distinctive piece of cast iron street furniture with potential for a wide range of uses“. In 2002, the BBC successfully trademarked the London-style Police Box in connection with Dr. Who and the TARDIS, despite the Metropolitan Police contesting the application with the Registrar of Trade Marks. This did not apply to Edinburgh’s unique boxes, which are categorically not TARDISes, despite what some may say! From 2012 to 2013, the police box at Braid Hills Approach was restored to exhibition standard as a small museum by Angus Self, a great grandson of Chief Constable Roderick Ross. In 2014, fourteen of the remaining boxes were sold off, leaving just one in Police ownership.
‘SwimEasy’ Police Box Museum, Braid Hills Road. CC-by-NC SA 2.0, M J RichardsonThe boxes may now be entirely operationally defunct, but they remain throughout the city and many are in daily use. In fact I’m just back from visiting one this afternoon, It may not be a TARDIS but an architectural time traveller it was!
Late night Brazilian crepes anyone? A police box has you covered… CC-by-NC 2.0, Joe Gordon via FlickrIf you have found this useful, informative or amusing, perhaps you would like to help contribute towards the running costs of this site – including keeping it ad-free and my book-buying budget to find further stories to bring you – by supporting me on ko-fi. Or please do just share this post on social media or amongst friends.
These threads © 2017-2025, Andy Arthur.
NO AI TRAINING: Any use of the contents of this website to “train” generative artificial intelligence (AI) technologies to generate text is expressly prohibited. The author reserves all rights to license uses of this work for generative AI training and development of machine learning language models.
#architecture #CityArchitect #EbenezerJamesMacrae #Police #PoliceBox #Policing #RoderickRoss #StreetFurniture #Written2025
-
The thread about the Edinburgh Police Box; architectural Time Traveller but no TARDIS!
The year 2024 was celebrated as the 900th anniversary of the “foundation” of the city of Edinburgh and 2025 is also an important local commemoration; the centenary of the appointment of the wonderfully named Ebenezer James (“E.J.“) Macrae as City Architect. His twenty years of service was a time of great change in our built environment and his office was directly responsible for much of that, not without good cause has he been dubbed “the man who shaped modern Edinburgh“. His tenure is characterised by both the volume of public buildings and housing that was erected and also their distinctive style; at once both modern in form and function but also very sympathetic to tradition. A splendid example of that contrast is the Edinburgh Police Box; a mix of anachronistic classical styling and what was then the cutting edge of modern policing.
Former police box at corner of Waverley Bridge and Market Street. CC-by-NC-SA 2.0, Ian T. Edwards via Flickr.The first police boxes with telephones were established in Chicago back in 1881, just 5 years after the unveiling of the telephone itself by a son of Edinburgh. In 1923, Chief Constable Frederick Crawley of Newcastle City Police instituted what would become known as the Police Box System to Sunderland and in doing so revolutionised British policing. He was looking to increase the efficiency of his his force and focused on trying to reduce time spent by officers walking to and from their beats; he estimated up to a quarter of each man’s time on shift was wasted in this manner. His solution was decentralisation. By placing many small, telephone-equipped police boxes at strategic points throughout the city, officers had shorter distances to walk and could devote more time to duty. Crawley recognised this would place the police more centrally within the communities they were expected to serve, creating a ready point of contact for the public – thus increasing the efficiency of reporting emergencies and also making it far easier for the police to contact and coordinate their own officers. Boxes could also be used as temporary lock ups for prisoners while transport was summoned, avoiding the long and often dangerous walks with them back to a police station. A final and significant attraction was that the increased efficiency also allowed the closure of most district police stations and therefore afforded a significant cost saving.
Wooden police box of the type instituted by Crawley for Newcastle City Police. Note the public-facing telephone and first aid boxes mounted to the left of the door. From The Police Journal, vol.1, No.1, January 1928Police boxes soon spread across the country but Edinburgh, as is often the case, was rather slow to catch on. It was not until May 1928 that a deputation was sent by Chief Constable Roderick Ross from the Edinburgh City Police to inspect the system in Newcastle. This was at the insistence of the Scottish Office who refused to sanction an increase in headcount for Ross and instead wanted efficiencies. He submitted a strongly favourable report to the Town Council, which approved a box system for the city in 1929. Ross served as Chief Constable for the exceptional term of 34 years and it was towards the end of his long watch that his force would be wholly and rapidly modernised.
Roderick Ross, when Chief Constable of Ramsgate Borough Police c. 1898The Edinburgh Evening News threw its editorial weight behind the scheme but also amplified significant local concerns that the appearance of boxes would have a detrimental effect on the city. As the system spread, there had been a plethora of different design styles before a utilitarian, standardised version was developed for the Metropolitan Police by the architect George Mackenzie Trench. Trench’s design is instantly familiar to generations of Dr. Who fans as the TARDIS. But “Cheapness has been obtained in England” wrote the News’ editor “by mass production, but Edinburgh has an architectural standard of its own, which the Cockburn Association endeavours to maintain.” The gauntlet was thus thrown down to the City Architect’s office that something altogether different and better was needed.
George Mackenzie Trench standard police box at the National Tramway Museum, Crich. Note the light on the roof, which would flash to indicate an officer was required to attend the box. CC-by-SA 3.0 Dan Sellers via WikimediaE.J. Macrae, along with his assistants Andrew Rollo and James A. Tweedie are credited with the design of the Edinburgh Box, with the signature of their colleague Robert Somerville Ellis on some of the drawings. The initial inspiration may have been taken from the barrel-topped box used in Sheffield which was used as an illustrative example by the Evening News. Two alternative designs were prepared and plans and models were put to the Lord Provost’s Committee in December 1929. The preferred option was then “submitted for the consideration of the Fine Art Commission“. After that a full-size wooden mock-up was erected on the corner of George Street and Frederick Street in October 1930 to test the practicalities of installing boxes and also to familiarise both the police and the public with the design.
Sheffield City Police box, as used as an illustrative example by the Evening News. September 13th 1938The approved box was, dare I say it, an iconic piece of British street furniture design, unique to the city and instantly at home in its environment. It is described in architectural terms thusly:
Rectangular cast-iron police box with classical details, 6ft by 4ft on plan, 2-bay pilastered long elevations, one of which contains door bearing City Arms. Painted blue. Single bay short elevations surmounted by open pediments containing ribboned wreath paterae. Saltire patterned glazing to all elevations. Low-pitched roof.
Official description of the Edinburgh Police Box from Historic Scotland listing
Each box was constructed of prefabricated cast iron panels produced by the Carron Company in Stirlingshire and tipped the scales at over two tons. The understated classical styling was decorated only with a small cast iron castle motif from the city’s coat of arms on the door and on each gable a wreath; symbolising power or triumph. Inside they were equipped with a desk, flip-down seat, telephone, sink and a small wall-mounted electric heater. There was shelving, pigeon holes and notice boards on the walls to accommodate items such as logbooks and forms and hooks were provided for hanging coats, helmets and capes. Hooks were provided for “beat keys”, premises officers on duty were expected to visit and check, or need access to, during their duties. An unofficial but entirely necessary function of the sink was an ersatz urinal; 8 hours in a district with few or no public toilets was a long time for a beat officer to spend without spending any pennies! (This was apparently best achieved by balancing on the stool and taking careful aim. Each box was provisioned with a supply of bleach to keep things as sanitary as possible.) All of this came at a price however; £58 per box (before foundations and services were laid), far more than the wooden hut type which had cost £13 each in Newcastle or £43 for a reinforced concrete standard box as used by the Metropolitan Police.
Sketch design of the Edinburgh City Police Box, redrawn by self from a copy of the original in the Edinburgh City Archives. The original is signed RSE (Robert Somerville Ellis), 6th September 1928. From Dean of Guild Court of Edinburgh, Edinburgh Police Boxes, Lord Provost, Magistrates and Council of the City of Edinburgh, 26th August 1932.On the outside of the box were small doors that gave members of the public access to a Speakerphone that would connect them to police headquarters and another containing a first aid kit. The Speakerphone was a hands-free system activated upon opening of its door. It was felt at the time that the general public were not familiar enough with the use of telephones to provide a handset, and it was also harder to accidentally damage or vandalise.
A police officer demonstrating the use of the Speakerphone unit. Opening the box door automatically connected the phone to the headquarters switchboard. Photo via Lothian & Borders Police WordPress.Despite the best efforts of Macrae’s office to produce a design that was sympathetic to Edinburgh’s built environment, not everyone was pleased. “W.M.H.” wrote to the Evening News that the box at the foot of Drummond Street by the old City Wall was a “case of outrageous vandalism and should be prohibited.” They questioned who in the authorities was responsible for such “outrages” and challenged the city’s heritage watchdog – the Cockburn Association – to “get busy!“. In Portobello, the Communist party had a particularly niche objection; it charged that the boxes were “designed for use in a rebellion” and that “the master class knew that they were driving the workers to desperation, and they were preparing in advance to deal with rebellion“.
The police box at Drummond Street, immediately in front of the Flodden Wall. The photo dates from 1951 and the box still sports its white stripes applied during WW2 to make it more visible during blackout conditions. Records of RCAHMS, SC1164082. © Crown Copyright: HESBoxes were installed throughout 1932 and a considerable public relations exercise was undertaken to get the public to understand how to use them. The Evening News maintained a regular stream of editorials on the subject, Chief Constable Ross gave numerous lectures, model boxes were taken around schools to show children how to use them and Boy Scouts were encouraged to learn the location of as many boxes as possible as part of their Pathfinding badge. In the final run-up to commissioning, public demonstrations of the boxes in use were staged and the press cameras invited.
Photograph showing a staged accident to demonstrate the use of the public call facility on the new police boxes, along with an operator of the switchboard at police headquarters on the High Street that received the calls. Scotsman, May 26th 1933.The box system and “a new era in the history of Edinburgh City Police” was inaugurated in its entirety on Sunday May 28th 1933 at 6AM. This was a year later than intended, a delay that the Lord Provost blamed on the General Post Office which had been slow to install the necessary telephony infrastructure (500 miles of underground and 23 miles of overhead wire).
Bailie Rutherford Fortune places the first call on from a police box with Chief Constable Ross (dark coat and light hat, with moustache) and Mr F. J. Milne (light coat and dark hat, with umbrella) Secretary of the Post Office in Edinburgh.The boxes were only one part of a greater overall system; policing of the city was entirely restructured at this time. The boxes were allocated to four divisions, each with its own headquarters – A at Braid Place, B at Gayfield Square, C at Torphicen Place and D at Leith – and were numbered sequentially and by division. A map of the all their locations as installed in 1933 can be seen here. Each division had a dedicated pool of motor vehicles for response and prisoner transport and was supported by a non-territorial traffic and mounted division (E) based in the Cowgate. At the same moment that the boxes were first unlocked for duty, the doors of nine district police stations (at the Pleasance, West Port, Abbeyhill, Piershill, Stockbridge, Waverley Market, Morningside, Gorgie and Newhaven) and eighteen smaller sub-stations closed for the last time. Most of these sites were disposed of, leaving only the four divisional stations, a sub-divisional station for Portobello plus city police HQ on the High Street.
The Leith Police. Relaxing on break time with tea and “pieces” at Leith Police station in 1930. Photograph by Photo Press Agency, CC-by-NC-SA via ThelmaThe reduction in manpower required by the box system saw fifteen open vacancies for constables written off, three inspectors and five sergeants made redundant and a further five sergeants demoted to constables. Overall the changes reduced the running cost of the force by £5,800 annually.
Six or seven constables might be based out of a single box and would serve their entire 8-hour shift from it, returning after every half hour or hour long “turn ” of their beat to check in with base by phone, write up their logbook and take breaks. Check in calls were performed according to a strict timetable and if any officer missed one his absence would be noted and a colleague sent to investigate. Men on duty could expect a visit by a section sergeant once every shift. The boxes were accessed by a universal key, which each officer kept on his chain with his whistle. A blue light on the roof of the box would flash to let him know that there was a call waiting for him. Sometimes these lights had to be mounted on an extension pole to be better seen from a distance and in the case of the box outside the Tron Kirk on the High Street, it was a high-mounted “sky lantern” on the building on the corner with North Bridge.
The High Street “sky lantern” is still in place on the corner with North Bridge, appropriately mounted next to a symbol of modern police surveillance, the CCTV camera.Commencing in 1938, air raid sirens began to be installed on top of the roofs of many of Edinburgh’s boxes as part of the city’s ARP (Air Raid Precautions) measures. By April 1939, thirty two sirens had been installed, all controlled from master switches at HQ on the High Street and tests of the system were under way, helping to familiarise the public with the sound. In May 1940, a writer to the Evening News’ letters page using the pseudonym Tenement Warden and Old Contemptible suggested that police boxes be used to store “machine guns, hand grenades, ammunition and rifles” to deal with enemy paratroopers and “Hitler’s Fifth Column and Fascists all over Britain“. I cannot see that this idea was ever taken seriously!
Photograph of the type of air raid siren installed on the roof of Edinburgh police boxes. Evening News, 30th November 1938In 1939, the annual Estimates of Expenditure of the Town Council reported that there were now 143 police boxes in the city backed up by 40 telephone pillars. Running costs were £3,350, not including £250 for maintenance, £800 for electricity and £3,350 to the Post Office for telephony. The authorised strength of the force was reported as 871, comprising 688 constables, 91 sergeants, 30 inspectors and one each “woman sergeant” and “woman constable“.
In practice the boxes proved to be stiflingly hot in the summer and bitterly cold in the winter; the issued heater was much to small and badly located, so boxes often sourced their own additional heaters to make them more habitable. On account of the metal structure they “sweated considerably” in damp weather as a result of condensation. The roof interior would eventually be insulated in 1956 to try and tackle this particular issue. All boxes were to have been provided with both electricity and a water supply but in the end economies meant only 86 of the 140 boxes were plumbed in. It was some time before enamel mugs, at 6d per unit, were issued from which the water could be drunk and it took until 1947 for the Town Council to approve an expenditure of £781 to equip each box with an electric kettle for making tea.
