Raph Levien

YouTube video by Arthur Gleckler I Want a Good Parallel Language, by Raph Levien (BALISP)

My talk on "I want a good parallel language" is now up: www.youtube.com/watch?v=0-eV...

Slides here: docs.google.com/presentation...

Thanks Arthur Gleckler and BALISP for hosting the talk!

The work is based on prototypes I did in the last few months, but is a complete reimplementation with lots of optimizations, including portable SIMD. I believe it's the fastest pure Rust CPU renderer, and also shares a lot of logic with vello_hybrid. 2/2

I'm thrilled that Laurenz Stampfl, a student at ETH Zurich, has completed his Masters degree, and has published the thesis: ethz.ch/content/dam/...

1/2

I Want a Good Parallel Language, RetroAI, lightning talks, and socializing, Sat, Nov 1, 2025, 12:00 PM | Meetup Hello, hacker ladies and gentlemen! On Saturday, November 1st, we'll have a talk by [Raph Levien](https://en.wikipedia.org/wiki/Raph_Levien), who is an expert in Rust and

I'll be giving at talk tomorrow at the Hacker Dojo in Mountain View, entitled "I want a good parallel language." Meetup link is here: www.meetup.com/balisp/event...

Greg Bovino Emerges as Trump’s Man For Conjuring Up a Blue-City Insurrection Elected officials, activists, and at least one former high-ranking Department of Homeland...

Talking Points Memo published this a few weeks ago: talkingpointsmemo.com/news/greg-bo...

Right, in that case if the video supports your new text, you're on firmer ground making the change. Another thing that might be useful is explaining in the talk page what you've done and why. Keep in mind I'm not an expert on WP (wondering if we should tag Molly White).

That said, in the spirit of WP:BOLD (en.wikipedia.org/wiki/Wikiped...) you could just edit the article. I've done that myself on occasion. Wikipedia depends on a nonzero fraction of its users being people who know what they're talking about. 3/2

I might be willing to help facilitate this. As you know, I'm fascinated by Larrabee as part of my "I want a good parallel computer" mission, and there's a dearth of good information out there. 2/2

Sadly it doesn't really work like this, skeets are not reliable sources. If you wanted to do this the right way, someone should interview you and publish it in a reputable place. For example, a lot of my bio information is from a TeX interview (www.tug.org/interviews/l...). 1/2

Trump Administration's Arrival on Bluesky Highlights Growing Pains for Open Networks | TechPolicy.Press The administration’s antagonistic entry to the platform is best understood as a game of chicken, writes Erin Kissane.

There was already a *lot* going on with Bluesky, and now it's under even more pressure. I wrote about how I think about moderation, affordances, expectations, human needs, and powerful trolls for @techpolicypress.bsky.social

www.techpolicy.press/trump-admini...

Some updates from the Linebender ecosystem: we've published our monthly update blog (linebender.org/blog/tmil-21/), and also released fearless_simd 0.3.0 (github.com/linebender/f...). The train keeps rolling!

YouTube video by Topos Institute [Berkeley Seminar] Raph Levien | How Rust won: the quest for performant, reliable software

I gave a seminar entitled "How Rust won: The quest for performant, reliable software" at the Topos Institute on Jun 3, and the video (youtu.be/k_-6KI3m31M) is now published. I hope people enjoy it!

GitHub - ztanml/fast-hash: Fast hash function learned using genetic programming Fast hash function learned using genetic programming - ztanml/fast-hash

Is this a microcontroller that has a 1-cycle multiply instruction, like RP2040 or better? In that case, FNV might be a good choice. If not, Jenkins is a possibility. You might also take a look at github.com/ztanml/fast-....

I've subscribed. We need more explorations of this space, especially from a human rather than business-centered perspective. I'm eager to see where it goes.

We have our usual monthly update for Linebender: linebender.org/blog/tmil-19/

A lot of good progress, especially on the renderer, and fearless_simd is coming along well. Also some personal news: I'm leaving Google and taking a new role at Canva, moving to Australia in January.

With Tom Lehrer's passing, I suppose this is a moment to share the story of the prank he played on the National Security Agency, and how it went undiscovered for nearly 60 years.

The pardon is the carrot. The stick, not just for Ghislaine, is the threat of being accused of treason (or some other offense with capital punishment) if you've got factual information and facilitate that being released to the public, as is being done with the Russia scandal. Timing no coincidence.

When I was little, the U.S. military came to our home at gunpoint and took me and my family away. We were imprisoned for years in barbed wire camps simply because we were Japanese American. I have spent my life telling that story, hoping it would never be repeated.

Cool to see this work happening, and look forward to some nice GPU compute acceleration for path rendering!

Topnotch reporting from Marisa Kabas on a very important topic. It's maddening that mainstream news sources are not meeting the moment. Support independent journalism by subscribing.

Took me a little while to place, but yeah. Thanks!

That was a fun Saturday morning exercise. I did it on pen and paper, got it mostly right (missed a sign, which is a weakness), then validated it in Desmos. I personally choose to consider this not cheating, it's my standard toolset for such things, but maybe I should practice doing it from scratch.

But (though I was kinda expecting to), I still can't repro. I just get an undefined behavior error in debug, and the intuitively correct answer in release. I'm afraid Rust just doesn't have the expressive power for eldritch horror that C does.

fn function(a: u16, b: u16) {
let c = unsafe { (a as i32).unchecked_mul(b as i32) as u32 };
if c < 2147483648 {
println!("{c} is less than 2147483648");
} else {
println!("{c} is greater than or equal to 2147483648");
}
}

fn main() {
function(65535, 65535);
}

I tried porting this to Rust and am unable to repro. What am I doing wrong?

Four here. I was hoping to get 3 by doing paper in collaboration with Norm Megill, back when I was working on Metamath, but sadly he has passed. Another reminder that the time we have here is precious, not to get maudlin on your thread.

...being able write super-performant libraries and then sufficient abstraction to use those libraries. In the limit, you get something like PyTorch (also where MLIR is going), where you write very high level stuff and magic happens to run it on your GPU efficiently.

Ah, makes sense. I think this question is conflating pedagogy ("how do you teach modern CPU and GPU performance") with language design ("how do you write programs that are maintainable and also exploit available performance"). The latter pulls you in a different direction, mostly an emphasis on...

If your question is, "how could CPUs be designed for radically better performance," then I think the answer is GPU-like parallelism but without the impoverished execution model. The problem is that we have no idea how to program such a beast, and it would be weak on existing workloads.

Depending on the exact workload of course, I suspect many problems could get a 2x or more speedup by adopting data oriented design and exploiting SIMD. Especially on modern SIMD with predication (AVX-512, SVE, RVV). Of course, in some domains (media codecs) this is already done.