James Snell
@jasnell
#FDJT. He/Him JavaScript In Depth
In the category of New-Awful-Things-You-Can-Do-With-JavaScript-Because-It-Just-Lets-You-But-Should-Not-Ever-Do Explicit Resource Management Edition:
Lots of people have been asking me about my dogs when I attend conferences. Since so many people have shown they care, I had stickers made! These will likely be making an appearance at @renderatl.com
Very happy to say that @cloudflare.social is a Diamond level sponsor for this year's @nodeconf.eu ... also thankful for all our sponsors!
Reviewing my kid's merge requests at work is fun ... When I hit "request changes" I get to pretend he actually has to listen!
You're not alone, @jenmsft.bsky.social ... My favorite one.... Which is obvious because taking your time and not rushing through it is always better.
The nest is officially empty. Youngest son driving away with the last of his stuff on his way to new adventures.
32 parallel server instances, with reusePort=2, simple hello world benchmark on localhost (h2load to node.js quic).. roughly 2 million http3 req/s ... not a bad result. No errors, no memory leaks, no crashes. That's enough perf tuning for the day.
There's a certain irony in these activity metrics on the quic blog post ...
About an hour or so before this it was hailing, thundering, pouring down rain, and we even had a tornado warning.
Great Node.js Collab Summit in London the past 2 days. Lots accomplished.
One thing I don't really talk about much is the fact that my oldest son (now 26) is autistic. He's still working on figuring the world out. He's needing to do his taxes and just sent me this. Which struck me as so pure and innocent and honest that I had to share.
When Amazon picks up on the fact we have dogs that like to chew on things....
This sits along side easily with the existing Node.js streams and web streams. It's still experimental but I updated the `FileHandle` API in `fs/promises` to support it as an additional option for reading and writing to files. The fun bit are the performance numbers:
Hey, @rauchg.blue ... As one of those Node.js core developers you're mentioning here... I'd appreciate it if you'd keep me out of this squabble. You've no idea what I want and I've got nothing to do with this conversation. I happen to believe in both Node.js AND Workers, which is why I work on both
My workflow with opencode/opus is always extremely interactive. It gets lost in the weeds so easily.
For the browser side... the benchmarks are a bit different but here's how it stacks up against chromium's impl of web streams:
Fixed up some perf issues and benchmark bugs in the new-streams reference impl ... some highlights running comparisons on @nodejs.org @deno.land and @bun.sh ... note each column is just looking at the one runtime, not comparing runtimes against each other ...
... Obviously, take all the usual "Benchmarks are usually fundamentally flawed" caveats into consideration when looking at those numbers. Also, these are the numbers when checked against Node.js. What do the numbers with Deno and Bun look like? Here are Deno's numbers for the same benchmarks: