Joe Wood

Useful summary of the #dataengineering stack. Missing for me are the extract, load tools like #airbyte, #meltano. Would have liked to see some governance tooling to support data mesh, catalog support too.

https://thenewstack.io/the-architects-guide-to-the-modern-data-stack/

#databuildtool

So it looks like the #binggpt #gpt #bingchat does parody Tweets (this was rumored to be part of the original demo, but we didn't see it). I asked it to do an #elonmusk impression. I think it's spot on:

“Just launched another Starship prototype. It exploded again, but that’s part of the learning process. We’ll get there eventually. Also, Dogecoin to the moon. #spacex #doge”

I'm in on the #bing #gpt #binggpt Beta. Super impressive and game changing not just for search, but for comparative research. The Bing integration is great, and the sidebar in Edge super useful.
The model looks to be up to date, but there's a gap. For example, I asked how are the political representatives for my area and the aggregate results were old (last year), but specific questions were correct (for the new session).

@phlox @andre #mkdocs is good. #docusaurus (version 2) worth a look at too. https://docusaurus.io/. If you want more control then maybe #Gatsby and the #remark components? #Sphinx also worth looking at (with a more Python slant).

@dantleech interested in understanding more people's experience with them. They have a special section in #backstageio. I've seen limited uptake. My concern is that ADRs would be seen as immutable - things change with fresh perspectives or changing ecosystem.

Giving #backstageio a quick try in the Christmas downtime. So far so good - impressed with how much work has been and the size of the ecosystem already.

Early thoughts about the need to think about degrading gracefully when critical systems are unavailable. I wonder if a Server Side Generated solution has been thought of as a fallback.