CKAJun 3
tobru 🇨🇭 🧑‍🚒

rsync and outrage. Andrew Tridgell | Jun, 2026 | Medium

"Anything you learned about this stuff last year might as well be from another planet. Oh, btw, I also did a PhD in CS, and yes, I actually did quite a lot of work on neural networks, and yes, all my knowledge from that time in my life is utterly out of date too."

Link: https://medium.com/@tridge60/rsync-and-outrage-d9849599e5a0

#linkdump #ai #blogpost #rsync

rsync and outrage

I gave up blogging a long time ago (apart from an occasional thing about ArduPilot), I tend to just write code and hope people find it…

Medium
Jun 3 at 9:11amWeb
Show threadDaniel BairdJun 3
@tobru @peturdainn that was a good read, I’m glad he posted all that.
Show threadCybarbieJun 3
@djb @tobru @peturdainn I've never liked rsync. I found it counter-intuitive, non-unixy and had to read the man page every time. I finally got pissed off with it destroying my data (probably mea culpa) and wrote my own 'copy' which is when you find out how non-trivial it really is. It's a bit sad and annoying that regardless of what people think of rsync past or present that he feels harangued and browbeaten. !Manners
Show threadDaniel BairdJun 3

@nf3xn @tobru @peturdainn

I agree there’s some weird non intuitive behaviour that rsync has. I use it, but it don’t love it.

we’re living in interesting times. There’s bad shit going down that people are justified reacting to. It’s getting harder and harder to not just REACT. Staying calm and figuring out what’s really going on is taking more and more effort.

Show threadAlessandro Corazza 🇨🇦Jun 3

@nf3xn

That's fair, but at the same time, if he'd rather be sailing that much, why not hand off the project to someone else?

@djb @tobru @peturdainn

Show threadpetur 😶🇺🇦🇵🇸🇹🇼Jun 3
@alessandro he did, but stepped in when the maintainer went away
@nf3xn @djb @tobru
Show threadmhoyeJun 3

@nf3xn @djb @tobru @peturdainn

I respectfully disagree. Once people depend on a piece of software - not "use" but "depend" - then we have different obligations around testing and deployment. "We've landed a bunch of changes, for security reasons, to code that is currently poorly tested,, that may break some use cases" should be well-preannounced, with much greater transparency and a longer pre-release shakeout period.

There's nothing like that visible in recent rsync changelogs.

Show threadmhoyeJun 3

@nf3xn @djb @tobru @peturdainn

Put differently, we live in a society and at some point "this software is provided as is with no warranty express or implied" is no more morally defensible a position than "this building is provided as is and if it falls down well you chose to live in it, whether or not I decide to change it however I like in the future."

... and I'm not sure where that line is, exactly, but rsync is on the far side of it.

Show threadCybarbieJun 3
@mhoye @djb @tobru @peturdainn I disagree in that if I release free software and you happen to use it that is on you. You have the option of not using it. I know the eurocrats have different opinions especially for the cybers but I respectfully disagree and under circumstances where I be legally required to provide some guarantee of fit for purpose or merchantability will simply not release. If people don't like the thing he made they are free to fork off and I believe some have.
Show threadmhoyeJun 3

@nf3xn @djb @tobru @peturdainn

I don't think it's a legal question; I think it's about transparency, risk management and community harm reduction. This was a breaking change that landed as a couple of bullet points in what was numbered as an inconsequential upgrade, 3.4.2 -> 3.4.3.

If this had been shipped as "3.4.3 -> 3.5.0-alpha1, significant upgrade please test", we wouldn't be having this conversation.

Show threadCybarbieJun 3
@mhoye @djb @tobru @peturdainn I still contend there is no onus or moral duty, requirement for transparency or to satisfy some abstract community unless you are actually paying the guy. I don't know the exact details as I personally don't use rsync at all for aforementioned reason-it is shit-but my understanding was that he had generated a TON of new code with AI. No idea it was borked but y'know again its his business. I defend people's right to freezepeach bad code without obligation. fwiw...
Show threadCybarbieJun 3
@mhoye @djb @tobru @peturdainn I don't think there is anything wrong with saying to him - dude that kind of sucks or you made me sad now or just hey we could have done with a heads up but I think it went a bit further than that into ad hominem attacks and accusations and other unpleasantness.
Show threadDaniel BairdJun 3

@mhoye @nf3xn @tobru @peturdainn a wicked problem this. I’m reminded of @daedalus

https://2025.pycon.org.au/program/A7XZLS/

Sustaining Open Source Software: Tools for Change—PyCon AU 2025

Show threadengtaoJun 4
@tobru @pjacock Oh no!