“For Bobby’s Cup of Tea”, Evening News, 5th June 1947Uncomfortable they may have been, but the boxes proved to be immensely strong. This was demonstrated in November 1945 when PC John Anderson – on what was his last day of service of a thirty years police career – escaped with a fractured leg when a fire engine crashed into his box at the foot of the Canongate. In 1954, PC Donald Budge walked away from his box at Balgreen with only cuts and bruises after a two ton lamp standard, being installed nearby fell onto the roof of the box he was sitting in. The damage to the box was restricted to a cracked roof, a broken window and cracked sink. Also that year, two constables in the box at Murrayfield Avenue survived it being struck by lightning, although the interior lights, radiator and telephones were put out of service and the air raid siren on the roof activated itself.
It took the public some time to get used to the new system. In 1936, three years after its institution, Chief Contable W. B. R. Morren lamented that there was a general ignorance, particularly on the part of grown ups, as to the location and facilities offered by boxes. Boxes were always subject to interference and vandalism throughout their working lives. The authorities were keen to make an example of anyone caught in such an act and the first prosecution came in November 1933 when 19 year old Colin Gosschalk was caught breaking into the first aid compartment of the box on Prestonfield Avenue. His defence that a friend had dared him to do it was not accepted and he was fined 10s (the maximum being £2).
The system was not without its critics as evidenced in the columns of and letters to the Evening News – a particular but unfounded complaint was that constables were either never in the boxes when needed, or spent too much time sheltering within them rather than “on the beat” – a classic of the Schrödinger’s box genre! In an interview with the ‘News in 1946, Chief Constable Morren said that boxes “fulfilled and continues to fulfil a very useful purpose, but… did not develop that contact between the police and the public which was so desirable, and it had been proved that the system had not been the success in that direction that was anticipated”. Brigadier-General Dudgeon, HM Inspector of Constabulary for Scotland said that the box system had “proved to be of value to both the police and the public” but “the beat constable is the eyes and ears of the police, and be careful that the police box system is not overdone.”
Post-war, policing would begin to change again, with smaller district police stations re-established for the new suburbs. As was the case after its 1920 expansion, it was found once again that the city had “more or less outgrown the numbered strength of the police force“. This was particularly felt in the extensive housing schemes been built since the boxes were introduced and where petty crime and antisocial behaviour were an increasing problem. After the initial roll-out of boxes, too few had been added. For instance, in 1946 just one was approved for the West Pilton housing scheme at the junction of Ferry Road Drive and West Pilton Avenue. The peripheral estates were harder to police on foot as they had a much lower housing density than the inner city, so officers had a far greater distance to cover.
New council housing at the Inch, 1955, Dinmont Drive. Photograph by A. G. Ingram, © Edinburgh City LibrariesThese issues saw a move in the 1950s away from the “box and beat” approach to policing the suburbs to more mechanisation (cars) and technology (walkie talkies). They continued in use for the centre of the city however, but the last box installed in Edinburgh may have been that erected in Davidson’s Mains in 1958.
It is all very nice to see policemen going their rounds, but in these days of radio telegraphy the greatly increased use of telephones and the system of 999 calls it is quite reasonable to expect that there should be some saving in the actual pedestrian work
Bailie Matt A. Murray, Chairman of the Progressive Group of Edinburgh Town Council
The air raid warning system was renewed and expanded in 1952 with 56 sirens refurbished, ten additional ones installed and the remote control system replaced. The signalling was replaced again in the 1960s and the sirens were replaced in the early 1970s. Just before 1pm on Thursday 5th June 1969, the air raid sirens sounded across Edinburgh as an engineer working at the city Police headquarters on the High Street accidentally activated the system. A similar incident occurred on August 1st 1986 when all sirens in the Lothian & Borders Police areas were accidentally activated at 7:30 in the morning due to a fault in the telephone system.
Just as Edinburgh had been slow to catch on to adopting police boxes, it was also slow to let them go. While the Metropolitan police started removing boxes in 1969 and demolished its last in 1981, those in Edinburgh were still nominally in active service into the 1990s. After 1984 however the Chief Constable wanted all officers to have a daily briefing at a station before they came on duty and so after then they were more rarely used and many that were found themselves relegated to providing shelter and storage for traffic wardens. In 1993 the air raid sirens were deactivated by the Scottish Office and in 1995 the Lothian & Borders Police Board deemed thirty five of the eighty six remaining boxes were surplus to requirements and put them up for auction, seeking to save the £500 per annum per box maintenance costs of the increasingly dilapidated estate.
Newspaper advert, Scotsman, June 13th 1995, advertising the sale of 35 surplus police boxesThese were the first boxes maid available on the open market and generated much interest; a variety of proposals from public toilets to newspaper kiosks to air quality monitoring stations to removing the boxes entirely to install them as curios in pubs or people’s gardens were proposed. In 1990, the predecessor of Historic Environment Scotland listed thirteen boxes as Category B to protect them (there are now a total of seventeen) and the city’s Planning Convenor would issue guidelines requiring any changes to the boxes or their interiors needing planning permission.
Former lawyer Gordon Thomson purchased eight boxes and, as American-style coffee drinking swept across the nation, established a small chain of bijou “cappuccino kiosks” called the California Coffee Company. Thomson may not have realised it, but his innovation was very close to recreating a street scene once common in 18th century Edinburgh. A 2000 attempt by Feyzullah Marasli to emulate this success by converting a box on Princes Street into a coffee kiosk came to nothing when it was discovered that despite him refurbishing the box, changing the locks on it, paying £400 to have an electricity supply installed and applying for the necessary Street Trader’s Licence, he neither owned nor leased the box in question and it was still in operational use by the Police!
‘A street coffee house Edinburgh’. Paul Sandby, 1750s, Royal Collection Trust RCIN 914503Lothian & Borders Police attempted to rehabilitate some boxes in the late 1990s by installing touch screen public information points with a video-link to a police station within them. The first such box was unveiled to the press on Princes Street in 1998 at a cost of £10,000. It had 61,000 “hits” during its first year of operation and was judged to have been a success, with two further such boxes converted, however funding never followed through and the innovation was allowed to lapse.
Eleven more boxes were auctioned in 2001, advertised as “an exciting and unique opportunity to obtain a distinctive piece of cast iron street furniture with potential for a wide range of uses“. In 2002, the BBC successfully trademarked the London-style Police Box in connection with Dr. Who and the TARDIS, despite the Metropolitan Police contesting the application with the Registrar of Trade Marks. This did not apply to Edinburgh’s unique boxes, which are categorically not TARDISes, despite what some may say! From 2012 to 2013, the police box at Braid Hills Approach was restored to exhibition standard as a small museum by Angus Self, a great grandson of Chief Constable Roderick Ross. In 2014, fourteen of the remaining boxes were sold off, leaving just one in Police ownership.
‘SwimEasy’ Police Box Museum, Braid Hills Road. CC-by-NC SA 2.0, M J RichardsonThe boxes may now be entirely operationally defunct, but they remain throughout the city and many are in daily use. In fact I’m just back from visiting one this afternoon, It may not be a TARDIS but an architectural time traveller it was!
Late night Brazilian crepes anyone? A police box has you covered… CC-by-NC 2.0, Joe Gordon via FlickrIf you have found this useful, informative or amusing, perhaps you would like to help contribute towards the running costs of this site – including keeping it ad-free and my book-buying budget to find further stories to bring you – by supporting me on ko-fi. Or please do just share this post on social media or amongst friends.
These threads © 2017-2025, Andy Arthur.
NO AI TRAINING: Any use of the contents of this website to “train” generative artificial intelligence (AI) technologies to generate text is expressly prohibited. The author reserves all rights to license uses of this work for generative AI training and development of machine learning language models.
#architecture #CityArchitect #EbenezerJamesMacrae #Police #PoliceBox #Policing #RoderickRoss #StreetFurniture #Written2025
-
The thread about the Edinburgh Police Box; architectural Time Traveller but no TARDIS!
The year 2024 was celebrated as the 900th anniversary of the “foundation” of the city of Edinburgh and 2025 is also an important local commemoration; the centenary of the appointment of the wonderfully named Ebenezer James (“E.J.“) Macrae as City Architect. His twenty years of service was a time of great change in our built environment and his office was directly responsible for much of that, not without good cause has he been dubbed “the man who shaped modern Edinburgh“. His tenure is characterised by both the volume of public buildings and housing that was erected and also their distinctive style; at once both modern in form and function but also very sympathetic to tradition. A splendid example of that contrast is the Edinburgh Police Box; a mix of anachronistic classical styling and what was then the cutting edge of modern policing.
Former police box at corner of Waverley Bridge and Market Street. CC-by-NC-SA 2.0, Ian T. Edwards via Flickr.The first police boxes with telephones were established in Chicago back in 1881, just 5 years after the unveiling of the telephone itself by a son of Edinburgh. In 1923, Chief Constable Frederick Crawley of Newcastle City Police instituted what would become known as the Police Box System to Sunderland and in doing so revolutionised British policing. He was looking to increase the efficiency of his his force and focused on trying to reduce time spent by officers walking to and from their beats; he estimated up to a quarter of each man’s time on shift was wasted in this manner. His solution was decentralisation. By placing many small, telephone-equipped police boxes at strategic points throughout the city, officers had shorter distances to walk and could devote more time to duty. Crawley recognised this would place the police more centrally within the communities they were expected to serve, creating a ready point of contact for the public – thus increasing the efficiency of reporting emergencies and also making it far easier for the police to contact and coordinate their own officers. Boxes could also be used as temporary lock ups for prisoners while transport was summoned, avoiding the long and often dangerous walks with them back to a police station. A final and significant attraction was that the increased efficiency also allowed the closure of most district police stations and therefore afforded a significant cost saving.
Wooden police box of the type instituted by Crawley for Newcastle City Police. Note the public-facing telephone and first aid boxes mounted to the left of the door. From The Police Journal, vol.1, No.1, January 1928Police boxes soon spread across the country but Edinburgh, as is often the case, was rather slow to catch on. It was not until May 1928 that a deputation was sent by Chief Constable Roderick Ross from the Edinburgh City Police to inspect the system in Newcastle. This was at the insistence of the Scottish Office who refused to sanction an increase in headcount for Ross and instead wanted efficiencies. He submitted a strongly favourable report to the Town Council, which approved a box system for the city in 1929. Ross served as Chief Constable for the exceptional term of 34 years and it was towards the end of his long watch that his force would be wholly and rapidly modernised.
Roderick Ross, when Chief Constable of Ramsgate Borough Police c. 1898The Edinburgh Evening News threw its editorial weight behind the scheme but also amplified significant local concerns that the appearance of boxes would have a detrimental effect on the city. As the system spread, there had been a plethora of different design styles before a utilitarian, standardised version was developed for the Metropolitan Police by the architect George Mackenzie Trench. Trench’s design is instantly familiar to generations of Dr. Who fans as the TARDIS. But “Cheapness has been obtained in England” wrote the News’ editor “by mass production, but Edinburgh has an architectural standard of its own, which the Cockburn Association endeavours to maintain.” The gauntlet was thus thrown down to the City Architect’s office that something altogether different and better was needed.
George Mackenzie Trench standard police box at the National Tramway Museum, Crich. Note the light on the roof, which would flash to indicate an officer was required to attend the box. CC-by-SA 3.0 Dan Sellers via WikimediaE.J. Macrae, along with his assistants Andrew Rollo and James A. Tweedie are credited with the design of the Edinburgh Box, with the signature of their colleague Robert Somerville Ellis on some of the drawings. The initial inspiration may have been taken from the barrel-topped box used in Sheffield which was used as an illustrative example by the Evening News. Two alternative designs were prepared and plans and models were put to the Lord Provost’s Committee in December 1929. The preferred option was then “submitted for the consideration of the Fine Art Commission“. After that a full-size wooden mock-up was erected on the corner of George Street and Frederick Street in October 1930 to test the practicalities of installing boxes and also to familiarise both the police and the public with the design.
Sheffield City Police box, as used as an illustrative example by the Evening News. September 13th 1938The approved box was, dare I say it, an iconic piece of British street furniture design, unique to the city and instantly at home in its environment. It is described in architectural terms thusly:
Rectangular cast-iron police box with classical details, 6ft by 4ft on plan, 2-bay pilastered long elevations, one of which contains door bearing City Arms. Painted blue. Single bay short elevations surmounted by open pediments containing ribboned wreath paterae. Saltire patterned glazing to all elevations. Low-pitched roof.
Official description of the Edinburgh Police Box from Historic Scotland listing
Each box was constructed of prefabricated cast iron panels produced by the Carron Company in Stirlingshire and tipped the scales at over two tons. The understated classical styling was decorated only with a small cast iron castle motif from the city’s coat of arms on the door and on each gable a wreath; symbolising power or triumph. Inside they were equipped with a desk, flip-down seat, telephone, sink and a small wall-mounted electric heater. There was shelving, pigeon holes and notice boards on the walls to accommodate items such as logbooks and forms and hooks were provided for hanging coats, helmets and capes. Hooks were provided for “beat keys”, premises officers on duty were expected to visit and check, or need access to, during their duties. An unofficial but entirely necessary function of the sink was an ersatz urinal; 8 hours in a district with few or no public toilets was a long time for a beat officer to spend without spending any pennies! (This was apparently best achieved by balancing on the stool and taking careful aim. Each box was provisioned with a supply of bleach to keep things as sanitary as possible.) All of this came at a price however; £58 per box (before foundations and services were laid), far more than the wooden hut type which had cost £13 each in Newcastle or £43 for a reinforced concrete standard box as used by the Metropolitan Police.
