Is there some clever way to make GNU Parallel (when running a list of bash commands in a file) remove the lines from the file as they complete with exit status 0?

Alternatively, some way to print *un*-completed commands to stdout when parallel gets a SIGTERM?

In either case, I'm trying to find a clean way to resume a set of non-idempotent operations that I have to temporarily ^C for some reason.

apetresc boosted

I'm planning on streaming some #Readlebee development on Thursday at around 7 or 8 pm Mountain Time, so stay tuned if you're interested! As it gets closer, I'll post the exact time and the link!

#liveCode #webdev

apetresc boosted

@apetresc @codesections no, he didn't change his mind. His website was defaced, presumably by volunteers. In the section "What's bad about" there was a Richard Stallman link and also a link to the first, defamatory, medium post that started the recent storm

@hntooter Love the feed, but is it really necessary to CW every single post? I don't see the benefit at all :(

apetresc boosted
if you want a synopses of who is on the fediverse; its people who would join internet forums for specific topics all jammed together into a not-actually-single social network. If we forget about federation, the functional difference is just how content is displayed and share.

But federation actually exists, so you got clusters of people with particular interests all in approximately the same space. We're federated GameFAQs.

The best part of Toy Story 4 is that it pushes out the date when Debian releases run out of new names by another ~60 years.

Completely agreed with this article:

It's surprising to me that none of the major application sandboxes generate an "audit log" that shows exactly which protected operations (full filesystem access, camera, location, etc) were used, and exactly when. Not just timestamps but also parameters, open windows, etc.

You could even imagine third-party services aggregating this log across users and time to detect anomalies and deviations.

It's exciting to see major IRC networks actually implementing new IRCv3 features!

Loving @mastonaut so far, but... is it really necessary to have this prompt every time the app is started up, in the case where the user only has one account?

@jamie @brandon I think in the ideal system, the criteria for federation doesn't stop at just "is the other instance literally criminal in their lack of moderation". It should aim at maintaining a cohesion of theme, values, interests, etc.

Just like how everyone back in the phpBB days quickly realized that it wasn't a good idea for *every* forum to have boards for *every* topic. Specialize, and who cares if your post count would've been higher if all your forums were combined instead.

@jamie @brandon I think it's so ironic how everyone complains bitterly when platforms like Google and Facebook try to unify identity everywhere, for all the right reasons – and then willingly commit to the exact same problem on the Fediverse.

@jamie @brandon I don't even think it's necessary to abstain from federating – just federate *selectively* and within the constraints of the instance's focus.

Needing to have multiple accounts across multiple instances shouldn't be regarded as a downside the way it is now. We should change the model so that identity is fragmented by design - high follower counts be damned.

Is the @funkwhale GitLab instance down right now, or has it moved somewhere else? (cc @eliotberriot)

If you read only one kernel patch note this year, make it this one:

tl;dr: 5.1.2 has a single mitigation= param to disable *all* of them, from retpolines to and everything in between 🙏

Has anyone tried out (and ideally even benchmarked) the new Intel microcode 3.20190514 update, that mitigates et al? (Details at

Is it as drastic as Apple's mitigation that disables SMT? That's way more of a performance penalty than I'm willing to pay for a purely theoretical vulnerability.

apetresc boosted

@ParadeGrotesque you can't make Microsoft go away. They're the most valuable company in the world. So your choices are to encourage them and appreciate their open source efforts, or to continue to disparage them. Which of these options encourages the engineers at Microsoft to keep pushing for open source? Reard them for the good they've done so they keep doing it. They can't un-open the things they've open sourced. They've made permanent and meaningful contributions to open source and I'm thankful to them for it.

Oh nice, the thumbnail does show up on the official web client at least 😀

Testing whether embeds work properly on which Mastodon clients...

Show more

Fosstodon is an English speaking Mastodon instance that is open to anyone who is interested in technology; particularly free & open source software.