Hacker News Reader

claytonjyon Fast

We have common words for those two flavors of “fast” already: latency and throughput. S3 has high latency (arguable!), but very very high throughput.

claytonjyon Linda Yaccarino is leaving X

how would you explain how hard he fought to NOT buy twitter?

people seem to forget he was legally forced to buy Twitter after he tried for months to get out of his joke bid, primarily through claiming he was misled about the extent of bots on the platform

claytonjyon Show HN: Tool to Automatically Create Organized Commits for PRs

I used to like allowing squashing or fast-forward merges. Most PRs would be squashed, because most developers write terrible commit messages and use merge where they could rebase. But, if you had a well-crafted set of commits, we could retain them.

I’ve recently switched to using conventional commits and release-please everywhere, but that pretty much forces us into a squash-only world, since even the devs who write nice commit messages don’t want to make each commit a conventional commit; much nicer to do it as the PR title, and more visible

claytonjyon Archaeological evidence of intensive indigenous farming in MI's Upper Peninsula

Calumet, in the Keweenaw Peninsula (just north of Michigan Tech) was nearly made the capital, instead of Lansing, because of this!

claytonjyon Pyrefly vs. Ty: Comparing Python's two new Rust-based type checkers

for decades, big tech contributed relatively little in the way of python ecosystem tooling. There’s Facebooks Pyre, but that’s about it. Nothing for package/dependency management, linting, formatting, so folks like those at Astral have stepped up to fill the gap.

why is type checking the exception? with google and facebook and astral all writing their own mypy replacements, i’m curious why this space is suddenly so busy

claytonjyon Run your GitHub Actions locally

that’s not really what’s new or special about pixi, is it? poetry (poethepoet) and uv can both do variations of this.

From the outside, pixi looks like a way to replace (conda + pip) with (not-conda + uv). It’s like uv-for-conda, but also uses uv internally.

Better task running is cool, but it would be odd to use pixi if you don’t otherwise need conda stuff. And extra super duper weird if you don’t have any python code!

claytonjyon Run GitHub Actions locally

anyone have any tips for testing actions locally, rather than workflows?

Despite the name, act is really only for the latter. You can try to test a local action by putting it in a workflow and calling that, but if you do a checkout in your workflow that will overwrite the mount of your local code into the act container, meaning you’ll get the version from the remote branch instead. Depending on the action, you may not be able to comment out the checkout step while testing.

claytonjyon Burrito Now, Pay Later

That’s one of the arguments in the substack post. BNPL is like credit cards, but more specific, and that specificity allows for more accurate pricing of the loan risk.

claytonjyon Big Book of R

Not OP, but i think python is very far above R for engineering stuff. I built my early career on R and ran R user groups. R is great for one-off analyses, or low-volume controlled repetition like running the same report with new inputs.

For engineering stuff i want strong static analysis (type hints, pydantic, mypy), observability (logfire, structlog), and support (can i upload a package to my cloud package registry?).

For ML stuff, i want the libraries everyone else uses (pytorch, huggingface) because popularity brings a lot of development and documentation and obscure github issues the R clones lack.

Userbase matters. In R, hardly any users are doing any engineering; most R code only needs to run successfully one time. The ecosystem reflects that. The python-based ML world has the same problem, but the broader sea of python engineers helps counterbalance.

claytonjyon It's all fun and games until somebody loses their retirement savings

money that you need soon shouldn’t be in the broader market.

for retirement savings, balance should shift to bonds or similar as you near retirement.

same principle applies to saving for other large events, like a down payment on a house. decrease risk as even nears.

no need to manage this yourself; every roboadvisor will do this for you rather cheaply

claytonjyon Ironwood: The first Google TPU for the age of inference

Because of the TPUs, or due to other factors?

What even is an AI data center? are the GPU/TPU boxes in a different building than the others?

claytonjyon Nvidia adds native Python support to CUDA

I’ve always thought it was weird GPU stuff in python doesn’t use asyncio, and mostly assumed it was because python-on-GPU predates asyncio. But I was hoping a new lib like this might right that wrong, but it doesn’t. Maybe for interop reasons?

Do other languages surface the asynchronous nature of GPUs in language-level async, avoiding silly stuff like synchronize?

claytonjyon Most promoted and blocked domains on Kagi

Not the point of the post, but i’m surprised to see less than a million queries per day. Is that all of Kagi, or just some subset?

claytonjyon Google’s two-year frenzy to catch up with OpenAI

Suggesting they’ll cancel gemini makes about as much sense as suggesting they’ll cancel gmail.

claytonjyon The Pain That Is GitHub Actions

How do you ensure what a dev builds and tags and pushes is coherent, meaning the tag matches the code commit it’s expected to?

