Replying to
@hywan@floss.social still "in discussion" though, together with the opposite proposal below
Remote
Simon Sapin
@simon@tutut.delire.party
mastodon 4.7.3🦀🦀🦀 abyss gazer by day
he (or they)
339 Followers
304 Following
31 Posts
Joined November 13, 2022
Blog:
Replying to
@sarahjamielewis@mastodon.social unfortunately i think "takes a village" needs to be read as "takes serious funding for a number of full-time people", more than servo currently has. especially over time as web sites adopt new platform features
Replying to
@david_chisnall@infosec.exchange @cstross@wandering.shop i took a eurostar direct from london to lyon a few times years ago. unfortunately i think it was an experiment they didn’t continue
Replying to
@0xabad1dea@infosec.exchange @tomoyo@tech.lgbt @afx@infosec.exchange thanks for the context, i’d only skimmed the original post and assumed this dude was involved
Replying to
@yosh@toot.yosh.is i don't like DRM and overreaching copyright, but i especially don't like how it's selectively applied
Replying to
@sarahjamielewis@mastodon.social two things can be true. there’s a lot of fluff that i’d prefer be removed, and there’s an incredible (and ever growing) amount of complexity in the engine to make the web work and make it fast
Replying to
@niconiconi@mk.absturztau.be my next rfc will mandate that implementations don’t have any bug
The Edge of Safe Rust
https://kyju.org/blog/tokioconf-2026/
excellent blog post by kyren / Catherine West showing common ways to deal with circular references with rust ownership and borrowing, and evolving "Vec + indexing" step by step to fix its downsides until reaching pretty much the design of the gc-arena crate
a garbage collector can be a fancy data structure, not necessarily a pervasive language characteristic!
Replying to
@MicroSFF i love the implications. is the fish a god? the only entity aware than going back happened at all?
Replying to
@kiplet@social.tchncs.de @mhoye@cosocial.ca wait, they actually picked a name that had drifted ages ago to mean "fabulation" and "lie"?
Replying to
@zyzzyxdonta@social.tchncs.de it’s blocking because you need to constantly and actively fight against them, like ads
Replying to
@whitequark @wffl for context: https://docs.rs/objc/latest/objc/macro.msg_send.html relies on type inference for the arguments and return type of an FFI call. with insufficient annotations, the return type would correctly fall back to () but that stabilization attempt changed it to ! somehow, even though the call *did* return. lots of arguably-wrong but innocent-looking calls became UB in nightly
Replying to
@ewjoachim@piaille.fr ok i looked up PEP 636
Matching against constants […] will work with any dotted name (like
math.pi). However an unqualified name (i.e. a bare name with no dots) will be always interpreted as a capture pattern, so avoid that ambiguity by always using qualified constants in patterns
that’s an unfortunate design choice
Replying to
@bodil @ryanprior since i never knew another meaning i must conclude that /usr/bin/bc is also named after you
Replying to
@svgeesus Fedora 43 offers to install 4.0.1 after "command not found"
on a Debian 13 server, the package available is 3.0.3
Replying to
@wffl @whitequark sometimes i wonder if i shouldn't have argued so loud against that previous attempt because of what it did to the objc crate
it was so long ago :(
Replying to
@0xabad1dea@infosec.exchange @SecureOwl@infosec.exchange @catsalad@infosec.exchange i once got email asking if it’d stopped using the machine (which would affect reimbursement). turned out the SIM card had fallen off the network for a couple weeks so the machine wasn’t sending data. the actual cpap worked at usual so i hadn’t noticed. the fix was to turn it off and on again
Replying to
@shravanrn@infosec.exchange @abhishekcs@infosec.exchange @anand@discuss.systems @taehyun@infosec.exchange @jandem@mastodon.social @mgaudet@discuss.systems @hovav@infosec.exchange
nice work!
i skimmed the paper and did not find discussion of: what benefit does this new isolation boundary bring over the existing boundary isolating "content processes" from the rest of firefox?
Replying to
@ewjoachim@piaille.fr it's been a long time since i've done much python but that output looks unexpected. can you say more about what's going on?