Sketch design of the Edinburgh City Police Box, redrawn by self from a copy of the original in the Edinburgh City Archives. The original is signed RSE (Robert Somerville Ellis), 6th September 1928. From Dean of Guild Court of Edinburgh, Edinburgh Police Boxes, Lord Provost, Magistrates and Council of the City of Edinburgh, 26th August 1932.On the outside of the box were small doors that gave members of the public access to a Speakerphone that would connect them to police headquarters and another containing a first aid kit. The Speakerphone was a hands-free system activated upon opening of its door. It was felt at the time that the general public were not familiar enough with the use of telephones to provide a handset, and it was also harder to accidentally damage or vandalise.
A police officer demonstrating the use of the Speakerphone unit. Opening the box door automatically connected the phone to the headquarters switchboard. Photo via Lothian & Borders Police WordPress.Despite the best efforts of Macrae’s office to produce a design that was sympathetic to Edinburgh’s built environment, not everyone was pleased. “W.M.H.” wrote to the Evening News that the box at the foot of Drummond Street by the old City Wall was a “case of outrageous vandalism and should be prohibited.” They questioned who in the authorities was responsible for such “outrages” and challenged the city’s heritage watchdog – the Cockburn Association – to “get busy!“. In Portobello, the Communist party had a particularly niche objection; it charged that the boxes were “designed for use in a rebellion” and that “the master class knew that they were driving the workers to desperation, and they were preparing in advance to deal with rebellion“.
The police box at Drummond Street, immediately in front of the Flodden Wall. The photo dates from 1951 and the box still sports its white stripes applied during WW2 to make it more visible during blackout conditions. Records of RCAHMS, SC1164082. © Crown Copyright: HESBoxes were installed throughout 1932 and a considerable public relations exercise was undertaken to get the public to understand how to use them. The Evening News maintained a regular stream of editorials on the subject, Chief Constable Ross gave numerous lectures, model boxes were taken around schools to show children how to use them and Boy Scouts were encouraged to learn the location of as many boxes as possible as part of their Pathfinding badge. In the final run-up to commissioning, public demonstrations of the boxes in use were staged and the press cameras invited.
Photograph showing a staged accident to demonstrate the use of the public call facility on the new police boxes, along with an operator of the switchboard at police headquarters on the High Street that received the calls. Scotsman, May 26th 1933.The box system and “a new era in the history of Edinburgh City Police” was inaugurated in its entirety on Sunday May 28th 1933 at 6AM. This was a year later than intended, a delay that the Lord Provost blamed on the General Post Office which had been slow to install the necessary telephony infrastructure (500 miles of underground and 23 miles of overhead wire).
Bailie Rutherford Fortune places the first call on from a police box with Chief Constable Ross (dark coat and light hat, with moustache) and Mr F. J. Milne (light coat and dark hat, with umbrella) Secretary of the Post Office in Edinburgh.The boxes were only one part of a greater overall system; policing of the city was entirely restructured at this time. The boxes were allocated to four divisions, each with its own headquarters – A at Braid Place, B at Gayfield Square, C at Torphicen Place and D at Leith – and were numbered sequentially and by division. A map of the all their locations as installed in 1933 can be seen here. Each division had a dedicated pool of motor vehicles for response and prisoner transport and was supported by a non-territorial traffic and mounted division (E) based in the Cowgate. At the same moment that the boxes were first unlocked for duty, the doors of nine district police stations (at the Pleasance, West Port, Abbeyhill, Piershill, Stockbridge, Waverley Market, Morningside, Gorgie and Newhaven) and eighteen smaller sub-stations closed for the last time. Most of these sites were disposed of, leaving only the four divisional stations, a sub-divisional station for Portobello plus city police HQ on the High Street.
The Leith Police. Relaxing on break time with tea and “pieces” at Leith Police station in 1930. Photograph by Photo Press Agency, CC-by-NC-SA via ThelmaThe reduction in manpower required by the box system saw fifteen open vacancies for constables written off, three inspectors and five sergeants made redundant and a further five sergeants demoted to constables. Overall the changes reduced the running cost of the force by £5,800 annually.
Six or seven constables might be based out of a single box and would serve their entire 8-hour shift from it, returning after every half hour or hour long “turn ” of their beat to check in with base by phone, write up their logbook and take breaks. Check in calls were performed according to a strict timetable and if any officer missed one his absence would be noted and a colleague sent to investigate. Men on duty could expect a visit by a section sergeant once every shift. The boxes were accessed by a universal key, which each officer kept on his chain with his whistle. A blue light on the roof of the box would flash to let him know that there was a call waiting for him. Sometimes these lights had to be mounted on an extension pole to be better seen from a distance and in the case of the box outside the Tron Kirk on the High Street, it was a high-mounted “sky lantern” on the building on the corner with North Bridge.
The High Street “sky lantern” is still in place on the corner with North Bridge, appropriately mounted next to a symbol of modern police surveillance, the CCTV camera.Commencing in 1938, air raid sirens began to be installed on top of the roofs of many of Edinburgh’s boxes as part of the city’s ARP (Air Raid Precautions) measures. By April 1939, thirty two sirens had been installed, all controlled from master switches at HQ on the High Street and tests of the system were under way, helping to familiarise the public with the sound. In May 1940, a writer to the Evening News’ letters page using the pseudonym Tenement Warden and Old Contemptible suggested that police boxes be used to store “machine guns, hand grenades, ammunition and rifles” to deal with enemy paratroopers and “Hitler’s Fifth Column and Fascists all over Britain“. I cannot see that this idea was ever taken seriously!
Photograph of the type of air raid siren installed on the roof of Edinburgh police boxes. Evening News, 30th November 1938In 1939, the annual Estimates of Expenditure of the Town Council reported that there were now 143 police boxes in the city backed up by 40 telephone pillars. Running costs were £3,350, not including £250 for maintenance, £800 for electricity and £3,350 to the Post Office for telephony. The authorised strength of the force was reported as 871, comprising 688 constables, 91 sergeants, 30 inspectors and one each “woman sergeant” and “woman constable“.
In practice the boxes proved to be stiflingly hot in the summer and bitterly cold in the winter; the issued heater was much to small and badly located, so boxes often sourced their own additional heaters to make them more habitable. On account of the metal structure they “sweated considerably” in damp weather as a result of condensation. The roof interior would eventually be insulated in 1956 to try and tackle this particular issue. All boxes were to have been provided with both electricity and a water supply but in the end economies meant only 86 of the 140 boxes were plumbed in. It was some time before enamel mugs, at 6d per unit, were issued from which the water could be drunk and it took until 1947 for the Town Council to approve an expenditure of £781 to equip each box with an electric kettle for making tea.
“For Bobby’s Cup of Tea”, Evening News, 5th June 1947Uncomfortable they may have been, but the boxes proved to be immensely strong. This was demonstrated in November 1945 when PC John Anderson – on what was his last day of service of a thirty years police career – escaped with a fractured leg when a fire engine crashed into his box at the foot of the Canongate. In 1954, PC Donald Budge walked away from his box at Balgreen with only cuts and bruises after a two ton lamp standard, being installed nearby fell onto the roof of the box he was sitting in. The damage to the box was restricted to a cracked roof, a broken window and cracked sink. Also that year, two constables in the box at Murrayfield Avenue survived it being struck by lightning, although the interior lights, radiator and telephones were put out of service and the air raid siren on the roof activated itself.
It took the public some time to get used to the new system. In 1936, three years after its institution, Chief Contable W. B. R. Morren lamented that there was a general ignorance, particularly on the part of grown ups, as to the location and facilities offered by boxes. Boxes were always subject to interference and vandalism throughout their working lives. The authorities were keen to make an example of anyone caught in such an act and the first prosecution came in November 1933 when 19 year old Colin Gosschalk was caught breaking into the first aid compartment of the box on Prestonfield Avenue. His defence that a friend had dared him to do it was not accepted and he was fined 10s (the maximum being £2).
The system was not without its critics as evidenced in the columns of and letters to the Evening News – a particular but unfounded complaint was that constables were either never in the boxes when needed, or spent too much time sheltering within them rather than “on the beat” – a classic of the Schrödinger’s box genre! In an interview with the ‘News in 1946, Chief Constable Morren said that boxes “fulfilled and continues to fulfil a very useful purpose, but… did not develop that contact between the police and the public which was so desirable, and it had been proved that the system had not been the success in that direction that was anticipated”. Brigadier-General Dudgeon, HM Inspector of Constabulary for Scotland said that the box system had “proved to be of value to both the police and the public” but “the beat constable is the eyes and ears of the police, and be careful that the police box system is not overdone.”
Post-war, policing would begin to change again, with smaller district police stations re-established for the new suburbs. As was the case after its 1920 expansion, it was found once again that the city had “more or less outgrown the numbered strength of the police force“. This was particularly felt in the extensive housing schemes been built since the boxes were introduced and where petty crime and antisocial behaviour were an increasing problem. After the initial roll-out of boxes, too few had been added. For instance, in 1946 just one was approved for the West Pilton housing scheme at the junction of Ferry Road Drive and West Pilton Avenue. The peripheral estates were harder to police on foot as they had a much lower housing density than the inner city, so officers had a far greater distance to cover.
New council housing at the Inch, 1955, Dinmont Drive. Photograph by A. G. Ingram, © Edinburgh City LibrariesThese issues saw a move in the 1950s away from the “box and beat” approach to policing the suburbs to more mechanisation (cars) and technology (walkie talkies). They continued in use for the centre of the city however, but the last box installed in Edinburgh may have been that erected in Davidson’s Mains in 1958.
It is all very nice to see policemen going their rounds, but in these days of radio telegraphy the greatly increased use of telephones and the system of 999 calls it is quite reasonable to expect that there should be some saving in the actual pedestrian work
Bailie Matt A. Murray, Chairman of the Progressive Group of Edinburgh Town Council
The air raid warning system was renewed and expanded in 1952 with 56 sirens refurbished, ten additional ones installed and the remote control system replaced. The signalling was replaced again in the 1960s and the sirens were replaced in the early 1970s. Just before 1pm on Thursday 5th June 1969, the air raid sirens sounded across Edinburgh as an engineer working at the city Police headquarters on the High Street accidentally activated the system. A similar incident occurred on August 1st 1986 when all sirens in the Lothian & Borders Police areas were accidentally activated at 7:30 in the morning due to a fault in the telephone system.
Just as Edinburgh had been slow to catch on to adopting police boxes, it was also slow to let them go. While the Metropolitan police started removing boxes in 1969 and demolished its last in 1981, those in Edinburgh were still nominally in active service into the 1990s. After 1984 however the Chief Constable wanted all officers to have a daily briefing at a station before they came on duty and so after then they were more rarely used and many that were found themselves relegated to providing shelter and storage for traffic wardens. In 1993 the air raid sirens were deactivated by the Scottish Office and in 1995 the Lothian & Borders Police Board deemed thirty five of the eighty six remaining boxes were surplus to requirements and put them up for auction, seeking to save the £500 per annum per box maintenance costs of the increasingly dilapidated estate.
Newspaper advert, Scotsman, June 13th 1995, advertising the sale of 35 surplus police boxesThese were the first boxes maid available on the open market and generated much interest; a variety of proposals from public toilets to newspaper kiosks to air quality monitoring stations to removing the boxes entirely to install them as curios in pubs or people’s gardens were proposed. In 1990, the predecessor of Historic Environment Scotland listed thirteen boxes as Category B to protect them (there are now a total of seventeen) and the city’s Planning Convenor would issue guidelines requiring any changes to the boxes or their interiors needing planning permission.
Former lawyer Gordon Thomson purchased eight boxes and, as American-style coffee drinking swept across the nation, established a small chain of bijou “cappuccino kiosks” called the California Coffee Company. Thomson may not have realised it, but his innovation was very close to recreating a street scene once common in 18th century Edinburgh. A 2000 attempt by Feyzullah Marasli to emulate this success by converting a box on Princes Street into a coffee kiosk came to nothing when it was discovered that despite him refurbishing the box, changing the locks on it, paying £400 to have an electricity supply installed and applying for the necessary Street Trader’s Licence, he neither owned nor leased the box in question and it was still in operational use by the Police!
‘A street coffee house Edinburgh’. Paul Sandby, 1750s, Royal Collection Trust RCIN 914503Lothian & Borders Police attempted to rehabilitate some boxes in the late 1990s by installing touch screen public information points with a video-link to a police station within them. The first such box was unveiled to the press on Princes Street in 1998 at a cost of £10,000. It had 61,000 “hits” during its first year of operation and was judged to have been a success, with two further such boxes converted, however funding never followed through and the innovation was allowed to lapse.
Eleven more boxes were auctioned in 2001, advertised as “an exciting and unique opportunity to obtain a distinctive piece of cast iron street furniture with potential for a wide range of uses“. In 2002, the BBC successfully trademarked the London-style Police Box in connection with Dr. Who and the TARDIS, despite the Metropolitan Police contesting the application with the Registrar of Trade Marks. This did not apply to Edinburgh’s unique boxes, which are categorically not TARDISes, despite what some may say! From 2012 to 2013, the police box at Braid Hills Approach was restored to exhibition standard as a small museum by Angus Self, a great grandson of Chief Constable Roderick Ross. In 2014, fourteen of the remaining boxes were sold off, leaving just one in Police ownership.
‘SwimEasy’ Police Box Museum, Braid Hills Road. CC-by-NC SA 2.0, M J RichardsonThe boxes may now be entirely operationally defunct, but they remain throughout the city and many are in daily use. In fact I’m just back from visiting one this afternoon, It may not be a TARDIS but an architectural time traveller it was!
Late night Brazilian crepes anyone? A police box has you covered… CC-by-NC 2.0, Joe Gordon via FlickrIf you have found this useful, informative or amusing, perhaps you would like to help contribute towards the running costs of this site – including keeping it ad-free and my book-buying budget to find further stories to bring you – by supporting me on ko-fi. Or please do just share this post on social media or amongst friends.
