Dear network switch vendors: Why do you always put the power connector on one side of the box and the network ports on the other? I somewhat could understand for a rackmount switch but for desktop ones it's always Always ALWAYS better to have all the cables on one side of a component.
Remote
Billy O'Neal
@malwareminigun@infosec.exchange
Dev at Microsoft on the vcpkg team. Former @VisualC STL maintainer. He/Him (Although I don’t care much)
675 Followers
327 Following
50 Posts
Joined December 18, 2022
GitHub:
Discord:
BillyONeal#4301
Open post
. @lmanul@macaw.social absolutely rocking https://lmanul.github.io/null-pointer/public_img/0011.png
2
2
0
0
Open post
Replying to
@lmanul@macaw.social You're very welcome, and thank you for making epic shit.
I have to make up for the crazy number of times I see your work show up uncredited.
1
0
0
0
Open post
Replying to
@vitaut@mastodon.social CMake has very good "tree" documentation and awful "forest" documentation.
2
0
0
0
Open post
Replying to
@vitaut@mastodon.social In fairness in my experience even people who eat CMake for breakfast every morning, like *me*, should know better about some things and don't.
1
2
0
0
Open post
It's almost like it has nothing to do with security and everything to do with kissing the ring
4
0
0
0
Open post
Replying to
@RomanOnARiver@mastodon.social makes sense and that's fair. Sorry for my confusion
1
1
0
0
Open post
Open post
Open post
Replying to
@CarePackage17@mastodon.gamedev.place Oh wait I just realized this is for everything including dependencies and I was comparing with just LLVM itself ... so this might be the 20ish percent faster I was expecting
1
3
0
0
Open post
Open post
Open post
Open post
Replying to
@pinskia@hachyderm.io I still think it should pass the \t rather than needing an option
0
1
0
0
Open post
My stepfather, Woodro “Woody” Riley, December 22 1970 - August 26 2026
0
0
0
0
Open post
Open post
Replying to
0
0
0
0
Open post
I keep seeing the Codeberg logo and thinking of the Myst opening https://youtu.be/8zSHBjab1t4
0
0
0
0
Open post
Replying to
@bpfitzgerald@infosec.exchange I mean, I don’t have an issue using it because I did. This isn’t about whether the policy is reasonable. It’s their project and they can set that policy if they want. This is about what happens after I realize I’ve fucked up.
I’m also asking about this because I strongly suspect this won’t be the last time. Since we started hanging those tools check vcpkg packages they have caught a LOT of bugs, including this one. But I do strongly care about contributors to other components not hating me/us
0
1
0
0
Open post
Replying to
. @masnick.com Oh, I disagree. Everyone who would have ever considered this a good idea, please keep doing it so that I know who I can avoid doing business with in the future
0
0
0
0
Open post
Replying to
@CarePackage17@mastodon.gamedev.place regular. I've never been able to measure a difference for the whole dev drive thing. Maybe makes a bigger difference for many small ports rather than 1 big one
0
0
0
0
Open post
Tired: Ridiculous
Wired: Ridiculounumbus
0
0
0
0
Open post
People who hate AI contributions. I accidentally had AI generate a fix for a project that explicitly forbids AI contributions. I wrote about it as https://github.com/icculus/physfs/issues/125 . Notwithstanding that I know you hate that I used AI for anything, was this an acceptable way to handle the situation, or do you think I should have done something different?
0
1
1
0
Open post
Replying to
@vitaut@mastodon.social I argue the implementations that try to preserve those as long double are the real whacky actors here. The "we're going to give the hardware a few more bits to make implementing transcendentals easier" seems like a reasonable idea
0
0
0
0
Open post
I love listening to Seriois Trouble and hearing Ken White say "was reported on by @molly0xfff@hachyderm.io and @krebs_on_security_feed@noc.social "
0
0
0
0
Open post
Replying to
Wow there's one that's even worse
0
0
0
0
Open post
Replying to
@GossiTheDog@cyberplace.social I always assumed killing Claude Code use is just because it's embarrassing to be using a competitor rather than it having anything to do with cost
0
0
0
0
Open post
Replying to
@pinskia@hachyderm.io Everyone has a different opinion about how many spaces to draw for a tab. Which is why tooling should just pass through the \t and let the user's configuration do what they want with it.
FWIW I am of the 4 spaces per tab religion rather than 8 but I fully understand the religion that it is
0
1
0
0
Open post
Replying to
@vitaut@mastodon.social (If 1 implementation does something dumb then blame that implementation, when every implementation does something dumb I think you are just https://xkcd.com/2501/ )
0
1
0
0
Open post
Replying to
@chandlerc@hachyderm.io I am also normally a spaces person but not to the point I would ignore what the user said
0
4
0
0
Open post
Replying to
It’s so frustrating to see the vendor who did music the most right just completely torpedo it slowly over time
0
0
0
0
Open post
Open post
Replying to
@CarePackage17@mastodon.gamedev.place This is on the Strix Halo machine again... so probably trades blows with 275HX. I was really expecting yours to be the ~20% faster I see in reviews but compilation and Cinebench have never been the best comparisons
0
2
0
0
Open post
Replying to
@RomanOnARiver@mastodon.social I would argue "vibe code" traditionally means not only using AI to write the code but also not reading it or having any specific understanding of the details of its function, so accusing someone of "vibe coding" additionally implies that they didn't even review it.
0
0
0
0
Open post
Replying to on mastodon.social
@RomanOnARiver@mastodon.social is it that you don't see the distinction, or is it that you don't believe it to be meaningful?
If truly the former, pretend it's OSS with no AI at all where one maintainer actually reviews and spots bugs in PRs vs. one who just rubber stamps and merges anything.
(To be clear, the only reason I replied here is because your toot was in the form of a question. FWIW I would describe my day job as mostly vibe coding these days; before AI most vcpkg PRs were a rubber stamp)
0
2
0
0
Open post
Replying to
@vitaut@mastodon.social "naive" ... "10x faster" ... I think there is some confusion here
0
3
0
0
Open post
Replying to
@vitaut@mastodon.social There's a lot of stuff in textbooks that is not "naive".
0
1
0
0
Open post
Replying to
@chandlerc@hachyderm.io (I'm looking at you GCC and your recent converting tabs to spaces in console output assuming that everyone follows GNU indentation conventions rather than just emitting the \t and letting the user's terminal/editor do what they configure it to do)
0
1
0
0
Open post
Man claims without evidence to have reinvented (at least) 45-year-old technology (https://en.wikipedia.org/wiki/STS-1) to much fanfare. For some reason.
Worked out great for Colombia. I'm pretty sure the Inconel they made the orbiters out of is a bit higher quality than the stainless steel sheet metal they're using. 🤷♂️
0
0
0
0
Open post
Replying to
@Migueldeicaza@mastodon.social I can't decide if I hate calling the orange pile of cheesy whatsits "emperor" is good for criticizing how undemocratic he is or bad because it suggests that the last shred of democracy is gone
0
1
0
0
Open post
Replying to
0
0
0
0
Open post
Open post
Open post
Replying to
@chandlerc@hachyderm.io I mean, sometimes they might not wrap in useful ways but in my experience tools that try to workaround this often do not help the situation: nothing is more confusing than double wrapping
0
7
0
0
Open post
Replying to
@bpfitzgerald@infosec.exchange I feel like you're kinda missing the point of my question, namely, that I'm asking about my behavior, not the maintainers'.
0
1
0
0
