snvzz

38 ms·

631.
by snvzz 1y ago
Erlang goodness. Say what you will about XMPP, but it Just Works, supported by robust software like this.
632.
by snvzz 1y ago
They lost me with the FreeBSD to Linux switch.
633.
by snvzz 1y ago
Preferred video8 and its successors, they were so much better. We had VHS and used it just to ingest tapes into Video8, keeping our collection purely video8.
634.
by snvzz 1y ago
Let's hope an example is made out of them, as well as obtaining the associated legal precedent. Furthermore, there's no reason we can't e.g. have a list of subscriptions and a button to cancel them, in every bank account&#x27
635.
by snvzz 1y ago
No, we won't w/o breaking changes. There's no way. Even ignoring pressing issues like lack of mandatory E2EE, SMTP requires encoding binary data into text. This includes the main body for most emails these days. Awfully waste
636.
by snvzz 1y ago
When it comes to SMTP for email, time has only served to highlight its inadequacy. DMAC, DKIM, SPF, S/MIME, PGP are all ugly workarounds. The issues are fundamental.
637.
by snvzz 1y ago
There's DIME[0]. I wish I heard about the effort more often, as it's sorely needed. 0. https://en.wikipedia.org/wiki/Dark_Mail_Alliance
638.
by snvzz 1y ago
+++ATH0
639.
by snvzz 1y ago
RISC-V is inevitable. It is sensible to focus all effort there.
640.
by snvzz 1y ago
TL;DR: RVV does much better than AVX2, in terms of how many loops can actually be vectorized, as well as the size of the resulting code. And, remarkably, some categories of loops cannot be vectorized in RVV due to a GCC limitation which the
641.
by snvzz 1y ago
The HURD has been around for a while, and its architecture is archaic. It was designed in a world where Mach is the microkernel, and thus microkernels are slow. There's many more options[0] these days. I'd like to highlight Genode
642.
by snvzz 1y ago
No such discussion is complete without mention of SENS[0] i.e. the rejuvenation approach. Not about delaying the inevitable, but about undoing the accumulated damage of aging. 0. https://en.wikipedia.org/wiki/SENS_Resea
643.
by snvzz 1y ago
For anything written from scratch, I would recommend fasm or nasm. I prefer the latter, because the documentation is better and there's a way to specify target cpu (e.g. 8086) and get errors when instructions aren't compliant.
644.
by snvzz 1y ago
Because she's cute; everybody loves her. It filters out people with issues.
645.
by snvzz 1y ago
Good. This nonsense, like religious scriptures, is also important history to be preserved.
646.
by snvzz 1y ago
clean mirror: https://archive.is/wMECM
647.
by snvzz 1y ago
It requires an extra click to get to the actual website. PuTTY's website is fairly clean and accessible, unlike this landing page.
648.
by snvzz 1y ago
The research (multiple papers; note they have published and presented more than I linked) they've done prove that hardware help is necessary. i.e. not about reducing cost, but about being able to kill timing side channels at all .
649.
by snvzz 1y ago
RVA23 is very important to the RISC-V ecosystem. Unlike RVA20 (which is almost the same as RV64GC), RVA23 is comparable to x86-64v4 and ARMv9. It being the baseline means the CPU will be running RVA23 code practically all the time. This mak
650.
by snvzz 1y ago
As a reminder, it would require solving the legal mess[0]. 0. https://sites.google.com/site/amigadocuments/
651.
by snvzz 1y ago
There's about one way[0] to fix timing side channels. The RISC-V ISA has an effort to standardize a timing fence[1][2], to take care of this once and for all. 0. https://tomchothia.gitlab.io/Papers/EuroSys19.pdf 1
652.
by snvzz 1y ago
the real problem is Linux is written in such manner that APIs change very often, and code all over the place including drivers is monkeyed without testing and often broken, sometimes in very subtle and hard to debug ways. It is an overall d
653.
by snvzz 1y ago
It is an Amiga, thus it does not have to draw everything via CPU like the PC version. Expecting 030 to work is not at all unreasonable, and it's also a very common A1200 configuration these days.
654.
by snvzz 1y ago
It has a distribution, Sculpt[0]. Many stories around using Genode in the Genodians[1] blog. 0. https://genode.org/download/sculpt 1. https://genodians.org/
655.
by snvzz 1y ago
As a reminder, this mess[0] needs to be untangled first. A tall order. 0. https://sites.google.com/site/amigadocuments/
656.
by snvzz 1y ago
>It would be cool to have a Hurd project with a verified microkernel like seL4. There's Genode[0]. Relative to the hurd, its design is much more advanced and it supports a range of modern microkernels including seL4. 0. https:&#x2F
657.
by snvzz 1y ago
The HURD has been around for a while, and its architecture is archaic. It's from a world where Mach is the microkernel, and thus microkernels are slow. There's many more options[0] these days. 0. https://www.microkernel
658.
by snvzz 1y ago
xorshift, a popular solution for pseudorandom also arm and m68k... and probably every architecture.
659.
by snvzz 1y ago
>and will likely be for the foreseeable future. Nah, high performance RISC-V (on RVA23 profile) is just around the corner. Might come as early as by year end. Early next year at worst. We've known about Windows for RISC-V since 202
660.
by snvzz 1y ago
>Apple isn't using ECC either. And thus their machines are expensive toys as well. (I type this on a system I built, a 9800x3d with 96GB (2x 48GB) ECC RAM.) >If your only aim is to run LLMs how much do you actually care about ECC

More ›