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.

Too tired to be useful tonight

@cstross @mrintergalactickeyboard @lauren @vvandinsky yes the 10” was much more practical. (You remind me about the tail end of that subsequent ultrabook era, the Samsung “ultrabook” fridge). I saw a GPT pocket 3 last FOSDEM and it was highly desirable!

@cstross @mrintergalactickeyboard @lauren @vvandinsky nightmares of that cramped keyboard

@mu yeah for papers I think printing and scribbling are still best for me too. Remarkable2 too small to comfortably do that for traditional journal layouts. Where it shined for me was reviewing my own thesis chapters, over and over

The Remarkable2 was incredibly useful whilst I was doing my PhD. Now I don’t really need it. So of course I’m looking at the remarkable pro…

tfw researching a topic you turn up a question you asked and have forgotten about for 15 years… https://exim-users.exim.narkive.com/BSnc11jb/smtp-time-spam-filtering-with-crm114-best-approach

@aeva I’d love your opinion on JACK. Signed: embarrassed, who has yet to get his synth to be audible from his Linux machine

…and I did it again, time to disable!

Sigh, Prusa Mini failed to print a simple trapezoid. I'm having a lot of failures since input shaping was added. #3dprinting

@StefanEJones who’s the poor guy in the lion suit?

@mark @mcc my supervisor has a habit of writing out “alpha”, “beta” etc for type variables. I struggled a lot with that

Just cut a new IkiWiki release. Hopefully I didn’t break anything. This is intended for Debian Trixie

@pndc it’s the bounce that concerns me. I could live with spooled retries. From the docs I think it might depend on the return code

@pndc i do! But as I understand it’s “pipe” command, I’m concerned about the behaviour if the command fails

@mcc I’m out of touch on current fashions, but my last project used stack and it solves a couple of the issues you’ve hit so far

@mcc @dysfun there’s quite a good chapter on parsec (megaparsec precursor) in “real world Haskell” (which is free online). I concur that a parsec variant is the best choice for this, even for micro-scale parsing . And I found that counter intuitive first time

@cstross mine’s always faintly lemon flavoured. I’m not sure I’d like mango more or less. I know I have to avoid lemon squash for months afterwards

@joostvb thanks! Out of curiosity, if you use “system some-command”, and some-command exits signalling failure, what does maildrop do?

@suihkulokki that’s the one I need to look at next. Thanks!

@noahm @suihkulokki I need to look at dovecot’s sieve in particular. I do want a pipe call-out (to a spam classifier) which exim’s sieve implementation doesn’t support (and exim’s own filter language has poor behaviour if the pipe fails: no handling in the filter script; generate a bounce to sender!). Thanks!

»