These threads © 2017-2025, Andy Arthur.
NO AI TRAINING: Any use of the contents of this website to “train” generative artificial intelligence (AI) technologies to generate text is expressly prohibited. The author reserves all rights to license uses of this work for generative AI training and development of machine learning language models.
#architecture #CityArchitect #EbenezerJamesMacrae #Police #PoliceBox #Policing #RoderickRoss #StreetFurniture #Written2025
-
The thread about the Edinburgh Police Box; architectural Time Traveller but no TARDIS!
The year 2024 was celebrated as the 900th anniversary of the “foundation” of the city of Edinburgh and 2025 is also an important local commemoration; the centenary of the appointment of the wonderfully named Ebenezer James (“E.J.“) Macrae as City Architect. His twenty years of service was a time of great change in our built environment and his office was directly responsible for much of that, not without good cause has he been dubbed “the man who shaped modern Edinburgh“. His tenure is characterised by both the volume of public buildings and housing that was erected and also their distinctive style; at once both modern in form and function but also very sympathetic to tradition. A splendid example of that contrast is the Edinburgh Police Box; a mix of anachronistic classical styling and what was then the cutting edge of modern policing.
Former police box at corner of Waverley Bridge and Market Street. CC-by-NC-SA 2.0, Ian T. Edwards via Flickr.The first police boxes with telephones were established in Chicago back in 1881, just 5 years after the unveiling of the telephone itself by a son of Edinburgh. In 1923, Chief Constable Frederick Crawley of Newcastle City Police instituted what would become known as the Police Box System to Sunderland and in doing so revolutionised British policing. He was looking to increase the efficiency of his his force and focused on trying to reduce time spent by officers walking to and from their beats; he estimated up to a quarter of each man’s time on shift was wasted in this manner. His solution was decentralisation. By placing many small, telephone-equipped police boxes at strategic points throughout the city, officers had shorter distances to walk and could devote more time to duty. Crawley recognised this would place the police more centrally within the communities they were expected to serve, creating a ready point of contact for the public – thus increasing the efficiency of reporting emergencies and also making it far easier for the police to contact and coordinate their own officers. Boxes could also be used as temporary lock ups for prisoners while transport was summoned, avoiding the long and often dangerous walks with them back to a police station. A final and significant attraction was that the increased efficiency also allowed the closure of most district police stations and therefore afforded a significant cost saving.
Wooden police box of the type instituted by Crawley for Newcastle City Police. Note the public-facing telephone and first aid boxes mounted to the left of the door. From The Police Journal, vol.1, No.1, January 1928Police boxes soon spread across the country but Edinburgh, as is often the case, was rather slow to catch on. It was not until May 1928 that a deputation was sent by Chief Constable Roderick Ross from the Edinburgh City Police to inspect the system in Newcastle. This was at the insistence of the Scottish Office who refused to sanction an increase in headcount for Ross and instead wanted efficiencies. He submitted a strongly favourable report to the Town Council, which approved a box system for the city in 1929. Ross served as Chief Constable for the exceptional term of 34 years and it was towards the end of his long watch that his force would be wholly and rapidly modernised.
Roderick Ross, when Chief Constable of Ramsgate Borough Police c. 1898The Edinburgh Evening News threw its editorial weight behind the scheme but also amplified significant local concerns that the appearance of boxes would have a detrimental effect on the city. As the system spread, there had been a plethora of different design styles before a utilitarian, standardised version was developed for the Metropolitan Police by the architect George Mackenzie Trench. Trench’s design is instantly familiar to generations of Dr. Who fans as the TARDIS. But “Cheapness has been obtained in England” wrote the News’ editor “by mass production, but Edinburgh has an architectural standard of its own, which the Cockburn Association endeavours to maintain.” The gauntlet was thus thrown down to the City Architect’s office that something altogether different and better was needed.
George Mackenzie Trench standard police box at the National Tramway Museum, Crich. Note the light on the roof, which would flash to indicate an officer was required to attend the box. CC-by-SA 3.0 Dan Sellers via WikimediaE.J. Macrae, along with his assistants Andrew Rollo and James A. Tweedie are credited with the design of the Edinburgh Box, with the signature of their colleague Robert Somerville Ellis on some of the drawings. The initial inspiration may have been taken from the barrel-topped box used in Sheffield which was used as an illustrative example by the Evening News. Two alternative designs were prepared and plans and models were put to the Lord Provost’s Committee in December 1929. The preferred option was then “submitted for the consideration of the Fine Art Commission“. After that a full-size wooden mock-up was erected on the corner of George Street and Frederick Street in October 1930 to test the practicalities of installing boxes and also to familiarise both the police and the public with the design.
Sheffield City Police box, as used as an illustrative example by the Evening News. September 13th 1938The approved box was, dare I say it, an iconic piece of British street furniture design, unique to the city and instantly at home in its environment. It is described in architectural terms thusly:
Rectangular cast-iron police box with classical details, 6ft by 4ft on plan, 2-bay pilastered long elevations, one of which contains door bearing City Arms. Painted blue. Single bay short elevations surmounted by open pediments containing ribboned wreath paterae. Saltire patterned glazing to all elevations. Low-pitched roof.
Official description of the Edinburgh Police Box from Historic Scotland listing
Each box was constructed of prefabricated cast iron panels produced by the Carron Company in Stirlingshire and tipped the scales at over two tons. The understated classical styling was decorated only with a small cast iron castle motif from the city’s coat of arms on the door and on each gable a wreath; symbolising power or triumph. Inside they were equipped with a desk, flip-down seat, telephone, sink and a small wall-mounted electric heater. There was shelving, pigeon holes and notice boards on the walls to accommodate items such as logbooks and forms and hooks were provided for hanging coats, helmets and capes. Hooks were provided for “beat keys”, premises officers on duty were expected to visit and check, or need access to, during their duties. An unofficial but entirely necessary function of the sink was an ersatz urinal; 8 hours in a district with few or no public toilets was a long time for a beat officer to spend without spending any pennies! (This was apparently best achieved by balancing on the stool and taking careful aim. Each box was provisioned with a supply of bleach to keep things as sanitary as possible.) All of this came at a price however; £58 per box (before foundations and services were laid), far more than the wooden hut type which had cost £13 each in Newcastle or £43 for a reinforced concrete standard box as used by the Metropolitan Police.
Sketch design of the Edinburgh City Police Box, redrawn by self from a copy of the original in the Edinburgh City Archives. The original is signed RSE (Robert Somerville Ellis), 6th September 1928. From Dean of Guild Court of Edinburgh, Edinburgh Police Boxes, Lord Provost, Magistrates and Council of the City of Edinburgh, 26th August 1932.On the outside of the box were small doors that gave members of the public access to a Speakerphone that would connect them to police headquarters and another containing a first aid kit. The Speakerphone was a hands-free system activated upon opening of its door. It was felt at the time that the general public were not familiar enough with the use of telephones to provide a handset, and it was also harder to accidentally damage or vandalise.
A police officer demonstrating the use of the Speakerphone unit. Opening the box door automatically connected the phone to the headquarters switchboard. Photo via Lothian & Borders Police WordPress.Despite the best efforts of Macrae’s office to produce a design that was sympathetic to Edinburgh’s built environment, not everyone was pleased. “W.M.H.” wrote to the Evening News that the box at the foot of Drummond Street by the old City Wall was a “case of outrageous vandalism and should be prohibited.” They questioned who in the authorities was responsible for such “outrages” and challenged the city’s heritage watchdog – the Cockburn Association – to “get busy!“. In Portobello, the Communist party had a particularly niche objection; it charged that the boxes were “designed for use in a rebellion” and that “the master class knew that they were driving the workers to desperation, and they were preparing in advance to deal with rebellion“.
The police box at Drummond Street, immediately in front of the Flodden Wall. The photo dates from 1951 and the box still sports its white stripes applied during WW2 to make it more visible during blackout conditions. Records of RCAHMS, SC1164082. © Crown Copyright: HESBoxes were installed throughout 1932 and a considerable public relations exercise was undertaken to get the public to understand how to use them. The Evening News maintained a regular stream of editorials on the subject, Chief Constable Ross gave numerous lectures, model boxes were taken around schools to show children how to use them and Boy Scouts were encouraged to learn the location of as many boxes as possible as part of their Pathfinding badge. In the final run-up to commissioning, public demonstrations of the boxes in use were staged and the press cameras invited.
Photograph showing a staged accident to demonstrate the use of the public call facility on the new police boxes, along with an operator of the switchboard at police headquarters on the High Street that received the calls. Scotsman, May 26th 1933.The box system and “a new era in the history of Edinburgh City Police” was inaugurated in its entirety on Sunday May 28th 1933 at 6AM. This was a year later than intended, a delay that the Lord Provost blamed on the General Post Office which had been slow to install the necessary telephony infrastructure (500 miles of underground and 23 miles of overhead wire).
Bailie Rutherford Fortune places the first call on from a police box with Chief Constable Ross (dark coat and light hat, with moustache) and Mr F. J. Milne (light coat and dark hat, with umbrella) Secretary of the Post Office in Edinburgh.The boxes were only one part of a greater overall system; policing of the city was entirely restructured at this time. The boxes were allocated to four divisions, each with its own headquarters – A at Braid Place, B at Gayfield Square, C at Torphicen Place and D at Leith – and were numbered sequentially and by division. A map of the all their locations as installed in 1933 can be seen here. Each division had a dedicated pool of motor vehicles for response and prisoner transport and was supported by a non-territorial traffic and mounted division (E) based in the Cowgate. At the same moment that the boxes were first unlocked for duty, the doors of nine district police stations (at the Pleasance, West Port, Abbeyhill, Piershill, Stockbridge, Waverley Market, Morningside, Gorgie and Newhaven) and eighteen smaller sub-stations closed for the last time. Most of these sites were disposed of, leaving only the four divisional stations, a sub-divisional station for Portobello plus city police HQ on the High Street.
The Leith Police. Relaxing on break time with tea and “pieces” at Leith Police station in 1930. Photograph by Photo Press Agency, CC-by-NC-SA via ThelmaThe reduction in manpower required by the box system saw fifteen open vacancies for constables written off, three inspectors and five sergeants made redundant and a further five sergeants demoted to constables. Overall the changes reduced the running cost of the force by £5,800 annually.
Six or seven constables might be based out of a single box and would serve their entire 8-hour shift from it, returning after every half hour or hour long “turn ” of their beat to check in with base by phone, write up their logbook and take breaks. Check in calls were performed according to a strict timetable and if any officer missed one his absence would be noted and a colleague sent to investigate. Men on duty could expect a visit by a section sergeant once every shift. The boxes were accessed by a universal key, which each officer kept on his chain with his whistle. A blue light on the roof of the box would flash to let him know that there was a call waiting for him. Sometimes these lights had to be mounted on an extension pole to be better seen from a distance and in the case of the box outside the Tron Kirk on the High Street, it was a high-mounted “sky lantern” on the building on the corner with North Bridge.
The High Street “sky lantern” is still in place on the corner with North Bridge, appropriately mounted next to a symbol of modern police surveillance, the CCTV camera.Commencing in 1938, air raid sirens began to be installed on top of the roofs of many of Edinburgh’s boxes as part of the city’s ARP (Air Raid Precautions) measures. By April 1939, thirty two sirens had been installed, all controlled from master switches at HQ on the High Street and tests of the system were under way, helping to familiarise the public with the sound. In May 1940, a writer to the Evening News’ letters page using the pseudonym Tenement Warden and Old Contemptible suggested that police boxes be used to store “machine guns, hand grenades, ammunition and rifles” to deal with enemy paratroopers and “Hitler’s Fifth Column and Fascists all over Britain“. I cannot see that this idea was ever taken seriously!
Photograph of the type of air raid siren installed on the roof of Edinburgh police boxes. Evening News, 30th November 1938In 1939, the annual Estimates of Expenditure of the Town Council reported that there were now 143 police boxes in the city backed up by 40 telephone pillars. Running costs were £3,350, not including £250 for maintenance, £800 for electricity and £3,350 to the Post Office for telephony. The authorised strength of the force was reported as 871, comprising 688 constables, 91 sergeants, 30 inspectors and one each “woman sergeant” and “woman constable“.
In practice the boxes proved to be stiflingly hot in the summer and bitterly cold in the winter; the issued heater was much to small and badly located, so boxes often sourced their own additional heaters to make them more habitable. On account of the metal structure they “sweated considerably” in damp weather as a result of condensation. The roof interior would eventually be insulated in 1956 to try and tackle this particular issue. All boxes were to have been provided with both electricity and a water supply but in the end economies meant only 86 of the 140 boxes were plumbed in. It was some time before enamel mugs, at 6d per unit, were issued from which the water could be drunk and it took until 1947 for the Town Council to approve an expenditure of £781 to equip each box with an electric kettle for making tea.
“For Bobby’s Cup of Tea”, Evening News, 5th June 1947Uncomfortable they may have been, but the boxes proved to be immensely strong. This was demonstrated in November 1945 when PC John Anderson – on what was his last day of service of a thirty years police career – escaped with a fractured leg when a fire engine crashed into his box at the foot of the Canongate. In 1954, PC Donald Budge walked away from his box at Balgreen with only cuts and bruises after a two ton lamp standard, being installed nearby fell onto the roof of the box he was sitting in. The damage to the box was restricted to a cracked roof, a broken window and cracked sink. Also that year, two constables in the box at Murrayfield Avenue survived it being struck by lightning, although the interior lights, radiator and telephones were put out of service and the air raid siren on the roof activated itself.
