#myai — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #myai, aggregated by home.social.
-
A Fitness Discussion with MyAI about Silos, Running, Cycling, and Strava
Reading Time: 4 minutesYears ago we had bulky GPS watches that we wore only when we were hiking or doing sports. They only tracked our speed, location, and elevation. With time they were made less hideous and more socially acceptable. We could go from wearing them when doing sports, to when doing nothing.
With time wrist HR monitoring and step counters were added and so these devices added to the quantified self habit. They counted how much we walked, how long we were still, and how how heart was doing for the entire day.
Companies like Apple and Google gameified our quantified self. You must stand for one minute an hour. You must be active for half an hour a day. You must burn 300 kcals per day. They went from watches that we could wear just when we were doing sports, to watches that quantified our activities 24/7. With gameificiation they almost forced us to wear their devices to feed their data silos.
Garmin has its silo. Suunto has another, and Apple has a third. The result is that, in theory, if you have devices for all three brands you need to wear all three devices simultaneously to feed each data silo.
In reality Suunto is happy to get data from the Apple watch via Sportstracker and it’s just as happy to get a workout file from Garmin. In essence Suunto is the ‘silo’ that is friendliest with external data, which makes it one of the most interesting. The interest comes from having the flexibility to wear a Garmin, Apple or Suunto device. All of them can be ingested.
With Garmin, if you import a GPX file it will log the distance, and other metrics, but will not count fatigue and other variables. It reserves certain metrics to its own sources.
Apple is the most paradoical of all. You buy an expensive phone, and then you buy an expensive watch that lasts for one day, before needing a charge. If you buy the SE charging is slower than if you buy the more “luxurious” models.
The paradox with Apple comes from it ingesting data from Suunto, from Garmin, and from other sources, but refusing to use that data for the fitness app. This is especially absurd since it actually has all of our fitness data. They want to force us to use an Apple watch.
Fitness and Freshness
Apps such as Strava, Garmin, Suunto and Fitness all quantify our fitness. It counts our walks, our runs, our climbs, our skateboarding, our cycling, our hiking and more. It then tells us how hard we worked and how fatigued we are.
The issue is that this information is siloed by each brand. This means that if you work out for four months with Apple, and then wear a Garmin for two days, then your Apple fitness will drop, because you were inactive, and Garmin will say “You’re overdoing it, consider resting” because of the leap from zero activity to some activity.
These watches all track fitness and freshness in their own silos, so if you slide between Apple for three days, and then suunto for two, and Garmin for one, let’s say because you walked with Apple, ran with Suunto, and cycled with Garmin, then each app sees a different fitness picture. It’s only by wearing your device(s) full time that you get a full picture, and not just full time, but for weeks, months or even years for an accurate evaluation of fatigue and exhaustion.
On the Topic of Sports Related fatigue
With Strava you can look at Fitness and Freshness overall, as well as by sport, whether running, swimming, or other. If you’re willing to pay for Strava Premium, then you can look at this data, whilst being platform independent, since data from Suunto, Garmin, Apple and many other sources is aggregated. The more data you add, the more complete the image. In essence you can track with a single device at a time, and get a good oversight, once you pay for premium.
The Apple Silo
The paradox is that Strava and Apple both have acess to the same user data, but whereas Apple chooses only to use its own data, Strava uses everyone’s data. If Apple opened up the Fitness App then it would compete directly with Strava for that niche.
In reality Suunto and Sportstracker are in a good position to compete with Strava and the Apple Fitness app because of how easy it is to import data from multiple sources. The key drawback is that it is a manual, rather than automated process, for now.
Why Would you Have Multiple Devices
There are people who go for a bike ride wearing an Apple watch. Other people who do a lot of hiking might have a Garmin Instinct or a Garmin Instinct 2. I use this device as an example because it’s cheap. Others might have a Garmin Fenix 7 or 7s because they have the budget and they buy one device every few years, so it’s justifiable to splurge. I noticed that one or three people use the Suunto App directly.
Others have Wahoo, Garmin, Cateye or other cycling specific devices that they use for navigation and climb information. That’s why I upgraded from an Explore to an Explore 2. Over time, as you switch devices you might switch brands, and that’s when you’re across two or more silos, and that’s when you have to decide whether to start from scratch or straddle two or more ecosystems. Tools exist to help you merge, in theory, but if that then propagates then you might end up with hundreds, or thousands of duplicates, with no practical way of tidying up.