I think builds must be possible locally, but i’d never rely on devs for the source of truth artifacts running in production, past a super early startup.

claytonjyon Switching from Pyenv to Uv

mise is higher level. i use it to install uv in projects with other non-python dependencies (helm, terraform, npm), which i also install with mise.

Then all the python dependencies are managed with uv.

For a non-python project which needs a python-based CLI tool, i’m not sure if i’d use mise or uv (uvx).

claytonjyon War rooms vs. deep investigations

Language can be tricky here. If I revert to an older commit, literally rewriting history to remove newer, bad commits, I think we’d all consider that a rollback. But if I instead add a new commit which undoes the bad commits, is that a rollback or a roll forward?

claytonjyon A year of uv: pros, cons, and should you migrate

yes, but if you’re not in their carefully directed “nemo environment” the nemo2riva command fails complaining about some hydra dependency. and on it goes…

claytonjyon A year of uv: pros, cons, and should you migrate

It does replace pipenv. They might not mention it because it's not widely used these days; most pipenv users switched to poetry sometime over the last half-decade or so.

There is a request for `uv shell` or similar[0], but it's trickier than it looks, and even poetry gave up `poetry shell` in their recent 2.0 release.

[0]: https://github.com/astral-sh/uv/issues/1910

claytonjyon A year of uv: pros, cons, and should you migrate

A scenario for "don't use uv" I hope none of you encounter: many nvidia libraries not packaged up in something better like torch.

Here's just one example, nemo2riva, the first in several steps to taking a trained NeMo model and making it deployable: https://github.com/nvidia-riva/nemo2riva?tab=readme-ov-file#...

before you can install the package, you first have to install some other package whose only purpose is to break pip so it uses nvidia's package registry. This does not work with uv, even with the `uv pip` interface, because uv rightly doesn't put up with that shit.

This is of course not Astral's fault, I don't expect them to handle this, but uv has spoiled me so much it makes anything else even more painful than it was before uv.

claytonjyon "We're building a new static type checker for Python"

the packaging one is an odd example because it’s hard to imagine a company that both needs an enterprise packaging solution external to their cloud provider, and is python only

i have a similar concern for the Pydantic folks commercializing via observability platform; how big is the market for python-only observability?

i think both cos will either look for money elsewhere, or find themselves having to deal with at least a few other language ecosystems. Even python+typescript would 10x the TAM or more

claytonjyon "We're building a new static type checker for Python"

is Django working towards using native types instead of all they stuff they invented before type hints existed?

claytonjyon Ruff: Python linter and code formatter written in Rust

has Astral confirmed they’re doing a type checker? i’m not surprised, everyone has been asking for it, but i haven’t seen them say they’re doing it

claytonjyon Uv's killer feature is making ad-hoc environments easy

this sounds like it’s asking for trouble! very easy to mess up your whole system

i would highly recommend only using —-system in a docker container or similar

claytonjyon Uv's killer feature is making ad-hoc environments easy

It is very difficult for most conda users to maintain conda environments. They use the same env for nearly all their work, don’t understand the hierarchical nature of conda envs, don’t know which one they’re installing into, install stuff with pip without recording it in their env file, etc. The worst local environment messes i’ve ever seen always involve conda.

It can be used effectively, but does not make it easy to do so.

uv makes itself harder to misuse

claytonjyon Making AMD GPUs competitive for LLM inference (2023)

I think you’ve mixed up your Triton’s; I’m talking about Triton Inference Server from NVIDIA while you’re talking about Triton the CUDA replacement from OpenAI

claytonjyon Making AMD GPUs competitive for LLM inference (2023)

The cuda situation is definitely better. The nvidia struggles are now with the higher-level software they’re pushing (triton, tensor-llm, riva, etc), tools that are the most performant option when they work, but a garbage developer experience when you step outside the golden path

claytonjyon MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUDA Moat Still Alive

I was surprised to hear recently that the same happens at NVIDIA! Hopefully less frequently, but I can understand why it's hard to keep many units on hand given the level of external demand.

claytonjyon Uv, a fast Python package and project manager

Have you tried pixi? I haven’t, and i’m hoping i can continue to avoid conda, but it’s a uv-using conda replacement: https://pixi.sh/latest/