pleroma.debian.social

pleroma.debian.social

Jonathan Dowland | @jmtd@pleroma.debian.social

Principal Software Engineer on #OpenJDK #RedHat. #Debian developer (dormant). Computer Science PhD student. Amateur Computing historian (Computer Science and H/W, esp. Commodore Amiga). Guerilla archivist.

@m @mwl yeah. My long term experience is Perl and ruby are fine; python not so much

@neil cheers!

Viva today!

@cameron_bosch @neal the bizarre, idiosyncratic and rich variety of WMs (and other stuff) was what hooked me on Linux, back in the 90s

@m @mwl counter-anecdote: I picked up maintaining a perl codebase last year, after at least 10 years off, and it was refreshingly still-working

I haven’t noticed whatever this big outage is and I feel that means I must be doing *something* right.

@algernon once upon a time one could tune Qt apps to resemble gtk ones (or vice versa). Is that a possibility?

@zhenech that was the one. Thanks!

I was enjoying a blog post on the Fediverse about Canonical’s recruitment process but I only half read it and my mastodon client has lost it. Does anyone know which one it might have been? Thanks!

@lproven @bloor @neil that’s likely who I mean, yes!

@lproven @bloor @neil the Prague-Brno train (the one run by students) remains my best train experience *ever*. Leaves first class uk trains in the dust

Rapidly trying to read up on School refusal :(

@brittcoxon if it’s a wire break (e.g. in the band) it’s likely possible. Other kinds of failure lower chance unfortunately

I've spent *hours* poking at podman behaviour when remapping UIDs (--userns=auto) and weirdness around file descriptors. Good news is it was fixed sometime between 4.9.3 and 5.4.2

Since I'm poking at httpds-inside-containers, I re-discovered Apache HTTPD's abuse of SIGWINCH. Resize the terminal running the apache container, and it dies: https://bz.apache.org/bugzilla/show_bug.cgi?id=50669

@mhoye the relevant tool is feh, and my script appears to be feh --no-fehbg --bg-fill --randomize /path/to/dir/of/wallpapers

@mart_brooks Yes I lean towards any logging going to stdout or stderr. If a user wants to distinguish error log from access log they should need to configure accordingly, provide a volume mount, etc.

Should a containerised httpd log to standard error/out (and let the container system handle logs)? For errors? For accesses?

@mdione @arrjay @mhoye I do that! But I tend to edit down copied-out history anyway (remove my hostname etc), so I might as well not be doing it

»