And Finally
My Apple Watch SE battery was so depleted a few months ago that it became unusable. Luckily I noticed when it was protected by Apple Care+ so replacing the battery was free. The side effect is that now the watch should be alive and well for another two or more years, hence having to decide which devices to prioritise.
In Conlusion
I know that “and finally” and “in Conclusion” are the same thing but I feel like embelishing. For a long time we used GPS watches as GPS watches, but with time, as they tracked the quantified self, and fitness progression so the need for complete data grew. That’s why wearing three watches simultaneously stops being absurd and starts being logical. The issue is that when we slide between platforms we lose continuity if we don’t wear devices A, B, and C. That’s where Strava give us the flexibility to slide between platforms, without losing the geostationary satellite view of our fitness progression.
It is absurd to wear two or three watches, but if the data is siloed, then either we’re eccentric, or we depend on paywall features from apps like Strava.
#AI #Apple #connect #connect #garminc #myai #suunto #swisscom -
OwnAether Personal AI Operating Systems- What if your entire digital life — your work, your income, your creativity, your health, your automation, your business — was orchestrated by a single intelligent layer that learns you, works for you, and evolves with you?
https://medium.com/@ownaether/the-personal-ai-os-revolution-37a475922aea
#AI #PersonalAI #IndividualAI #MyAI #YourAI #LocalAI #DesktopAI #AIApps #PrivateAI #LLMs #SLMs #AIModels #PersonalAIAssistant #PersonalAIApp
-
OwnAether Personal AI Operating Systems- What if your entire digital life — your work, your income, your creativity, your health, your automation, your business — was orchestrated by a single intelligent layer that learns you, works for you, and evolves with you?
https://medium.com/@ownaether/the-personal-ai-os-revolution-37a475922aea
#AI #PersonalAI #IndividualAI #MyAI #YourAI #LocalAI #DesktopAI #AIApps #PrivateAI #LLMs #SLMs #AIModels #PersonalAIAssistant #PersonalAIApp
-
OwnAether Personal AI Operating Systems- What if your entire digital life — your work, your income, your creativity, your health, your automation, your business — was orchestrated by a single intelligent layer that learns you, works for you, and evolves with you?
https://medium.com/@ownaether/the-personal-ai-os-revolution-37a475922aea
#AI #PersonalAI #IndividualAI #MyAI #YourAI #LocalAI #DesktopAI #AIApps #PrivateAI #LLMs #SLMs #AIModels #PersonalAIAssistant #PersonalAIApp
-
OwnAether Personal AI Operating Systems- What if your entire digital life — your work, your income, your creativity, your health, your automation, your business — was orchestrated by a single intelligent layer that learns you, works for you, and evolves with you?
https://medium.com/@ownaether/the-personal-ai-os-revolution-37a475922aea
#AI #PersonalAI #IndividualAI #MyAI #YourAI #LocalAI #DesktopAI #AIApps #PrivateAI #LLMs #SLMs #AIModels #PersonalAIAssistant #PersonalAIApp
-
OwnAether Personal AI Operating Systems- What if your entire digital life — your work, your income, your creativity, your health, your automation, your business — was orchestrated by a single intelligent layer that learns you, works for you, and evolves with you?
https://medium.com/@ownaether/the-personal-ai-os-revolution-37a475922aea
#AI #PersonalAI #IndividualAI #MyAI #YourAI #LocalAI #DesktopAI #AIApps #PrivateAI #LLMs #SLMs #AIModels #PersonalAIAssistant #PersonalAIApp
-
An Eleventy Strava Experiment in Three Parts
Reading Time: 3 minutesFor the last two days I have been experimenting with reproducing my Strava timeline in Eleventy using a CSV file, for a change, as well as a JavaScript file to import new GPX files and map them correctly for the CSV file. That CSV file is then read by Eleventy to populate a site.
Part 1. Getting the Data
Initially I was going to use the API. Due to the API site not working I switched to exporting my strava history and downloading it. I created eleventy-strava and then added _data/raw and that’s where the csv files live. I have activities, privacy, bikes, shoes, and one or two other csv files.
With a few lines of JavaScript and Nunjuck I am able to extract the photos, the title, average heart rate, max heart rate and the comment/description for each activity.
The result is that within four seconds my entire Strava history is created, going all the way back to 2008.