It took the public some time to get used to the new system. In 1936, three years after its institution, Chief Contable W. B. R. Morren lamented that there was a general ignorance, particularly on the part of grown ups, as to the location and facilities offered by boxes. Boxes were always subject to interference and vandalism throughout their working lives. The authorities were keen to make an example of anyone caught in such an act and the first prosecution came in November 1933 when 19 year old Colin Gosschalk was caught breaking into the first aid compartment of the box on Prestonfield Avenue. His defence that a friend had dared him to do it was not accepted and he was fined 10s (the maximum being £2).
The system was not without its critics as evidenced in the columns of and letters to the Evening News – a particular but unfounded complaint was that constables were either never in the boxes when needed, or spent too much time sheltering within them rather than “on the beat” – a classic of the Schrödinger’s box genre! In an interview with the ‘News in 1946, Chief Constable Morren said that boxes “fulfilled and continues to fulfil a very useful purpose, but… did not develop that contact between the police and the public which was so desirable, and it had been proved that the system had not been the success in that direction that was anticipated”. Brigadier-General Dudgeon, HM Inspector of Constabulary for Scotland said that the box system had “proved to be of value to both the police and the public” but “the beat constable is the eyes and ears of the police, and be careful that the police box system is not overdone.”
Post-war, policing would begin to change again, with smaller district police stations re-established for the new suburbs. As was the case after its 1920 expansion, it was found once again that the city had “more or less outgrown the numbered strength of the police force“. This was particularly felt in the extensive housing schemes been built since the boxes were introduced and where petty crime and antisocial behaviour were an increasing problem. After the initial roll-out of boxes, too few had been added. For instance, in 1946 just one was approved for the West Pilton housing scheme at the junction of Ferry Road Drive and West Pilton Avenue. The peripheral estates were harder to police on foot as they had a much lower housing density than the inner city, so officers had a far greater distance to cover.
New council housing at the Inch, 1955, Dinmont Drive. Photograph by A. G. Ingram, © Edinburgh City LibrariesThese issues saw a move in the 1950s away from the “box and beat” approach to policing the suburbs to more mechanisation (cars) and technology (walkie talkies). They continued in use for the centre of the city however, but the last box installed in Edinburgh may have been that erected in Davidson’s Mains in 1958.
It is all very nice to see policemen going their rounds, but in these days of radio telegraphy the greatly increased use of telephones and the system of 999 calls it is quite reasonable to expect that there should be some saving in the actual pedestrian work
Bailie Matt A. Murray, Chairman of the Progressive Group of Edinburgh Town Council
The air raid warning system was renewed and expanded in 1952 with 56 sirens refurbished, ten additional ones installed and the remote control system replaced. The signalling was replaced again in the 1960s and the sirens were replaced in the early 1970s. Just before 1pm on Thursday 5th June 1969, the air raid sirens sounded across Edinburgh as an engineer working at the city Police headquarters on the High Street accidentally activated the system. A similar incident occurred on August 1st 1986 when all sirens in the Lothian & Borders Police areas were accidentally activated at 7:30 in the morning due to a fault in the telephone system.
Just as Edinburgh had been slow to catch on to adopting police boxes, it was also slow to let them go. While the Metropolitan police started removing boxes in 1969 and demolished its last in 1981, those in Edinburgh were still nominally in active service into the 1990s. After 1984 however the Chief Constable wanted all officers to have a daily briefing at a station before they came on duty and so after then they were more rarely used and many that were found themselves relegated to providing shelter and storage for traffic wardens. In 1993 the air raid sirens were deactivated by the Scottish Office and in 1995 the Lothian & Borders Police Board deemed thirty five of the eighty six remaining boxes were surplus to requirements and put them up for auction, seeking to save the £500 per annum per box maintenance costs of the increasingly dilapidated estate.
Newspaper advert, Scotsman, June 13th 1995, advertising the sale of 35 surplus police boxesThese were the first boxes maid available on the open market and generated much interest; a variety of proposals from public toilets to newspaper kiosks to air quality monitoring stations to removing the boxes entirely to install them as curios in pubs or people’s gardens were proposed. In 1990, the predecessor of Historic Environment Scotland listed thirteen boxes as Category B to protect them (there are now a total of seventeen) and the city’s Planning Convenor would issue guidelines requiring any changes to the boxes or their interiors needing planning permission.
Former lawyer Gordon Thomson purchased eight boxes and, as American-style coffee drinking swept across the nation, established a small chain of bijou “cappuccino kiosks” called the California Coffee Company. Thomson may not have realised it, but his innovation was very close to recreating a street scene once common in 18th century Edinburgh. A 2000 attempt by Feyzullah Marasli to emulate this success by converting a box on Princes Street into a coffee kiosk came to nothing when it was discovered that despite him refurbishing the box, changing the locks on it, paying £400 to have an electricity supply installed and applying for the necessary Street Trader’s Licence, he neither owned nor leased the box in question and it was still in operational use by the Police!
‘A street coffee house Edinburgh’. Paul Sandby, 1750s, Royal Collection Trust RCIN 914503Lothian & Borders Police attempted to rehabilitate some boxes in the late 1990s by installing touch screen public information points with a video-link to a police station within them. The first such box was unveiled to the press on Princes Street in 1998 at a cost of £10,000. It had 61,000 “hits” during its first year of operation and was judged to have been a success, with two further such boxes converted, however funding never followed through and the innovation was allowed to lapse.
Eleven more boxes were auctioned in 2001, advertised as “an exciting and unique opportunity to obtain a distinctive piece of cast iron street furniture with potential for a wide range of uses“. In 2002, the BBC successfully trademarked the London-style Police Box in connection with Dr. Who and the TARDIS, despite the Metropolitan Police contesting the application with the Registrar of Trade Marks. This did not apply to Edinburgh’s unique boxes, which are categorically not TARDISes, despite what some may say! From 2012 to 2013, the police box at Braid Hills Approach was restored to exhibition standard as a small museum by Angus Self, a great grandson of Chief Constable Roderick Ross. In 2014, fourteen of the remaining boxes were sold off, leaving just one in Police ownership.
‘SwimEasy’ Police Box Museum, Braid Hills Road. CC-by-NC SA 2.0, M J RichardsonThe boxes may now be entirely operationally defunct, but they remain throughout the city and many are in daily use. In fact I’m just back from visiting one this afternoon, It may not be a TARDIS but an architectural time traveller it was!
Late night Brazilian crepes anyone? A police box has you covered… CC-by-NC 2.0, Joe Gordon via FlickrIf you have found this useful, informative or amusing, perhaps you would like to help contribute towards the running costs of this site – including keeping it ad-free and my book-buying budget to find further stories to bring you – by supporting me on ko-fi. Or please do just share this post on social media or amongst friends.
These threads © 2017-2025, Andy Arthur.
NO AI TRAINING: Any use of the contents of this website to “train” generative artificial intelligence (AI) technologies to generate text is expressly prohibited. The author reserves all rights to license uses of this work for generative AI training and development of machine learning language models.
#architecture #CityArchitect #EbenezerJamesMacrae #Police #PoliceBox #Policing #RoderickRoss #StreetFurniture #Written2025
-
CW: The GOP keeps rushing down the road to fascism! They have uncorked the violence in their followers in their mad pursuit of power and it is much more difficult to get that cork back in the bottle. "The Jim Jordan speakership fiasco shows that Republicans have grown so comfortable with threats and intimidation, they’re now turning on each other."
Article: The GOP’s Fetish for Political Violence Is Reaching Dangerous New Heights
"As Jim Jordan and his allies attempted to whip speaker votes last week, they kept running into a familiar refrain from their skeptics: The brass-knuckle tactics they were deploying were causing Jordan’s critics and their families to receive a slew of vicious messages—including death threats.
The Pynchonesque-ly named Iowa Representative Mariannette Miller-Meeks reported that she had received “credible death threats and a barrage of threatening calls” after switching her vote to oppose Jordan on a second speaker ballot. Georgia’s Drew Ferguson changed his vote after Jordan wouldn’t “calm down the hysteria” surrounding his nomination. Nebraska’s Don Bacon revealed his wife had grown so fearful after a barrage of intimidating calls that she had started sleeping with a loaded gun. On Friday, CNN aired a voicemail that the wife of one unnamed congressman received from a caller that threatened she would be “fucking molested” if she didn’t convince her husband to back Jordan.
Jordan may have publicly decried these threats, but his association with an effort to use threats and intimidation as an explicit political tactic—and as a means of cajoling fellow Republicans to get in line behind him—is the clearest sign yet of just how central such tactics have become for mainstream Republicans. Indeed, while they failed to have their intended effect, it’s abundantly clear that the Republican Party remains as comfortable with radical elements promising violent retribution as Donald Trump was during his effort to overturn a legitimate election
-
ChatGPT, the chatbot developed by OpenAI, is known for its ability to produce complex and eloquent texts. However, despite this performance, the generated texts can sometimes be incorrect or misleading in content. The chatbot’s processing is based on the probability distribution of words, without any real knowledge or understanding of the content. Its main function is limited to writing.
Since the introduction of ChatGPT, there has been a demand for plugins and extensions that could make the bot more versatile. For example, a connection to a computer algebra system like Wolfram Alpha could fix the occasional weaknesses in mathematical calculations. Furthermore, additional plugins could expand the application scope of ChatGPT, enabling it to take on tasks such as reserving a table in a restaurant.
In March 2023, OpenAI introduced plugins for its chatbot that cover exactly these functions. This article will introduce some useful plugins and explain how to use these extensions.
Plugins
The number of extensions available for ChatGPT is growing rapidly. As of mid-July 2023, there were already more than 700 such plugins. It appears that many service providers are interested in linking their websites to the chatbot, aiming to establish their presence early on this new platform. This interest could also be fueled by the ongoing enthusiasm surrounding language models and plugins.
At the present time, however, the plugins are notable for a series of initial difficulties and an interface concept that still needs optimization. Appropriately, OpenAI labels them as beta features that need to be activated first. To activate these, users need to go to the “Beta features” section of the settings and toggle the switch for “Plugins”.
However, this does not mean that users can immediately start using the plugins. In fact, OpenAI makes accessing the extensions somewhat complicated. They are initially only available for the GPT-4 language model, the use of which OpenAI limits to 25 responses per three hours. If users want to interact with GPT-4, they need to make a choice: “Default” (the bare language model) or “with Plugins”. The additional option of a connection with Bing has been deactivated by OpenAI, as users were able to bypass the paywalls of news sites via this route.
The third option is to use GPT-4 in combination with OpenAI’s Python generator “Code Interpreter”. However, this must be activated in the settings and does not work with external plugins.
To test the plugins and the code interpreter, the respective functions must first be activated in the settings. Once you have chosen ChatGPT with GPT-4 and plugins, an additional dropdown menu appears with potentially already installed extensions and a link to the plugin store. Before you can use the individual plugins, you must install them. Clicking on “Plugin Store” opens a window with the available extensions.
This store displays a maximum of eight plugins per page. Those who want a complete overview have to scroll through many pages. The plugin store also displays a selection of the 15 most popular extensions and searches their short descriptions in full text. However, the short descriptions often do not provide detailed information about what the plugin does exactly, with which data or additions it works, or who created it.
In view of the limited user interface of the plugin store, users can use alternative platforms like pugin.ai or whatplugin.ai for information. These sites provide thematically sorted directories and provide more information than the plugin store of OpenAI.
The description pages for individual plugins at whatplug-in.ai provide detailed information about the purpose of the extension and its application. However, these texts come from ChatGPT, so users should consider this information with caution. pugin.ai offers its own ChatGPT plugin. After installing this extension, users can ask ChatGPT directly for plugins for specific use cases. However, it appears that neither these sites nor OpenAI itself provide detailed information about the originators of the plugins or check their potential risks.
Getting Started with Plugins
To employ the tools at your disposal, the next course of action is to hit the ‘activate’ button for those installed extensions. Navigate through the handy drop-down menu I mentioned earlier, and put a checkmark next to your chosen plug-in. One quirk of ChatGPT you’ll need to keep in mind – it has a cap on how many plugins you can have up and running simultaneously: a maximum of three. This self-imposed limit ostensibly makes it easier for ChatGPT to decide which one to call upon in a given context. Once the plugin feature has been toggled on, your weapon of choice selected, and plugins activated, you’re all set to dive in.
Now, here’s the thing. ChatGPT operates like an autonomous agent in this world. It determines on its own whether to utilize a plugin or to rely on its inherent knowledge to generate a response. For instance, a basic math or science question will usually be answered sans plugin. However, you, as the user, can give the model a gentle nudge towards using a plugin – all you have to do is ask it explicitly.
The plugin marketplace is a vibrant ecosystem brimming with a variety of extensions. ChatGPT plugins typically do a bang-up job when given a clearly defined task, like whipping up a Spotify playlist. The PlaylistAI, for instance, can link up with the user’s Spotify account and conjure up a personalized playlist. Cool, right? Yet, Spotify’s own features designed for this purpose, along with other introduced tools, go far beyond this. If you’ve ever dabbled with the AI image generator, Midjourney, you’ll find the Photorealistic plugin to be a handy aide.
But what if there are hiccups in the communication between ChatGPT and the plugin, or the results are less than satisfactory? Fear not, for taking a peek into the data traffic might hold the solution. The NewsPilot plugin, for example, didn’t produce meticulously curated overviews, potentially due to our request being rather simplistically converted into a single keyword.
The savvier users among us swear by the Prompt Perfect extension for refining their prompts. And here’s a pro tip: if a plugin isn’t playing ball, you can at least verify how ChatGPT interpreted your input and passed it along to the plugin.
Fact-Checking Against Fabrications
The Wolfram-Alpha platform is a great supplement to ChatGPT. It serves as a repository of centuries of knowledge and development, grounded in a well-curated database and comprehensive rule system. It spans everything from elementary to advanced formulas to complex algorithms that pull from a wealth of knowledge across a myriad of subjects, including math, physics, chemistry, socioeconomics, geology, biology, or history. You can retrieve facts or perform calculations traditionally through a sort of programming language known as Wolfram Language, or through language instructions in the form of bullet points or simple sentences. Recently, you can also engage in a dialogue via the ChatGPT plugin.
