Interested in distributed CI? Radicle? CI in general?
I organize a weekly office hour, for questions about Radicle CI and related topics, in https://meet.ffmuc.net/radicleciofficehour (Jitsi, only needs browser, mic and webcam optional but helpful).
This week: Wednesday, 07 UTC (09 CEST, 10 EEST).
Please join if you have questions, are curious, or need help.
How to backup audio shenanigans like skins presets and shit?
This how I do it in macOS:
$ cat audio-dirs.txt
Library/Application Support/Sonic Academy/
Library/Application Support/u-he/
Library/Audio/Presets/u-he/
Library/Audio/Presets/Xfer Records/
Users/jarkko/Documents/FabFilter/
Users/jarkko/Library/Application Support/com.xfer.serum2/
Users/jarkko/Library/Application Support/com.xferrecords.serum/
Users/jarkko/Library/Application Support/FabFilter/
Users/jarkko/Library/Application Support/Sonic Academy/
Users/jarkko/Library/Application Support/ToguAudioLine/
Users/jarkko/Library/Application Support/u-he/
Users/jarkko/Library/Audio/Presets/u-he/
Users/jarkko/Library/Preferences/com.fabfilter.*.plist
Users/jarkko/Library/Preferences/Serum2Prefs.json
Users/jarkko/Library/Preferences/SerumPrefs.json
$ tar -cvjpf audio-dirs-$(date +%F).tar.bz2 -T audio-dirs.txt
Controversial take but as Flatpak completely bans AI assisted code, it will put people in inequal position:
I’m not claiming that defining AI metrics would be easy define but objectively this at least won’t play out well.
There’s an AI startup in my hometown of Tampere called GitHits. I see their product’s value and a glimpse of actually “doing better for the ecosystem” overall. I got in touch with them because I had one very specific use case in mind.
While at Intel, it was underlined that even copy-pasting code from Stack Overflow is something you should not be doing, given the unknown source of origin. It got really delivered for me through annual trainings, and as a consequence have never copy-pasted anything from there :-)
Where I see GitHits as useful is on the legal/licensing side of software IP i.e., tracking and documenting known sources and providing better legal and licensing feedback.
For instance, even a “permissive” license might require attribution despite allowing unrestricted use otherwise. Having these details screened is crucial software when it has actual IP value.
I believe doing the legally right thing with IP almost unconditionally also improves the legitimate use of open source.
"K&R C" is just LSD with all of the letters decremented by one. makes u think