Using MyAI and Gemini
When I was starting the experiment I ran out of tokens with MyAI almost instantly so I fell back to Gemini. For the intial setup Gemini worked fantastically, but when I wanted to add more functionality it provided me with erroneous code over, and over again so my enthusiasm has waned. We’re dealing with a lot of data, and I expect that by sharing data too generously I got Gemini narced, due to the pressure of all that data. I’m alluding to nitrogen narcosis and scuba diving in this image, not DEA activities.
GPX Imports
As I mentioned before, I’m experimenting with a static CSV file, rather than the API so the site is “dead”, in that data is refreshed only when I refresh it. That’s why I have the GPX import.js file. The idea is simple. I go for a run, a walk, a hike, a bike ride, or another sport and I come home. The activity is uploaded from my Suunto, Garmin, Apple watch or other to Strava and creates an individual activity. This activity has a strava ID and I give it a title.
The next step is that I download the gpx file from Strava. I rename it activity-id-name-of-activity.gpx, and my import.js file sees the id and title, parses, it, and then populates the CSV file with the name, title, hr info and more, in the appropriate 101 fields, before Eleventy rebuilds the site with the latest activity.
That’s when clever things happen
Automatic Hyperlinking to Strava
If I wanted to use GPX files directly, then I’d have several thousand data points per activity, several million over twenty years. If I wantd to map my activities it would take time, and effort. Instead I have my activities automatically connect with their Strava couterpart. This means that you have the Eleventy version that is light, fast, and easy to access and search, and the Strava version that is great for seeing the activity’s map and more. I’ve offloaded part of the work.
Part 2. Future Pairing With my Static Blog
For now the logic for my activities and my blog is seperate but I could very easily see myself pairing my daily blog post with the daily sporting activity. As you read the day’s post, you could then be drawn to the day’s sporting activity, and vice versa. With a build time of four seconds this doesn’t add much to the overhead of building my eleventy-blog site.
Part 3. Plenty of Flexibility
For now I have a summary page where I can see total distance for each sport where the distance isn’t 0km as well as the total climbed meters per sport. In theory, according to the data I have travelled 40,821km and climbed four hundred and sixty five metres. I could add information about average per year and more.
Proper Search
One of the key things missing from Strava, and Komoot, is the ability to search one’s own activities to find hikes. Several times I tried to find the Lammenhütte hike without much luck. With the flexibility offered by Eleventy I could add pagefind, and add search functionality quite easily, with filters for location and more. Even without search I wrote “lammen” and within a few milliseconds it found the lammenhutte hike that would take half an hour, using Strava.
And Finally
I’m only two days into this project. I could add monthly and yearly summaries for each sport, as well as searchability, and integration with my blog. I could also integrate my scuba diving log.
There are opportunities for me to explore.
#eleventy #gemini #JSON #myai #strava -
Refactoring Eleventy with Swisscom MyAI
Reading Time: 3 minutesYesterday afternoon I decided to experiment with MyAI and Eleventy because I was getting an error message. The issue is that when I refactored my Eleventy files with Gemini it removed code that broke the build process. I asked MyAI for help and it went beyond what I asked of it. Specifically, not only did it help me to fix the bug but it also helped me go from a build time of 35+ seconds, down to 25 seconds, and eventually down to 15 seconds. When you’re running build for each blog post that soon saves a lot of time.
A testament to MyAI’s Improvement
When I started playing with MyAI several months ago it felt quite limited, but yesterday the opposite is true. Today it feels as though MyAI could be serious competition for the US based AI options.
Early Adopter with Pro
It’s worth noting that because I signed up for MyAI Pro as an early adopter I was included within the MyAI Pro free of charge until 31.12.2026. When I was about to unsubscribe from pro I noticed the “free” opportunity and grabbed it.
It’s worth noting that I was using the fast bunny mode, not the “thoughtful pro” mode.
Asking for Clarity
I like that the model didn’t seem to hallucinate. I especially like that it asked for file A, and when it saw references to file B and C it asked for those, to understand the context of the code I was working with. As a result it kept things tidy and within scope.
It noticed loops that were called thousands of times, wastefully, and it refactored and optimised them with a minimum of effort.When Gemini saw the legacy code in Latin-8 it wanted to create to go through all of the blog posts on each build. As a human I thought “let’s do this just once”, but not Gemini. MyAI created a file, and within a few iterations it had fixed all of the errors. In one case we used a SED command for a single exception.
The result was clean markdown, and a far more efficient build process.
When it noticed that reading time was called several thousand times it refactored the code so that it was mapped, and thus called half as often.