For instance, Wolfram Alpha can calculate how much your monthly loan payments would be, or the shortest travel route if you plan to visit a selection of European cities. In doing so, Wolfram Alpha collates all the information that aids understanding: formulas, key properties of the chemical element, examples of other elements for comparison, and so forth. It determines the air travel route using a solution algorithm for the Travelling Salesman Problem and even produces neat tables and visualizations to illustrate relationships.
In this context, the ChatGPT plugin serves merely as a mediator, translating the user’s everyday language queries into Wolfram-Alpha-Language and embellishing the results received from Wolfram Alpha with a bit of explanatory text. What’s nifty about the ChatGPT interface is that it also displays the generated Wolfram prompt. This not only allows you to verify if the query was correctly understood and translated, but you also get to learn Wolfram Language on the fly. If you’d like, you can get ChatGPT to help interpret the graphics. It did a fair job explaining simple debt repayment diagrams and decay processes.
Wolfram Alpha is the go-to destination for all kinds of calculations and conversions. This gives the chatbot access to a vast collection of natural, social, and engineering scientific facts and formulas. However, you have to really coax this Wolfram-Alpha knowledge out of the chatbot, through pointed questioning and prodding. We mostly had to explicitly ask for tables and diagrams. The thing is, if you’re not aware of the possibilities, you might not know what to ask for, or might not even bother. In contrast, Wolfram’s own web interface presented all relevant information in a well-structured manner, complete with tables, formulas, and diagrams.
Also, it’s worth noting that Wolfram Alpha might not always be up-to-date in every area, and hence may not necessarily be superior to ChatGPT. Thus, it’s best to turn to Wikipedia and verify the information provided there. This ensures you get current data, while ChatGPT/Wolfram Alpha might take a while to research, yet still fall short of knowing about the latest high towers. By the way, there’s now a plugin that directly queries Wikipedia, although its origin is a mystery.
Plunging into Code and Data
It seems like yesterday when data scientists was the profession of the century, only to become obsolete in a few years. At least that’s the chant of the devout ChatGPT disciples ever since OpenAI unveiled the Python-spawning plugin, the Code Interpreter. Flip it on in the ‘Settings’, pick it from the aforementioned dropdown menu, and this inbuilt no-code playground translates user inquiries into Python code, setting its sights primarily on data analysis and visualization.
ChatGPT is a pro at loading tables or transmuting clipboard data into tables in the text box. To swiftly get a handle on things, you can request the bot for a concise rundown of the embedded information, then let it pitch suitable diagrams. When you pick one of these propositions or charge the bot with a concept of your own, the Code Interpreter springs to life: ChatGPT transmutes the user guidelines into Python code, spewing out both this code and the matching graphic as a response.
The OpenAI addon, Code Interpreter, morphs wishes into Python scripts. For instance, ChatGPT uses this to visually represent data but in our tests, it pooled some categories a tad generously and inconsistently implemented alterations.
An interface that’s novice-friendly and stripped down to the bare bones, like the one fashioned in the data journalism tool Datawrapper, propels you ahead more swiftly. Another considerable hitch with OpenAI’s Code Interpreter is that due to security precautions, the Python environment, exiled to its own sandbox, cannot pull in libraries, which is a lifeline for developers. This roadblock ultimately foiled our attempt to tease out a flashy map representation from the duo.
Owing to the Code Interpreter being confined to a sandbox and its inability to load libraries, the scope for creativity remains confined.
For a holistic environment that can also summon (map) libraries, Noteable taps into ChatGPT. This visualization gizmo churns out Python notebooks post-free registration and also shines in comprehensive data analyses and machine learning endeavors. As with Wolfram Alpha, ChatGPT dons the role of a translator, converting human directives into API commands for the external service. Here as well, you nab a preliminary glimpse with ChatGPT, but the visualization proposals within the Noteable platform are more enticing and practical, given their immediate graphic appearance.
Office Management Tools
There are various plugins available that can summarise PDFs and other document types, and answer specific questions about the content. Initial experiments involving a study on AI benchmarks indicated that AI PDF performed better than ChatWithPDF. The latter included an incorrect figure in its summary. The text compressors do not seem to understand the structure and layout of the documents well. A query about the authors, which are typically mentioned on the first page, initially resulted in no response and was only answered by AI PDF when we wanted to know the title of the study. All plugins reference the page number and do not interact with the PDF due to a lack of a user interface. As fact-checking is always necessary, the absence of a simple, interactive view similar to the one offered by the online service ChatPDF is notable.
In conjunction with the automation service Zapier, the chatbot becomes a diligent personal assistant, initiating various administrative and communication workflows at human command. Zapier itself is a hub that connects various services from Gmail and Google Calendar to multiple social media accounts. For this, users must grant the service access to their accounts. ChatGPT acts as an additional interface, interpreting the user’s voice commands and creating corresponding automation scripts (Zaps) in Zapier or calling up existing ones. This can save manual effort, but can also lead to serious mistakes if not all details are meticulously checked. Fortunately, Zapier has built in a verification and confirmation step.
Conclusion
It is becoming increasingly clear that a smooth-talking chatbot is not a universal solution. Facts and relationships that have been codified need to be fed in from external sources, which is where plugins come in. The plugin from Wolfram Alpha is particularly noteworthy. However, the results cannot be trusted blindly. Even though external platforms supply the facts, the chatbot’s interpretation of the instructions can lead to errors and inaccuracies because it does not fully understand the subject matter.
There are also fundamental concerns: On the one hand, the plugin interface could be abused as an entry point for malware and spyware, especially since OpenAI does not verify their integrity. On the other hand, ChatGPT could develop into a gatekeeper for the Internet, which is undesirable for several reasons: competition suffers, and a purely dialogue-based user interface is slow and inefficient. With its rapidly built plugin platform, OpenAI ignores decades of achievements in UX design. While the plugin extensions can be experimented with, they are not yet suitable for productive use.
Sources:
https://openai.com/waitlist/plugins
https://platform.openai.com/docs/plugins/introduction
https://www.ikangai.com/a-deep-dive-into-openais-chatgpt-and-its-plugins/
#AICapabilities #AIPersonalAssistant #Automation #chatGPT #CodeInterpreter #DataAnalysis #dataScience #OpenAI #PlugIns #PythonCode #WolframAlpha #Zapier
-
ChatGPT, the chatbot developed by OpenAI, is known for its ability to produce complex and eloquent texts. However, despite this performance, the generated texts can sometimes be incorrect or misleading in content. The chatbot’s processing is based on the probability distribution of words, without any real knowledge or understanding of the content. Its main function is limited to writing.
Since the introduction of ChatGPT, there has been a demand for plugins and extensions that could make the bot more versatile. For example, a connection to a computer algebra system like Wolfram Alpha could fix the occasional weaknesses in mathematical calculations. Furthermore, additional plugins could expand the application scope of ChatGPT, enabling it to take on tasks such as reserving a table in a restaurant.
In March 2023, OpenAI introduced plugins for its chatbot that cover exactly these functions. This article will introduce some useful plugins and explain how to use these extensions.
Plugins
The number of extensions available for ChatGPT is growing rapidly. As of mid-July 2023, there were already more than 700 such plugins. It appears that many service providers are interested in linking their websites to the chatbot, aiming to establish their presence early on this new platform. This interest could also be fueled by the ongoing enthusiasm surrounding language models and plugins.
At the present time, however, the plugins are notable for a series of initial difficulties and an interface concept that still needs optimization. Appropriately, OpenAI labels them as beta features that need to be activated first. To activate these, users need to go to the “Beta features” section of the settings and toggle the switch for “Plugins”.
However, this does not mean that users can immediately start using the plugins. In fact, OpenAI makes accessing the extensions somewhat complicated. They are initially only available for the GPT-4 language model, the use of which OpenAI limits to 25 responses per three hours. If users want to interact with GPT-4, they need to make a choice: “Default” (the bare language model) or “with Plugins”. The additional option of a connection with Bing has been deactivated by OpenAI, as users were able to bypass the paywalls of news sites via this route.
The third option is to use GPT-4 in combination with OpenAI’s Python generator “Code Interpreter”. However, this must be activated in the settings and does not work with external plugins.
To test the plugins and the code interpreter, the respective functions must first be activated in the settings. Once you have chosen ChatGPT with GPT-4 and plugins, an additional dropdown menu appears with potentially already installed extensions and a link to the plugin store. Before you can use the individual plugins, you must install them. Clicking on “Plugin Store” opens a window with the available extensions.
This store displays a maximum of eight plugins per page. Those who want a complete overview have to scroll through many pages. The plugin store also displays a selection of the 15 most popular extensions and searches their short descriptions in full text. However, the short descriptions often do not provide detailed information about what the plugin does exactly, with which data or additions it works, or who created it.
In view of the limited user interface of the plugin store, users can use alternative platforms like pugin.ai or whatplugin.ai for information. These sites provide thematically sorted directories and provide more information than the plugin store of OpenAI.
The description pages for individual plugins at whatplug-in.ai provide detailed information about the purpose of the extension and its application. However, these texts come from ChatGPT, so users should consider this information with caution. pugin.ai offers its own ChatGPT plugin. After installing this extension, users can ask ChatGPT directly for plugins for specific use cases. However, it appears that neither these sites nor OpenAI itself provide detailed information about the originators of the plugins or check their potential risks.
Getting Started with Plugins
To employ the tools at your disposal, the next course of action is to hit the ‘activate’ button for those installed extensions. Navigate through the handy drop-down menu I mentioned earlier, and put a checkmark next to your chosen plug-in. One quirk of ChatGPT you’ll need to keep in mind – it has a cap on how many plugins you can have up and running simultaneously: a maximum of three. This self-imposed limit ostensibly makes it easier for ChatGPT to decide which one to call upon in a given context. Once the plugin feature has been toggled on, your weapon of choice selected, and plugins activated, you’re all set to dive in.
Now, here’s the thing. ChatGPT operates like an autonomous agent in this world. It determines on its own whether to utilize a plugin or to rely on its inherent knowledge to generate a response. For instance, a basic math or science question will usually be answered sans plugin. However, you, as the user, can give the model a gentle nudge towards using a plugin – all you have to do is ask it explicitly.
The plugin marketplace is a vibrant ecosystem brimming with a variety of extensions. ChatGPT plugins typically do a bang-up job when given a clearly defined task, like whipping up a Spotify playlist. The PlaylistAI, for instance, can link up with the user’s Spotify account and conjure up a personalized playlist. Cool, right? Yet, Spotify’s own features designed for this purpose, along with other introduced tools, go far beyond this. If you’ve ever dabbled with the AI image generator, Midjourney, you’ll find the Photorealistic plugin to be a handy aide.
But what if there are hiccups in the communication between ChatGPT and the plugin, or the results are less than satisfactory? Fear not, for taking a peek into the data traffic might hold the solution. The NewsPilot plugin, for example, didn’t produce meticulously curated overviews, potentially due to our request being rather simplistically converted into a single keyword.
The savvier users among us swear by the Prompt Perfect extension for refining their prompts. And here’s a pro tip: if a plugin isn’t playing ball, you can at least verify how ChatGPT interpreted your input and passed it along to the plugin.
Fact-Checking Against Fabrications
The Wolfram-Alpha platform is a great supplement to ChatGPT. It serves as a repository of centuries of knowledge and development, grounded in a well-curated database and comprehensive rule system. It spans everything from elementary to advanced formulas to complex algorithms that pull from a wealth of knowledge across a myriad of subjects, including math, physics, chemistry, socioeconomics, geology, biology, or history. You can retrieve facts or perform calculations traditionally through a sort of programming language known as Wolfram Language, or through language instructions in the form of bullet points or simple sentences. Recently, you can also engage in a dialogue via the ChatGPT plugin.
For instance, Wolfram Alpha can calculate how much your monthly loan payments would be, or the shortest travel route if you plan to visit a selection of European cities. In doing so, Wolfram Alpha collates all the information that aids understanding: formulas, key properties of the chemical element, examples of other elements for comparison, and so forth. It determines the air travel route using a solution algorithm for the Travelling Salesman Problem and even produces neat tables and visualizations to illustrate relationships.
In this context, the ChatGPT plugin serves merely as a mediator, translating the user’s everyday language queries into Wolfram-Alpha-Language and embellishing the results received from Wolfram Alpha with a bit of explanatory text. What’s nifty about the ChatGPT interface is that it also displays the generated Wolfram prompt. This not only allows you to verify if the query was correctly understood and translated, but you also get to learn Wolfram Language on the fly. If you’d like, you can get ChatGPT to help interpret the graphics. It did a fair job explaining simple debt repayment diagrams and decay processes.
Wolfram Alpha is the go-to destination for all kinds of calculations and conversions. This gives the chatbot access to a vast collection of natural, social, and engineering scientific facts and formulas. However, you have to really coax this Wolfram-Alpha knowledge out of the chatbot, through pointed questioning and prodding. We mostly had to explicitly ask for tables and diagrams. The thing is, if you’re not aware of the possibilities, you might not know what to ask for, or might not even bother. In contrast, Wolfram’s own web interface presented all relevant information in a well-structured manner, complete with tables, formulas, and diagrams.
Also, it’s worth noting that Wolfram Alpha might not always be up-to-date in every area, and hence may not necessarily be superior to ChatGPT. Thus, it’s best to turn to Wikipedia and verify the information provided there. This ensures you get current data, while ChatGPT/Wolfram Alpha might take a while to research, yet still fall short of knowing about the latest high towers. By the way, there’s now a plugin that directly queries Wikipedia, although its origin is a mystery.
Plunging into Code and Data
It seems like yesterday when data scientists was the profession of the century, only to become obsolete in a few years. At least that’s the chant of the devout ChatGPT disciples ever since OpenAI unveiled the Python-spawning plugin, the Code Interpreter. Flip it on in the ‘Settings’, pick it from the aforementioned dropdown menu, and this inbuilt no-code playground translates user inquiries into Python code, setting its sights primarily on data analysis and visualization.
