sgammon

fun and beautifully done. i learned a bit about the breach from this

Tables with headers that scroll differently from their body is a horror I had not imagined until I clicked this link

This is literally just a scheduler over firecracker man what

> The big companies can afford the compliance teams and lawyers to keep up, and each new requirement makes it harder for anyone else to catch them

It also takes a lot to do anything with a 1T+ parameter model. Arguably there is still an immense capital moat. So I'm not sure it would even have the effect the author intends.

Releasing the weights for a 1T+ parameter model doesn't really help "the little guy" when it costs ~$50K per H200 and 8 of them aren't enough to run a single node.

Somehow this plan manages to lean into every single bad outcome

Chrome was invented in 2008. Abseil[1] didn't exist publicly until 2017, derived partly from Chromium.

HarfBuzz wasn't ported to C++ until 2010. I don't know the entire constellation of libs Chrome would use if it was designed in C++ today, but I bet a lot of them didn't exist before Chrome or were even created partly as a downstream byproduct of Chromium.

In other ecosystems: if Chrome were designed in Rust in 2008, there would be no crates.io to pull from. There would be no NPM, since Node followed V8/Chrome's invention in 2009. Maven Central had ~50K packages in 2009[2], which was a lot back then. In 2026, Maven Central hosts over 3 million.

[1]: https://github.com/abseil/abseil-cpp

[2]: https://mvnrepository.com/repos/central

That’s right, they are. I don’t think I said they weren’t.

If you are selling the code as the product, it could be said to be an “actual” software product, rather than a mechanism for selling other products, or software as a substrate for some other business.

Those other businesses aren’t any less valid, of course. They just aren’t software businesses. They are businesses that (quite sensibly) use software.

We live in a time now where libraries are free and plentiful and pretty good. It is easy to forget that this was not the case for the vast majority of time that computers have existed.

Chrome’s original design definitely predates our current rich ecosystem. Plus, V8 and Chrome’s performance profile are demanding enough that there would be a pretty high bar for those dependencies anyway.

so… commercializing the same features that are free in omp? what’s the business model? you guys aren’t open source?

omp mentioned :)

love this, and the name "nub" is conceptually perfect

(1) adopts existing tech and improves, rather than a greenfield project with worse compat

(2) the way it uses existing node mechanisms to implement itself (+oxc) is very smart

(3) i like that the node.js-forever crowd has a way to use the newest stuff

success has many fathers

nice to see JS/python interop becoming a thing

Yes, there are many ways to do that now, in under 5 minutes. Cloudflare will set all of that up just fine. GSuite is much easier to set up than CPanel.

End of an era

did you even read and edit the title of this post?

bad news for big shoe

shoe shine boy is trying to sell me compute

Amazon

Would have been much cheaper in the first case on Cloudflare

> Claude simply cheats here and calls out to GCC for this phase

I see

Wow what a terrible and misleading article

is it not a file exfiltrator, as a product

Suggesting this in 2025 is wild

People absolutely might confuse a DI framework for Android with this, seeing as lots of teams use both? They are both software products, to start?

> > can’t live up to owning it.

> ?

Dagger, the DI framework, is very popular, supported by Google, and the primary way to do DI on Android (for many). It is not going anywhere. It is unlikely that this tool or product will create a bigger audience. Therefore, it might never "live up to the name," compared to the benchmark set by current Dagger. It's a simple concept.

People aren’t going to confuse a physical knife with this.

Android is the most popular operating system on earth. Names can overlap but you shouldn’t choose one where you can’t live up to owning it.

“There’s nuance” yeah as much in my comment as the headline at the top of the article

No it isn’t, otherwise Closure Compiler would (still) be running MS365

Dagger is already a very popular thing -- a DI framework