AI As Micromanager
Working with AI is quite ironic. Last night, when I was pair coding with MyAI there was a moment when I thought, “It’s behaving like a boss with it’s ‘now change this, and that, and then these two things’.” We could almost say that MyAI was micromanaging me. In the end, when we had fixed the reading time problem I told MyAI “It’s 18:57, I’m calling it a day” and MyAI said “See you tomorrow — the ‘image’ shortcode is next.”
MyAI personal Summary
Great progress today:
? Fixed encoding in 443+ source files ? Cleaned up filters.js with proper memoization ? Moved excerpt to computed data ? Eliminated the O(n2)O(n^2)O(n2) related posts loop ? Removed redundant readingTime from base.njkSelf-Perception
When asked whether we had refactored its response was:
Performance Refactoring
Moving expensive operations out of render loops Precomputing data at build time instead of runtime Memoizing repeated filter callsData Quality Refactoring
Fixing encoding at the source rather than patching it at render timeThe classic definition applies perfectly here:
“Improving the internal structure of code without changing its external behaviour”And Finally
This was a positive experience for me. The character of the AI agent didn’t get on my nerves like Gemini often does, and I didn’t hit the token wall like I do with Euria, Claude and other solutions and that does make a difference, when you can make real progress. It’s also worth noting that I was using the fast, which I assume is the free option, rather than the “thoughtful pro” model.
What I saw yesterday afternoon, into the evening, leads me to believe that MyAI will be a useful companion for AI related tasks.
#AI #eleventy #myai #refactoring #swisscom -
Gobind: My-AI Standards To Serve As Trust Infrastructure For AI Development #ai #artificialintelligence #myai
-
A Year to Experiment with Swisscom MyAI
Reading Time: 2 minutesA few days ago I was looking at MyAI and I was going to unsubscribe from the "Pro" account before they charged me, before realising that as an early adopter I have a year to play with MyAI for "free". When I say that I can play with it for free I mean that I have been a Swisscom client for decades by now, so I paid for it through years of loyal custom.
Early Adoption of MyCloud
At the same time as an early adopter I played with Swisscom MyCloud when it was free, and eventually, when they switched to a paying solution I was caught within. Luckily I had been careful not to store terabytes of data on their service. I'm on the 200gb plan.
Based on Claude
MyAI runs with Anthropic's Claude and although the company is based in the US it has European offices in London, Dublin, Zurich, Paris and Munich. In contrast Euria by Infomaniak is based on open-source models such as Qwen3 and operates 'within an ethical Swiss-hosted infrastructure" according to itself(Euria). MyAI is less Swiss centric than Euria.
Playing With MyAI
I played with MyAI by asking what I should write about today and it gave a native answer, without looking outside of its model. When I asked it about a specific run that I did yesterday it provided me with a summary, after retrieving the data from Strava. When I provided it with a link to my profile it was able to answer questions like "How long would it take if I walked yesterday's route" as well as "how far do you think I will run this year. It answered that it could be between 1000-1300km.
When I asked it about Swisscom and Wingo it could provide plenty of information because both are Swisscom products. When I asked it about Infomaniak's Euria AI it didn't have much information and did not look for it.
When I asked it about the Swisscom WLAN box 2 it gave me an answer in German, so I told it to please remember to answer in French or English, and then it gave me the answer in English.
Worth Noting
As I looked through my conversation history with MyAI I notice that the first answers were sometimes in German, and that I had to ask for the answer in English. I also feel that as MyAI matures, so the quality of answers is improving.
And Finally
A few days ago I was going to switch back to the free tier to avoid being charged by mistake. I then noticed that for a few more days we can use the app for free for a year, so I am taking advantage of that opportunity. In my eyes I have been a Swisscom customer for long enough for to be justifiable.
It works well and provides us with an alternative to ChatGPT and other US based solutions.
#AI #european #myai #swiss #swisscom -
#Swisscom startet künstliche Intelligenz #myAI in der Schweiz - pctipp.ch https://www.pctipp.ch/news/kuenstliche-intelligenz/swisscom-startet-kuenstliche-intelligenz-myai-in-schweiz-2973039.html #ArtificialIntelligence #Swissness
-
Swisscom MyAI as tutor
Reading Time: 2 minutes
Yesterday I asked MyAI by Swisscom, which is still in Beta, if it could help me write a JavaScript app to generate passwords and it did, with ease. It provided me with the javascript code I needed so I could cut and paste it, and then use node to run it straight from terminal.
People often think that AI will replace us, and if certain companies in the US have their way, then it will. On the flipside I believe that AI will replace Googling for snippets of code that help us reach our coding goals.
Google Fu and AI
In many podcasts that I listened to over the years about JavaScript, Ruby On Rails and others I often heard the term Google Fu. It's the concept that people know what they want to do but they use Google to find and help write functions that help them achieve what they want.
In some cases it would take hours to find useful help when stuck on a problem. That's where skill and experience in searching for useful answers came in.
With AI, even with MyAI which is still in Beta, you can save hours of searching and trial and error. When I asked MyAI to write the app it provided me with code, which I could read and understand, and check for anomalies. Through studying a number of JavaScript courses I understand what the code is doing.
Tutoring
AI works as a tutor, where we can ask it to help us achieve what we want, and we can cut and paste things but we can also elaborate.
Eventually I asked about the limitations of this code, and whether I should use bcrypt or something similar so it added that in. It also added password hashing.
It provided me with three suggestions of how to use it, whether in the command line, via a web page or one or two other uses. In the end this took half an hour to an hour to do.
More than a Function at a Time
When I first started experimenting with AI I asked it to create one function at a time. Now I see that we can ask it to write an entire app, and tweak it through an iterative process to get it to do what we want it to do. AI can help save time.
Existing Libraries
While we can get AI to help us write a password generator we have the option of finding the right package. With Node generate-password is an existing tool that does the same thing. Whilst it's good to know how AI can help build apps from scratch it makes sense to see if an app already exists that can do what we want it to.
And Finally
Although US based AI models hog the limelight there are European solutions that offer some of the same assistance, without having to leave Swiss or European servers. We should not automatically go for ChatGPT, Claude and other solutions. Le Chat and MyAI provide good alternatives. We don't need to default to US based projects.
-
Playing with MyAI by Swisscom
Reading Time: < 1 minute
If we use Claude, Gemini, ChatGPT or a few other AI models we are using AI that has data centres in the US. If we use Le Chat by Mistral or MyAI (Beta) by Swisscom we are using AI that is based in Europe, or Switzerland. The data stays here.
The first thing to note is that MyAI is in public beta. It is still at a 0.0. version number rather than a 1. something version. As such this means that it can answer questions but we can't yet give it default data to work with like we can with Claude, Gemini and other AI solutions.
It cannot generate images or provide you with maps. It can answer a question such as "Could you help me write a password generator using javascript?" with ease. It is aware of Hugo, the open source project so could help you generate a theme.
It is available in French, German, Italian and English and I played with it in English.
It is good that Le Chat by Mistral, MyAI by Swisscom and AITools is available via Infomaniak. We have three options of local AI models to experiment with and use, and this will grow with time.
-
CNBC: Snap shares drop as FTC refers MyAI chatbot complaint to the DOJ. “Snap shares closed down 5% on Thursday after the Federal Trade Commission said it would refer a complaint against the company to the Department of Justice. The FTC’s non-public complaint involves allegations that Snapchat’s My AI chatbot poses ‘risks and harms to young users,’ the commission said in a statement.”
-
En zo snel gaat de introductie van LLMs dus..., Snapchat AI bot, https://nos.nl/artikel/2472026-een-chatbot-die-in-het-echt-wil-afspreken-ook-snapchat-vindt-het-wat-ver-gaan
-
سنابشات أطلقت #MyAI المبني على نماذج OpenAI.. وعدد كبير من المستخدمين مبسوطين عليه، سنابشات ما خلته مجاني للكل من فراغ، هي قاعدة عم تجمع البيانات لتحسين الخدمات والإعلانات فانتبهوا من مشاركة أي أمور شخصية.
الكلام مذكور في موقعهم بشكل صريح 👇🏻
-
Plenty of competition for #OpenAI's #ChatGPT arriving:
#Microsoft - #Bing
#Google - #Bard
#AnthropicAI - #Claude
#Salesforce - #EinsteinGPT
#Baidu_Inc - #Ernie
#Snapchat - #MyAI
#Meta - #Galactica
#Huawei - #PanguAlpha
#nvidia - #MegatronNLG -
Snapchat adds OpenAI-powered chatbot and proactively apologizes for what it might say
My AI will be available for Snapchat+ subscribers this week.
#Snapchat #MyAI #AI #OpenAI
https://www.engadget.com/snapchat-adds-openai-powered-chatbot-and-proactively-apologizes-for-what-it-might-say-180507261.html