Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
Twirrim
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
78 ms
·
631.
▲
by
Twirrim
3y ago
Should be flagged as 2020. Rust is currently on v1.70!
632.
▲
by
Twirrim
3y ago
I've found myself in the past in a situation where I needed an offline regex debugger (dealing with air-gapped networks, so the folks that needed to use the tool literally couldn't get at any online site, nor would they be permitt
633.
▲
by
Twirrim
3y ago
It's not about you. It's about the person/company that signs the CLA. They're the ones that end up in potential legal nightmares, unless the CLA specifically promises to continue to indemnity.
634.
▲
by
Twirrim
3y ago
Lawyers at my job pointed out one particularly nasty aspect of CLAs: the lack of liability controls. All open source licenses have some form of legal liability wording in them, e.g. from GPL v2: > 12. IN NO EVENT UNLESS REQUIRED BY APPLI
635.
▲
by
Twirrim
3y ago
After the first couple of years, Debian Security stops patching it. "Debian LTS is not handled by the Debian Security team, but by a separate group of volunteers and companies interested in making it a success.",from https:/
636.
▲
by
Twirrim
3y ago
> Debian LTS only has 5 years of support, without guarantees, on a volunteer basis. Less than that if you care about security patches. Debian's Security team stop after ~18-24 months, at which point it transitions off to, quote, &q
637.
▲
by
Twirrim
3y ago
The "problem" with Debian is the lifecycle. You get about a 12-24 months at most of security patches, and that's effectively it (the patching model after that is unreliable). That means you've got to allocate resources
638.
▲
by
Twirrim
3y ago
> Please, dear Red Hat, just give me a DigitalOcean VM (or something comparable) that costs a few dollars more and has automatic RHEL licensing. Make deals with the VM providers, work something out. Most of the major clouds have this p
639.
▲
by
Twirrim
3y ago
I mentioned this in another thread here before, but I've had 3 Pixel 5a devices completely die on me. Dealing with Google support was such a pain I gave up. My first one died nicely within warranty, and they sent me to a "local&
640.
▲
by
Twirrim
3y ago
The way that government procurement often works, it isn't the cheap contractors you end up with, either. Not many companies have the resources or willingness to stick through the long time it often takes to go through the bidding proc
641.
▲
by
Twirrim
3y ago
NUMA can be such a pain, especially because it's really difficult to account for in code, and all sorts of stuff you'd never think of can end up impacting your performance too, that usually you'd never even think of, e.g. the
642.
▲
by
Twirrim
3y ago
To the contrary, the YouTube algorithm strongly discourages it. Longer videos are much less likely to appear in the suggested feed. The degree to which varies, but YouTubers are feeling very heavily forced towards < 5 minute content.
643.
▲
by
Twirrim
3y ago
Why would you expect US laws to apply to China?
644.
▲
by
Twirrim
3y ago
> Well, I, for one, don't fucking care about third party apps (which make money mind you) Those third party apps are central to how the entire content works, they're absolutely critical for the moderators that try valiantly to
645.
▲
by
Twirrim
3y ago
That one didn't. The bug wasn't actually in the kernel, it was in the klibc / klibc-utils. https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=852480 was the ultimate bug/patch that came out of it, C
646.
▲
by
Twirrim
3y ago
Oh that's really interesting. Thanks for sharing! I might revisit this whole post & bug and try using User-mode, see if I can reproduce it that way.
647.
▲
by
Twirrim
3y ago
Original post author here. Not sure what sparked this to be posted now. Happy to answer questions if folks have them.
648.
▲
by
Twirrim
3y ago
I've been having flashbacks to troubleshooting some particularly thorny unreliable boot stuff several years ago. In the end tracked that one down to the fact that device order was changing somewhat randomly between commits (determinis
649.
▲
by
Twirrim
3y ago
> - I get an analysis that I like, but there isn't a good way to share it with others, so I end up just taking screenshots. Jupyter supports printing out as a PDF, an all-in-one-webpage and all sorts of other formats. I've use
650.
▲
by
Twirrim
3y ago
There's whitepapers and articles every few years on Holographic storage, it's one of those perpetually just around the corner things. See also memristors, that HP in particular have been saying is coming in the next couple of year
651.
▲
by
Twirrim
3y ago
Argh. Apologies to the author. I didn't see the original when I looked before submitting, and I expected HN to pop up a link to the existing post, like it usually does. Wasn't my intention to post a dupe :(
652.
▲
A performance analysis of Intel x86-SIMD-sort (AVX-512)
(github.com)
129 points
by
Twirrim
3y ago
|
47 comments
653.
▲
by
Twirrim
3y ago
A one paragraph news article with effectively zero details? Keeping that quality bar nice and high at techcrunch.
654.
▲
by
Twirrim
3y ago
> Ah yes "there is always x, so why shy from this x?", I wonder if there's a named logical fallacy for this, I think the closest might be "Moral Equivalence" I absolutely did not say that, and would not say that
655.
▲
by
Twirrim
3y ago
So this is another example of HN commenters getting up on their high horse before gathering facts? Say it isn't so!
656.
▲
by
Twirrim
3y ago
If it involves people, it will be political. There's absolutely zero way to avoid it short of it being a single-person project (and arguably even _that_ will be Political at the very least).
657.
▲
by
Twirrim
3y ago
Decades ago they relentlessly optimised that page, shaving it down smaller and smaller. Then suddenly they stopped caring, round about the time they gave people customised landing pages with extra junk littering the page. It must have mad
658.
▲
by
Twirrim
3y ago
Thanks for sharing. Lots to chew over. This bit stood out to me as the main thing that's probably at least rationally explainable: > Why did RustConf leadership go along with this decision and not protect the speaker? Why wasn'
659.
▲
by
Twirrim
3y ago
I think OP was suggesting that instead of overflowing, the API could have instead returned 2^32 / 4 GiB.
660.
▲
by
Twirrim
3y ago
Sadly, 1.0 is _still_ around. Rare, but it happens. I remember being gobsmacked several years ago that F5 LBs only supported HTTP/1.0 health checks. Doing HTTP/1.1 health checks required writing one specifically for it. Somethi
More ›