Luckily you can turn it off and use the standard ‘add’ workflow. I did that almost reflexively when I started trying to use jj. (snapshot.auto-track)

However, over time, and once I got the .gitignore fully set up for bigger projects, I’ve come around on re-enabling autocommit for more of my repos. It does flow pretty naturally once you have an established process. I find it enables both better ‘undo’, and more seamless context-switching.

You can also set a more specific snapshot.auto-track on a repo or user basis for personal tooling conventions that don’t make sense to gitignore.

https://xkcd.com/600/

Out of curiosity, did it find a source for those specs that wasn’t indexed well elsewhere?

This change doesn’t even have anything to do with customers as far as I can tell? Sounds like they pissed off their resellers by cutting them out of deals with their biggest accounts, and are regretting burning all those bridges.

Doesn’t look like it, from their docs:

Non-goals

Patching. Difftastic output is intended for human consumption, and it does not generate patches that you can apply later. Use diff if you need a patch.

It worked great for me years ago, but all the US-based banks I use have since killed off their OFX Direct Connect programs.

In case you are unaware, make sure to override DNS on any web browsers or other programs that might be skipping OS configured DNS servers to use hard-coded DNS over HTTPS servers.

If you’re running your own DNS resolver you can hint this to some applications in your network via a canary domain

Cool news, I hope it’s able to be reproduced!

I appreciate this approach to science writing. Gives context and conveys excitement about the science being done, without resorting to hyperbole about how this is guaranteed to change the world as we know it. And very well distilled for someone like me with no background in materials science!