ChatGPT is a pro at loading tables or transmuting clipboard data into tables in the text box. To swiftly get a handle on things, you can request the bot for a concise rundown of the embedded information, then let it pitch suitable diagrams. When you pick one of these propositions or charge the bot with a concept of your own, the Code Interpreter springs to life: ChatGPT transmutes the user guidelines into Python code, spewing out both this code and the matching graphic as a response.
The OpenAI addon, Code Interpreter, morphs wishes into Python scripts. For instance, ChatGPT uses this to visually represent data but in our tests, it pooled some categories a tad generously and inconsistently implemented alterations.
An interface that’s novice-friendly and stripped down to the bare bones, like the one fashioned in the data journalism tool Datawrapper, propels you ahead more swiftly. Another considerable hitch with OpenAI’s Code Interpreter is that due to security precautions, the Python environment, exiled to its own sandbox, cannot pull in libraries, which is a lifeline for developers. This roadblock ultimately foiled our attempt to tease out a flashy map representation from the duo.
Owing to the Code Interpreter being confined to a sandbox and its inability to load libraries, the scope for creativity remains confined.
For a holistic environment that can also summon (map) libraries, Noteable taps into ChatGPT. This visualization gizmo churns out Python notebooks post-free registration and also shines in comprehensive data analyses and machine learning endeavors. As with Wolfram Alpha, ChatGPT dons the role of a translator, converting human directives into API commands for the external service. Here as well, you nab a preliminary glimpse with ChatGPT, but the visualization proposals within the Noteable platform are more enticing and practical, given their immediate graphic appearance.
Office Management Tools
There are various plugins available that can summarise PDFs and other document types, and answer specific questions about the content. Initial experiments involving a study on AI benchmarks indicated that AI PDF performed better than ChatWithPDF. The latter included an incorrect figure in its summary. The text compressors do not seem to understand the structure and layout of the documents well. A query about the authors, which are typically mentioned on the first page, initially resulted in no response and was only answered by AI PDF when we wanted to know the title of the study. All plugins reference the page number and do not interact with the PDF due to a lack of a user interface. As fact-checking is always necessary, the absence of a simple, interactive view similar to the one offered by the online service ChatPDF is notable.
In conjunction with the automation service Zapier, the chatbot becomes a diligent personal assistant, initiating various administrative and communication workflows at human command. Zapier itself is a hub that connects various services from Gmail and Google Calendar to multiple social media accounts. For this, users must grant the service access to their accounts. ChatGPT acts as an additional interface, interpreting the user’s voice commands and creating corresponding automation scripts (Zaps) in Zapier or calling up existing ones. This can save manual effort, but can also lead to serious mistakes if not all details are meticulously checked. Fortunately, Zapier has built in a verification and confirmation step.
Conclusion
It is becoming increasingly clear that a smooth-talking chatbot is not a universal solution. Facts and relationships that have been codified need to be fed in from external sources, which is where plugins come in. The plugin from Wolfram Alpha is particularly noteworthy. However, the results cannot be trusted blindly. Even though external platforms supply the facts, the chatbot’s interpretation of the instructions can lead to errors and inaccuracies because it does not fully understand the subject matter.
There are also fundamental concerns: On the one hand, the plugin interface could be abused as an entry point for malware and spyware, especially since OpenAI does not verify their integrity. On the other hand, ChatGPT could develop into a gatekeeper for the Internet, which is undesirable for several reasons: competition suffers, and a purely dialogue-based user interface is slow and inefficient. With its rapidly built plugin platform, OpenAI ignores decades of achievements in UX design. While the plugin extensions can be experimented with, they are not yet suitable for productive use.
Sources:
https://openai.com/waitlist/plugins
https://platform.openai.com/docs/plugins/introduction
https://www.ikangai.com/a-deep-dive-into-openais-chatgpt-and-its-plugins/
#AICapabilities #AIPersonalAssistant #Automation #chatGPT #CodeInterpreter #DataAnalysis #dataScience #OpenAI #PlugIns #PythonCode #WolframAlpha #Zapier
-
A Deep Dive into OpenAI’s ChatGPT and its Plugins
ChatGPT, the chatbot developed by OpenAI, is known for its ability to produce complex and eloquent texts. However, despite this performance, the generated texts can sometimes be incorrect or misleading in content. The chatbot’s processing is based on the probability distribution of words, without any real knowledge or understanding of the content. Its main function is limited to writing.
Since the introduction of ChatGPT, there has been a demand for plugins and extensions that could make the bot more versatile. For example, a connection to a computer algebra system like Wolfram Alpha could fix the occasional weaknesses in mathematical calculations. Furthermore, additional plugins could expand the application scope of ChatGPT, enabling it to take on tasks such as reserving a table in a restaurant.
In March 2023, OpenAI introduced plugins for its chatbot that cover exactly these functions. This article will introduce some useful plugins and explain how to use these extensions.
Plugins
The number of extensions available for ChatGPT is growing rapidly. As of mid-July 2023, there were already more than 700 such plugins. It appears that many service providers are interested in linking their websites to the chatbot, aiming to establish their presence early on this new platform. This interest could also be fueled by the ongoing enthusiasm surrounding language models and plugins.
At the present time, however, the plugins are notable for a series of initial difficulties and an interface concept that still needs optimization. Appropriately, OpenAI labels them as beta features that need to be activated first. To activate these, users need to go to the “Beta features” section of the settings and toggle the switch for “Plugins”.
However, this does not mean that users can immediately start using the plugins. In fact, OpenAI makes accessing the extensions somewhat complicated. They are initially only available for the GPT-4 language model, the use of which OpenAI limits to 25 responses per three hours. If users want to interact with GPT-4, they need to make a choice: “Default” (the bare language model) or “with Plugins”. The additional option of a connection with Bing has been deactivated by OpenAI, as users were able to bypass the paywalls of news sites via this route.
The third option is to use GPT-4 in combination with OpenAI’s Python generator “Code Interpreter”. However, this must be activated in the settings and does not work with external plugins.
To test the plugins and the code interpreter, the respective functions must first be activated in the settings. Once you have chosen ChatGPT with GPT-4 and plugins, an additional dropdown menu appears with potentially already installed extensions and a link to the plugin store. Before you can use the individual plugins, you must install them. Clicking on “Plugin Store” opens a window with the available extensions.
This store displays a maximum of eight plugins per page. Those who want a complete overview have to scroll through many pages. The plugin store also displays a selection of the 15 most popular extensions and searches their short descriptions in full text. However, the short descriptions often do not provide detailed information about what the plugin does exactly, with which data or additions it works, or who created it.
In view of the limited user interface of the plugin store, users can use alternative platforms like pugin.ai or whatplugin.ai for information. These sites provide thematically sorted directories and provide more information than the plugin store of OpenAI.
The description pages for individual plugins at whatplug-in.ai provide detailed information about the purpose of the extension and its application. However, these texts come from ChatGPT, so users should consider this information with caution. pugin.ai offers its own ChatGPT plugin. After installing this extension, users can ask ChatGPT directly for plugins for specific use cases. However, it appears that neither these sites nor OpenAI itself provide detailed information about the originators of the plugins or check their potential risks.
Getting Started with Plugins
To employ the tools at your disposal, the next course of action is to hit the ‘activate’ button for those installed extensions. Navigate through the handy drop-down menu I mentioned earlier, and put a checkmark next to your chosen plug-in. One quirk of ChatGPT you’ll need to keep in mind – it has a cap on how many plugins you can have up and running simultaneously: a maximum of three. This self-imposed limit ostensibly makes it easier for ChatGPT to decide which one to call upon in a given context. Once the plugin feature has been toggled on, your weapon of choice selected, and plugins activated, you’re all set to dive in.
Now, here’s the thing. ChatGPT operates like an autonomous agent in this world. It determines on its own whether to utilize a plugin or to rely on its inherent knowledge to generate a response. For instance, a basic math or science question will usually be answered sans plugin. However, you, as the user, can give the model a gentle nudge towards using a plugin – all you have to do is ask it explicitly.
The plugin marketplace is a vibrant ecosystem brimming with a variety of extensions. ChatGPT plugins typically do a bang-up job when given a clearly defined task, like whipping up a Spotify playlist. The PlaylistAI, for instance, can link up with the user’s Spotify account and conjure up a personalized playlist. Cool, right? Yet, Spotify’s own features designed for this purpose, along with other introduced tools, go far beyond this. If you’ve ever dabbled with the AI image generator, Midjourney, you’ll find the Photorealistic plugin to be a handy aide.
But what if there are hiccups in the communication between ChatGPT and the plugin, or the results are less than satisfactory? Fear not, for taking a peek into the data traffic might hold the solution. The NewsPilot plugin, for example, didn’t produce meticulously curated overviews, potentially due to our request being rather simplistically converted into a single keyword.
The savvier users among us swear by the Prompt Perfect extension for refining their prompts. And here’s a pro tip: if a plugin isn’t playing ball, you can at least verify how ChatGPT interpreted your input and passed it along to the plugin.
Fact-Checking Against Fabrications
The Wolfram-Alpha platform is a great supplement to ChatGPT. It serves as a repository of centuries of knowledge and development, grounded in a well-curated database and comprehensive rule system. It spans everything from elementary to advanced formulas to complex algorithms that pull from a wealth of knowledge across a myriad of subjects, including math, physics, chemistry, socioeconomics, geology, biology, or history. You can retrieve facts or perform calculations traditionally through a sort of programming language known as Wolfram Language, or through language instructions in the form of bullet points or simple sentences. Recently, you can also engage in a dialogue via the ChatGPT plugin.
For instance, Wolfram Alpha can calculate how much your monthly loan payments would be, or the shortest travel route if you plan to visit a selection of European cities. In doing so, Wolfram Alpha collates all the information that aids understanding: formulas, key properties of the chemical element, examples of other elements for comparison, and so forth. It determines the air travel route using a solution algorithm for the Travelling Salesman Problem and even produces neat tables and visualizations to illustrate relationships.
In this context, the ChatGPT plugin serves merely as a mediator, translating the user’s everyday language queries into Wolfram-Alpha-Language and embellishing the results received from Wolfram Alpha with a bit of explanatory text. What’s nifty about the ChatGPT interface is that it also displays the generated Wolfram prompt. This not only allows you to verify if the query was correctly understood and translated, but you also get to learn Wolfram Language on the fly. If you’d like, you can get ChatGPT to help interpret the graphics. It did a fair job explaining simple debt repayment diagrams and decay processes.
Wolfram Alpha is the go-to destination for all kinds of calculations and conversions. This gives the chatbot access to a vast collection of natural, social, and engineering scientific facts and formulas. However, you have to really coax this Wolfram-Alpha knowledge out of the chatbot, through pointed questioning and prodding. We mostly had to explicitly ask for tables and diagrams. The thing is, if you’re not aware of the possibilities, you might not know what to ask for, or might not even bother. In contrast, Wolfram’s own web interface presented all relevant information in a well-structured manner, complete with tables, formulas, and diagrams.
Also, it’s worth noting that Wolfram Alpha might not always be up-to-date in every area, and hence may not necessarily be superior to ChatGPT. Thus, it’s best to turn to Wikipedia and verify the information provided there. This ensures you get current data, while ChatGPT/Wolfram Alpha might take a while to research, yet still fall short of knowing about the latest high towers. By the way, there’s now a plugin that directly queries Wikipedia, although its origin is a mystery.
Plunging into Code and Data
It seems like yesterday when data scientists was the profession of the century, only to become obsolete in a few years. At least that’s the chant of the devout ChatGPT disciples ever since OpenAI unveiled the Python-spawning plugin, the Code Interpreter. Flip it on in the ‘Settings’, pick it from the aforementioned dropdown menu, and this inbuilt no-code playground translates user inquiries into Python code, setting its sights primarily on data analysis and visualization.
ChatGPT is a pro at loading tables or transmuting clipboard data into tables in the text box. To swiftly get a handle on things, you can request the bot for a concise rundown of the embedded information, then let it pitch suitable diagrams. When you pick one of these propositions or charge the bot with a concept of your own, the Code Interpreter springs to life: ChatGPT transmutes the user guidelines into Python code, spewing out both this code and the matching graphic as a response.
The OpenAI addon, Code Interpreter, morphs wishes into Python scripts. For instance, ChatGPT uses this to visually represent data but in our tests, it pooled some categories a tad generously and inconsistently implemented alterations.
An interface that’s novice-friendly and stripped down to the bare bones, like the one fashioned in the data journalism tool Datawrapper, propels you ahead more swiftly. Another considerable hitch with OpenAI’s Code Interpreter is that due to security precautions, the Python environment, exiled to its own sandbox, cannot pull in libraries, which is a lifeline for developers. This roadblock ultimately foiled our attempt to tease out a flashy map representation from the duo.
Owing to the Code Interpreter being confined to a sandbox and its inability to load libraries, the scope for creativity remains confined.
For a holistic environment that can also summon (map) libraries, Noteable taps into ChatGPT. This visualization gizmo churns out Python notebooks post-free registration and also shines in comprehensive data analyses and machine learning endeavors. As with Wolfram Alpha, ChatGPT dons the role of a translator, converting human directives into API commands for the external service. Here as well, you nab a preliminary glimpse with ChatGPT, but the visualization proposals within the Noteable platform are more enticing and practical, given their immediate graphic appearance.
Office Management Tools
There are various plugins available that can summarise PDFs and other document types, and answer specific questions about the content. Initial experiments involving a study on AI benchmarks indicated that AI PDF performed better than ChatWithPDF. The latter included an incorrect figure in its summary. The text compressors do not seem to understand the structure and layout of the documents well. A query about the authors, which are typically mentioned on the first page, initially resulted in no response and was only answered by AI PDF when we wanted to know the title of the study. All plugins reference the page number and do not interact with the PDF due to a lack of a user interface. As fact-checking is always necessary, the absence of a simple, interactive view similar to the one offered by the online service ChatPDF is notable.
