nyrikki

5 ms·

31.
by nyrikki 4mo ago
For simple cases I just launch podman containers on long lived hosts with ansible. You can still add pods if needed and the systemd integration works. Plus you can actually improve isolation by co-hosting services under separate UIDs. Like
32.
by nyrikki 4mo ago
You can get 60tps with three 1080tis and the sparse model, and I bet two 16gb 5060tis would do the same for ~1200. One 3090 is enough for a useful system, even on an old am4 host.
33.
by nyrikki 4mo ago
My workflow is too different right now (gradually constrained to network less builds for reasons) but I am really enjoying how zeds agents have worked out in the past few weeks. I have 27b, 35B-A3B and a cpu backed gpt-oss configured and us
34.
by nyrikki 4mo ago
More that LoC is a simple metric that has always been a problem. Non-Functional requirements is a vestigial term from ‘function point analysis’ which is from the late 70s, and which also ended up being a proxy for LoC. The entire industry i
35.
by nyrikki 4mo ago
Actually Linux was very SysV like back in the day, so it was more like the stuffy OS's that people liked. GCC was the real catalyst, With even SUN which had used bundled dev tools as a early selling point was unbundling them and char
36.
by nyrikki 4mo ago
I think you are a possibly a decade off on the timing here. USL v. BSDi is what impacted the BSD side, and it was during that lawsuit before Novell bought USL etc.... that the problems were that allowed Linux to make gains while the net&#
37.
by nyrikki 4mo ago
Luckily someone with a time machine saw your post and added it to POSIX.1-2001 :) (Sorry if you weren't joking) but yes, posix_spawn() has been a thing and in glibc fork is just a alias to clone() Not exactly that OP idea, but fork&#x2
38.
by nyrikki 5mo ago
It is hard to incentivize docs, and often when there are incentives they make writing docs painful (sharepoint) or time intensive (run books). I am not convinced that just adding llm summaries to a commit will have long term value, especial
39.
by nyrikki 5mo ago
While it may prevent sleep, one of the frustrations I have on an old skylake desktop with a bunch of 1080ti GPUs is that when set to power saving modes, the GPUs sit at 1x2.0 pcie lanes and low wattage. That slows down loading quite a bit.
40.
by nyrikki 5mo ago
It is being moved not going away. It is just a wrapper around zig translate-c, moving it to the build system moves the resulting clang dependency to a place where it can be optional. The functionality is staying, it is just moving.
41.
by nyrikki 5mo ago
The one glaring omission the jumps out to me is Holmström's theorem like effects of incentives. There has been a huge shift towards metrics, to the point that managers are often forced to or at least commonly believe they are forced to
42.
by nyrikki 5mo ago
Two things are being conflated here. Try writing the grammar of natural language in EBNF if you want a fast but unsatisfying answer. The univalence axiom and HoTT is another path. Or just trying to find out why two programs in lambda calcul
43.
by nyrikki 5mo ago
Especially because it will reduce the entropy that constrains the big guys from building a Dutch book (money pump) against the little guy. I am sure there are some very happy people in the larger firms due to this news.
44.
by nyrikki 5mo ago
Currently, with zig 0.16 the agent has to have access to the zig compiler and std library to even produce code that will compile. If you have zig installed, you can run ‘zig std’ to see that. You still have the limitations of attention etc…
45.
by nyrikki 5mo ago
I don’t, it is mostly used for ideas, review, etc… Getting the agent to grep std, example code, comments that reference inaccessible security or bugs etc.. help a little. But for my needs, not refactoring would just be stepping over dollars
46.
by nyrikki 5mo ago
They are getting better at historical data, not at the fundamental issue. As a recent example, I recently had to abandon the multiple LLM reviewer/verifier model I was using because zig 0.16 was released with major changes. I actually
47.
by nyrikki 5mo ago
While this article raises some pain points, it is really bad advice and while it does seem to come up every now and then, the initial flaw with the argument, which typically leads to it confusing newbies while making more experianced users
48.
by nyrikki 5mo ago
The Curry-Howard correspondence formal relationship between intuitionistic logic and typed lambda calculus. Intuitionistic logic doesn’t hold PEM as a priori. Due to many issues like [0] you actually lose the univalent in HoTT. That is what
49.
by nyrikki 5mo ago
The Modern Player Marauder was one of the few Chinese made, Fender branded guitars. I have one and outside of the frets, which wore out, and lower quality fake bone nut it was higher quality than the Mexican ones, when that plant was having
50.
by nyrikki 5mo ago
The naming is bad here. On Linux, most people only install the Docker Engine , unless they want the GUI. On MacOS or Windows you have to install Docker Desktop which spins up a VM running linux. You installed Docker's "Docker
51.
by nyrikki 5mo ago
> Docker Sandboxes require Docker Desktop 4.58+ on macOS or Windows. Linux is not supported since Docker Desktop uses platform-specific virtualization (Apple Virtualization.framework on macOS, Hyper-V on Windows). Docker can launch mach
52.
by nyrikki 5mo ago
There is no security barrier at all in UNIX(-like) Os's between a caller/callee, this is not thing that python can just fix. There are ways to harden and/or reduce privileges, but shells/scripting languages will always h
53.
by nyrikki 5mo ago
It was an abandoned path even before 3.10, it just took longer to implement 649 and 749 than they expected. But this is a "...will continue to exist with its current behavior at least ..." is an important bit there. From pep-0749
54.
by nyrikki 5mo ago
As TLA+ uses state machines that can define infinite state spaces, checking arbitrary temporal logic formulas is undecidable in the general case. Using a TLC model checker to verify invariants or properties, finding a valid counterexample c
55.
by nyrikki 5mo ago
Yes, there are better tools with ggml-org/gpt-oss-20b-GGUF where you can see a less terse refusal for the prompt "Did the FBI send a letter and audio tapes from a wiretap to MLK jr. telling him to commit suicide or they wo
56.
by nyrikki 5mo ago
That config looked too complicated, getting rid of the --prio 3 and --poll 100, setting the draft-n-max to now recommended values, etc... kicked it up to 61 t/s I think that was all about some earlier crashes. podman run --devic
57.
by nyrikki 5mo ago
(Note UPDATED config) Ya, if you are using the CPU it may slowdown quick. This may be a bit huge and overcomplicated, on this host I am running it on a AMD Ryzen 7 5700G so that I can use the APU to dedicate the 3090. podman run --dev
58.
by nyrikki 5mo ago
Bubblewrap tends to be better at defaults than docker/rancher/podman, where the users rarely use `USER` and/or drop elevated privileges, it still has the same limitations. It is just the reality that namespace/seccomp&#x
59.
by nyrikki 5mo ago
You can get all the Qwen 3.x models up to ~1 million tokens using YaRN with llama.cpp.[0] Personally I am using `--no-context-shift` and feeding in context back in on failure at the harness level. I have 2x1080ti + 1xTitanV that have a full
60.
by nyrikki 5mo ago
Namespaces and cgroups allow for resource accounting and limited isolation between trusted processes. It is only through hard work and luck that they have been usable in the k8s/docker world. To be 100% clear, namespaces are not

More ›