Twitter | Pretraživanje | |
Lin Clark
stuffing my head with code and turning it into . also, tinkering with WebAssembly, and a little JavaScript at
7.978
Tweetovi
779
Pratim
32.436
Osobe koje vas prate
Tweetovi
Lin Clark proslijedio/la je tweet
Heather Miller 28. sij
Heya 👋 Want to get research experience at CMU? If you're interested in PL, distributed systems, software engineering, etc we have a program that pays you to come learn to do research with us over the summer! Happy to answer Qs about it! RTs welcome🤗
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 26. sij
Odgovor korisniku/ci @copyconstruct @tbmcmullen
Oooo, fantastic list! I'm looking forward to digging in and watching them 😁 I'm not surprised that 's CodeMesh talk might have made it on this list... his talks are 💯. I wonder if he might have the slides up somewhere that isn't indexed by Google
Reply Retweet Označi sa "sviđa mi se"
Lin Clark proslijedio/la je tweet
HJ Chen 23. sij
"Our paper suggests that the way we train bees to develop numerical abilities has implications not only for human learning but also for artificial intelligence and design."-
Reply Retweet Označi sa "sviđa mi se"
Lin Clark proslijedio/la je tweet
Surma 14. sij
Odgovor korisniku/ci @linclark
It’s only appropriate for the first ’s opening keynote to be given by someone who has been part of WebAssembly from the very start. We are honored to announce as our opening keynote speaker!
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 7. sij
Odgovor korisniku/ci @bsletten @libel_vox
Thank you! 😊 The WebAssembly posts are all on the Hacks blog (I need to update the code cartoons site to point to them one of these days)
Reply Retweet Označi sa "sviđa mi se"
Lin Clark proslijedio/la je tweet
The Best or My Favourite? 3. sij
This is a good time to brush up on non aggressive bystander intervention below. The safest practices center the target for support, not escalating tension with the attacker. Also, don’t expect gratitude from the target who may be confused or traumatized. You’re not a hero.
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 3. sij
Odgovor korisniku/ci @chad_b_morrow @kmelve
Thank you for the kind words! 😊
Reply Retweet Označi sa "sviđa mi se"
Lin Clark proslijedio/la je tweet
Jenn Turner 30. pro
Hi again 👋 I’m starting a new project highlighting your favorite folks in tech who *aren’t* engineers/devs/programmers. If you know someone who I should definitely interview, please let me know here:
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 1. sij
Odgovor korisniku/ci @MaxGraey @MoreSocialism
Indeed, WebAssembly is sandboxed by default in browsers in a way that JS is not, so the only way a wasm module can do *anything* except heat up your computer is if your JS gives it the capability to do so. I write more about that in this post:
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 1. sij
Odgovor korisniku/ci @MoreSocialism @MaxGraey
If you don't care about meaningful variable names, then you can use a decompiler like wasmdec to achieve the same results with wasm
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 1. sij
Odgovor korisniku/ci @MoreSocialism @MaxGraey
I'm not clear on what you're getting at here. When I was talking about source maps, I was talking about deobfuscation of minified JS. While there have been some experiments around using source maps for Wasm, it wasn't a good fit, IIUC
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 1. sij
Odgovor korisniku/ci @MoreSocialism
Almost all JS that is shipped to browsers is minified, so can't be read. The way that developer tools show the original JS is using debug info (source maps) to de-minify it. In the same way, debug info can be used with WebAssembly so that you can view the original source code
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 31. pro
Odgovor korisniku/ci @munificentbob
So exciting, congrats!! 🎉
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 23. pro
Odgovor korisniku/ci @kevmoo
- indeed, a sufficiently clever runtime could optimize things quite a bit, but there are some limits to what can be done (as compared to something like Rust compiled to native code). Since these modules are already compiled to wasm, some of the required information is lost
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 23. pro
Odgovor korisniku/ci @kevmoo
- developers could use wasm threads in their code, but nanoprocesses aren't run on different threads by default - args aren't always copied. This one's a little compilicated but the interface types explainer has more detail:
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 23. pro
Odgovor korisniku/ci @kevmoo
To answer a few of your questions: - nanoprocesses are meant to run in-process, though there would also be an option to make individual nanoprocesses run in their own separate process (for when timing protection is required)
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 23. pro
Odgovor korisniku/ci @kevmoo
It's still pretty early days for this architectural pattern, so the technical details are mostly spread out across the different proposals
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 23. pro
Odgovor korisniku/ci @kevmoo
This is a fantastic compliment 😊 Thank you for the kind words!
Reply Retweet Označi sa "sviđa mi se"
Lin Clark 23. pro
Odgovor korisniku/ci @Sorseg @carols11cents @steveklabnik
Not at all, please feel free to!
Reply Retweet Označi sa "sviđa mi se"
Lin Clark proslijedio/la je tweet
WebAssembly SF 18. pro
Code cartoons. A safer future. nanoprocesses & - what more can you ask for!?
Reply Retweet Označi sa "sviđa mi se"