Automated Visual Regression Testing With Playwright, by @csstricks:

https://css-tricks.com/automated-visual-regression-testing-with-playwright/

How to Do Visual Regression Testing in Vue With Vitest?, by @alexanderopalic@x.com:

https://alexop.dev/posts/visual-regression-testing-with-vue-and-vitest-browser/

#WTF
#Nextcloud Android App

Severe #regressions

Photo uploads:
EXIF GPS data is stripped. WTF1

Autouploads:
no-media files are not automatically uploaded WTF2

Already filed and discussed in several issues like recently in

https://github.com/nextcloud/android/issues/14199#issuecomment-2558771115

and previously in

https://github.com/nextcloud/android/issues/6248

'Causal Discovery with Generalized Linear Models through Peeling Algorithms', by Minjie Wang, Xiaotong Shen, Wei Pan.

http://jmlr.org/papers/v25/23-1228.html

#causal #alzheimer #regressions

The two #Bluetooth #regressions[1] recently affecting many people in #Linux mainline and various stable/longterm #kernel series should soon be a thing of the past:

Linus merged fixes into #LinuxKernel 6.12-rc4[2] and Greg queued them for the next stable releases[3].

Wondering what we should learn from this wrt to handling such regressions quickly and ideally even preventing them from hitting stable.

[1] https://lore.kernel.org/all/4e1977ca-6166-4891-965e-34a6f319035f@leemhuis.info/

[2] https://git.kernel.org/torvalds/c/d7f513ae7b108f953cceec8bc96d2e5e83c3ccd0

[3] https://lore.kernel.org/all/2024102100-spongy-etching-0695@gregkh/

I love #TDD! So happy i get to do that properly in my own projects. Refactoring without fear of uncaught #regressions is amazing.

Can you come up with convincing *stats* on why backporting a huge pile of patches (say 800+, like in the case of #Linux 6.10.3) to stable #kernel series just days after being mainlined *during the merge window* and thus included in a mainline -rc1 is *worse* than quickly backporting changes mainlined *during the rest of the development cycle*?

Then tell me about it, because then I'll bring this up on the maintainers summit while talking about #regressions.

1/ FWIW, this is what I tried: I…

Seems I got a new nickname:

"regzcop"

Thx Ard Biesheuvel, I like that! https://lore.kernel.org/all/CAMj1kXHLSROujjc_OifzVmuAd7BpA%2Bb4D2JrPGLJJYj2au7SfQ@mail.gmail.com/

Screenshot from https://mastodon.online/@heftig/112592068865597095

#Sigh

Ever wondered why @torvalds coined the #Linux #kernel's "no regressions" rule? He just explained it again here: https://lore.kernel.org/all/CAHk-=wgtb7y-bEh7tPDvDWru7ZKQ8-KMjZ53Tsk37zsPPdwXbA@mail.gmail.com/

'"[…] I introduced that "no regressions" rule something like two decades ago, because people need to be able to update their kernel without fear of something they relied on suddenly stopping to work. […]"'

Follow the link for context and other statements that did not fit into a toot.

I might regret this later, but... What are the pros and cons of five-point vs. seven-point #Likert scales, and what are the risks of changing from whatever a measure was developed for? I've seen opinionated blog posts on this, but is there actual #research on this, or #simulations or the like? We're working on a #survey, data will be used for multivariate #analysis like #regressions, #SEM and the like. On one hand, we'd like to preserve the scale options the different included measures were developed for. On the other, we want to standardize response options across the survey.

one of today's lessons is that if each FOSS package had their own test suite for performance regressions one might catch an odd spike in CPU/latency/memory/network that is a symptom of a backdoor or attack attempt. if its new AND unexpected one should dig in

I'm the author of a FOSS Golang latency instrumentation lib designed especially to be used by perf regression tests:
https://github.com/mkramlich/LatLearn

#cve20243094
#xz
#lzma
#liblzma
#backdoor
#openssh

#Golang
#latency
#performance
#Regressions

Today I take on the AI Ouroboros. Can 2024's AI Build 2014's AI? Spoiler alert: kind of. https://sadlynothavocdinosaur.com/posts/data-science/

Hello crowdbrain!
Is there any good source for information about macOS update regressions?

Really glad that @gregkh refuses[1] to pick up changes for stable/longterm #Linux series (say 5.4.y), if they have not yet been applied to all newer series (e.g. 5.10.y, 5.15.y, …). This is crucial, as people otherwise would run into #regressions if they'd sooner or later update from the older to one of the newer series.

[1] every few weeks (or even more often) some developer learns that aspect through a mail like https://lore.kernel.org/all/2024022603-speculate-staple-4f54@gregkh/