Mastodawn

Today marks the 18th anniversary of the first commit of what should eventually become Spring Data JPA. What a ride! 🥳🍃🛢️ https://github.com/synyx/hades/commit/fca7af83e57188097310e50a30670360853b57ff #SpringData #JPA #Synyx

* First implementation of a spring based generic JPA dao as well as s… · synyx/hades@fca7af8

…implified spring configuration * Implementation is based on code published by Eberhard Wolff in JavaMagazin git-svn-id: svn+ssh://svn.synyx.de/var/svn/synyx/spring-jpa-dao/trunk@186 5a64d73e-33d...

GitHub

Mark Paluch 👨‍💻,🎹&🚑Dec 28

Marc Philipp Dec 28

✨ New blog post: "STF Milestone 7: Safe cancellation"

Introducing a safe way to cancel #JUnit test execution early, e.g. after the first test failed, but still execute all cleanup logic.

👉 https://marcphilipp.de/blog/2025/12/28/stf-milestone-7-safe-cancellation/

Thanks to the @sovtechfund for their support!

STF Milestone 7: Safe cancellation • Marc Philipp

Personal website

Show thread

Mark Paluch 👨‍💻,🎹&🚑Dec 8

@javahippie H…auch so eine großartige deutsche Erfindung. Ach nee, es war ja nur Sauklaue und deswegen ist das B auch wo anders 🤷

Mark Paluch 👨‍💻,🎹&🚑Nov 22

🇺🇦 Anna Filina Nov 22

Looks like PRs (and other contributions) written by LLMs are causing damage everywhere. I invite OSS projects to prohibit LLMs as a policy. Just treat everything that comes out of LLMs as invalid and be done with it.

Giving people the benefit of the doubt and then seeing such entitled and bad faith behavior is exhausting. I'm sure maintainers have better things to do than politely sift though towering piles of garbage.

https://chaos.social/@dpk/115589097803252590

Daphne Preston-Kendal (@[email protected])

Fucking. Hell. https://github.com/ocaml/ocaml/pull/14369

chaos.social

Mark Paluch 👨‍💻,🎹&🚑Nov 22

Damals… @rotnroll666

Mark Paluch 👨‍💻,🎹&🚑Sep 19, 2025

@rjayasinghe Moin! Das geht hier zu jeder Zeit.

Show thread

Mark Paluch 👨‍💻,🎹&🚑Sep 4, 2025

@amine_ If you want to contribute to OSS, the first two paragraphs would be a neat description for a pull request. Anything else would go into code and you’d had a much more pleasant contribution experience.

Mark Paluch 👨‍💻,🎹&🚑Sep 4, 2025

@amine_ Welcome as first-time-contributor.

My points mentioned are still valid, AI generates weird content. We consider this ticket a bug, yet the form and way how AI renders your research input causes hesitation to put a lot of effort initially. Several tickets with such a shape have turned out to be hallucinations keeping us maintainers busy.

Show thread

Mark Paluch 👨‍💻,🎹&🚑Sep 4, 2025

@amine_ another aspect to consider is the additional load to consume such reports. While I genuinely believe you wanted to contribute useful and helpful things to the project, LLMs are and will remain content generators that reward themselves by generating a lot of content. Using AI and editing its output to augment yourself is a good practice. Posting unedited AI generated content results (happens often) on our side in a DDoS on our time and motivation. Likely unintentional effect of AI users.

Mark Paluch 👨‍💻,🎹&🚑Sep 4, 2025

@amine_ do you have a link handy? We’ve been receiving several tickets and pull requests having the same characteristics in terms of overly long descriptions, bullet-point-driven summaries, subtitles with each word capitalized and content that isn’t relevant to the project, the issue, nor our code at all.

Happy to review the ticket and apologies in advance if I made a mistake.