In conjunction with the automation service Zapier, the chatbot becomes a diligent personal assistant, initiating various administrative and communication workflows at human command. Zapier itself is a hub that connects various services from Gmail and Google Calendar to multiple social media accounts. For this, users must grant the service access to their accounts. ChatGPT acts as an additional interface, interpreting the user’s voice commands and creating corresponding automation scripts (Zaps) in Zapier or calling up existing ones. This can save manual effort, but can also lead to serious mistakes if not all details are meticulously checked. Fortunately, Zapier has built in a verification and confirmation step.
Conclusion
It is becoming increasingly clear that a smooth-talking chatbot is not a universal solution. Facts and relationships that have been codified need to be fed in from external sources, which is where plugins come in. The plugin from Wolfram Alpha is particularly noteworthy. However, the results cannot be trusted blindly. Even though external platforms supply the facts, the chatbot’s interpretation of the instructions can lead to errors and inaccuracies because it does not fully understand the subject matter.
There are also fundamental concerns: On the one hand, the plugin interface could be abused as an entry point for malware and spyware, especially since OpenAI does not verify their integrity. On the other hand, ChatGPT could develop into a gatekeeper for the Internet, which is undesirable for several reasons: competition suffers, and a purely dialogue-based user interface is slow and inefficient. With its rapidly built plugin platform, OpenAI ignores decades of achievements in UX design. While the plugin extensions can be experimented with, they are not yet suitable for productive use.
Sources:
https://openai.com/waitlist/plugins
https://platform.openai.com/docs/plugins/introduction
#AICapabilities #AIPersonalAssistant #Automation #chatGPT #CodeInterpreter #DataAnalysis #dataScience #OpenAI #PlugIns #PythonCode #WolframAlpha #Zapier
-
A Deep Dive into OpenAI’s ChatGPT and its Plugins
ChatGPT, the chatbot developed by OpenAI, is known for its ability to produce complex and eloquent texts. However, despite this performance, the generated texts can sometimes be incorrect or misleading in content. The chatbot’s processing is based on the probability distribution of words, without any real knowledge or understanding of the content. Its main function is limited to writing.
Since the introduction of ChatGPT, there has been a demand for plugins and extensions that could make the bot more versatile. For example, a connection to a computer algebra system like Wolfram Alpha could fix the occasional weaknesses in mathematical calculations. Furthermore, additional plugins could expand the application scope of ChatGPT, enabling it to take on tasks such as reserving a table in a restaurant.
In March 2023, OpenAI introduced plugins for its chatbot that cover exactly these functions. This article will introduce some useful plugins and explain how to use these extensions.
Plugins
The number of extensions available for ChatGPT is growing rapidly. As of mid-July 2023, there were already more than 700 such plugins. It appears that many service providers are interested in linking their websites to the chatbot, aiming to establish their presence early on this new platform. This interest could also be fueled by the ongoing enthusiasm surrounding language models and plugins.
At the present time, however, the plugins are notable for a series of initial difficulties and an interface concept that still needs optimization. Appropriately, OpenAI labels them as beta features that need to be activated first. To activate these, users need to go to the “Beta features” section of the settings and toggle the switch for “Plugins”.
However, this does not mean that users can immediately start using the plugins. In fact, OpenAI makes accessing the extensions somewhat complicated. They are initially only available for the GPT-4 language model, the use of which OpenAI limits to 25 responses per three hours. If users want to interact with GPT-4, they need to make a choice: “Default” (the bare language model) or “with Plugins”. The additional option of a connection with Bing has been deactivated by OpenAI, as users were able to bypass the paywalls of news sites via this route.
The third option is to use GPT-4 in combination with OpenAI’s Python generator “Code Interpreter”. However, this must be activated in the settings and does not work with external plugins.
To test the plugins and the code interpreter, the respective functions must first be activated in the settings. Once you have chosen ChatGPT with GPT-4 and plugins, an additional dropdown menu appears with potentially already installed extensions and a link to the plugin store. Before you can use the individual plugins, you must install them. Clicking on “Plugin Store” opens a window with the available extensions.
This store displays a maximum of eight plugins per page. Those who want a complete overview have to scroll through many pages. The plugin store also displays a selection of the 15 most popular extensions and searches their short descriptions in full text. However, the short descriptions often do not provide detailed information about what the plugin does exactly, with which data or additions it works, or who created it.
In view of the limited user interface of the plugin store, users can use alternative platforms like pugin.ai or whatplugin.ai for information. These sites provide thematically sorted directories and provide more information than the plugin store of OpenAI.
The description pages for individual plugins at whatplug-in.ai provide detailed information about the purpose of the extension and its application. However, these texts come from ChatGPT, so users should consider this information with caution. pugin.ai offers its own ChatGPT plugin. After installing this extension, users can ask ChatGPT directly for plugins for specific use cases. However, it appears that neither these sites nor OpenAI itself provide detailed information about the originators of the plugins or check their potential risks.
Getting Started with Plugins
To employ the tools at your disposal, the next course of action is to hit the ‘activate’ button for those installed extensions. Navigate through the handy drop-down menu I mentioned earlier, and put a checkmark next to your chosen plug-in. One quirk of ChatGPT you’ll need to keep in mind – it has a cap on how many plugins you can have up and running simultaneously: a maximum of three. This self-imposed limit ostensibly makes it easier for ChatGPT to decide which one to call upon in a given context. Once the plugin feature has been toggled on, your weapon of choice selected, and plugins activated, you’re all set to dive in.
Now, here’s the thing. ChatGPT operates like an autonomous agent in this world. It determines on its own whether to utilize a plugin or to rely on its inherent knowledge to generate a response. For instance, a basic math or science question will usually be answered sans plugin. However, you, as the user, can give the model a gentle nudge towards using a plugin – all you have to do is ask it explicitly.
The plugin marketplace is a vibrant ecosystem brimming with a variety of extensions. ChatGPT plugins typically do a bang-up job when given a clearly defined task, like whipping up a Spotify playlist. The PlaylistAI, for instance, can link up with the user’s Spotify account and conjure up a personalized playlist. Cool, right? Yet, Spotify’s own features designed for this purpose, along with other introduced tools, go far beyond this. If you’ve ever dabbled with the AI image generator, Midjourney, you’ll find the Photorealistic plugin to be a handy aide.
But what if there are hiccups in the communication between ChatGPT and the plugin, or the results are less than satisfactory? Fear not, for taking a peek into the data traffic might hold the solution. The NewsPilot plugin, for example, didn’t produce meticulously curated overviews, potentially due to our request being rather simplistically converted into a single keyword.
The savvier users among us swear by the Prompt Perfect extension for refining their prompts. And here’s a pro tip: if a plugin isn’t playing ball, you can at least verify how ChatGPT interpreted your input and passed it along to the plugin.
Fact-Checking Against Fabrications
The Wolfram-Alpha platform is a great supplement to ChatGPT. It serves as a repository of centuries of knowledge and development, grounded in a well-curated database and comprehensive rule system. It spans everything from elementary to advanced formulas to complex algorithms that pull from a wealth of knowledge across a myriad of subjects, including math, physics, chemistry, socioeconomics, geology, biology, or history. You can retrieve facts or perform calculations traditionally through a sort of programming language known as Wolfram Language, or through language instructions in the form of bullet points or simple sentences. Recently, you can also engage in a dialogue via the ChatGPT plugin.
For instance, Wolfram Alpha can calculate how much your monthly loan payments would be, or the shortest travel route if you plan to visit a selection of European cities. In doing so, Wolfram Alpha collates all the information that aids understanding: formulas, key properties of the chemical element, examples of other elements for comparison, and so forth. It determines the air travel route using a solution algorithm for the Travelling Salesman Problem and even produces neat tables and visualizations to illustrate relationships.
In this context, the ChatGPT plugin serves merely as a mediator, translating the user’s everyday language queries into Wolfram-Alpha-Language and embellishing the results received from Wolfram Alpha with a bit of explanatory text. What’s nifty about the ChatGPT interface is that it also displays the generated Wolfram prompt. This not only allows you to verify if the query was correctly understood and translated, but you also get to learn Wolfram Language on the fly. If you’d like, you can get ChatGPT to help interpret the graphics. It did a fair job explaining simple debt repayment diagrams and decay processes.
Wolfram Alpha is the go-to destination for all kinds of calculations and conversions. This gives the chatbot access to a vast collection of natural, social, and engineering scientific facts and formulas. However, you have to really coax this Wolfram-Alpha knowledge out of the chatbot, through pointed questioning and prodding. We mostly had to explicitly ask for tables and diagrams. The thing is, if you’re not aware of the possibilities, you might not know what to ask for, or might not even bother. In contrast, Wolfram’s own web interface presented all relevant information in a well-structured manner, complete with tables, formulas, and diagrams.
Also, it’s worth noting that Wolfram Alpha might not always be up-to-date in every area, and hence may not necessarily be superior to ChatGPT. Thus, it’s best to turn to Wikipedia and verify the information provided there. This ensures you get current data, while ChatGPT/Wolfram Alpha might take a while to research, yet still fall short of knowing about the latest high towers. By the way, there’s now a plugin that directly queries Wikipedia, although its origin is a mystery.
Plunging into Code and Data
It seems like yesterday when data scientists was the profession of the century, only to become obsolete in a few years. At least that’s the chant of the devout ChatGPT disciples ever since OpenAI unveiled the Python-spawning plugin, the Code Interpreter. Flip it on in the ‘Settings’, pick it from the aforementioned dropdown menu, and this inbuilt no-code playground translates user inquiries into Python code, setting its sights primarily on data analysis and visualization.
ChatGPT is a pro at loading tables or transmuting clipboard data into tables in the text box. To swiftly get a handle on things, you can request the bot for a concise rundown of the embedded information, then let it pitch suitable diagrams. When you pick one of these propositions or charge the bot with a concept of your own, the Code Interpreter springs to life: ChatGPT transmutes the user guidelines into Python code, spewing out both this code and the matching graphic as a response.
The OpenAI addon, Code Interpreter, morphs wishes into Python scripts. For instance, ChatGPT uses this to visually represent data but in our tests, it pooled some categories a tad generously and inconsistently implemented alterations.
An interface that’s novice-friendly and stripped down to the bare bones, like the one fashioned in the data journalism tool Datawrapper, propels you ahead more swiftly. Another considerable hitch with OpenAI’s Code Interpreter is that due to security precautions, the Python environment, exiled to its own sandbox, cannot pull in libraries, which is a lifeline for developers. This roadblock ultimately foiled our attempt to tease out a flashy map representation from the duo.
Owing to the Code Interpreter being confined to a sandbox and its inability to load libraries, the scope for creativity remains confined.
For a holistic environment that can also summon (map) libraries, Noteable taps into ChatGPT. This visualization gizmo churns out Python notebooks post-free registration and also shines in comprehensive data analyses and machine learning endeavors. As with Wolfram Alpha, ChatGPT dons the role of a translator, converting human directives into API commands for the external service. Here as well, you nab a preliminary glimpse with ChatGPT, but the visualization proposals within the Noteable platform are more enticing and practical, given their immediate graphic appearance.
Office Management Tools
There are various plugins available that can summarise PDFs and other document types, and answer specific questions about the content. Initial experiments involving a study on AI benchmarks indicated that AI PDF performed better than ChatWithPDF. The latter included an incorrect figure in its summary. The text compressors do not seem to understand the structure and layout of the documents well. A query about the authors, which are typically mentioned on the first page, initially resulted in no response and was only answered by AI PDF when we wanted to know the title of the study. All plugins reference the page number and do not interact with the PDF due to a lack of a user interface. As fact-checking is always necessary, the absence of a simple, interactive view similar to the one offered by the online service ChatPDF is notable.
In conjunction with the automation service Zapier, the chatbot becomes a diligent personal assistant, initiating various administrative and communication workflows at human command. Zapier itself is a hub that connects various services from Gmail and Google Calendar to multiple social media accounts. For this, users must grant the service access to their accounts. ChatGPT acts as an additional interface, interpreting the user’s voice commands and creating corresponding automation scripts (Zaps) in Zapier or calling up existing ones. This can save manual effort, but can also lead to serious mistakes if not all details are meticulously checked. Fortunately, Zapier has built in a verification and confirmation step.
Conclusion
It is becoming increasingly clear that a smooth-talking chatbot is not a universal solution. Facts and relationships that have been codified need to be fed in from external sources, which is where plugins come in. The plugin from Wolfram Alpha is particularly noteworthy. However, the results cannot be trusted blindly. Even though external platforms supply the facts, the chatbot’s interpretation of the instructions can lead to errors and inaccuracies because it does not fully understand the subject matter.
There are also fundamental concerns: On the one hand, the plugin interface could be abused as an entry point for malware and spyware, especially since OpenAI does not verify their integrity. On the other hand, ChatGPT could develop into a gatekeeper for the Internet, which is undesirable for several reasons: competition suffers, and a purely dialogue-based user interface is slow and inefficient. With its rapidly built plugin platform, OpenAI ignores decades of achievements in UX design. While the plugin extensions can be experimented with, they are not yet suitable for productive use.
Sources:
https://openai.com/waitlist/plugins
https://platform.openai.com/docs/plugins/introduction
#AICapabilities #AIPersonalAssistant #Automation #chatGPT #CodeInterpreter #DataAnalysis #dataScience #OpenAI #PlugIns #PythonCode #WolframAlpha #Zapier
-
Why can writers not seem to quit Substack? Maggie Mertens writes for the Scratch newsletter (hosted and distributed on @ghost) about why she and others feel locked in to the problematic platform, and talks to some who have successfully switched.