d@nny "disc@" mcClanahanJan 8, 2024
according to this post, the point of webassembly is to disenfranchise the existing toolchains, which explains perfectly well why there's no documentation about how to use it to integrate existing C/C++ from emscripten into a rust-based wasm project except for a github comment from a microsoft employee, so in that sense it's extremely useful. using the term "chauvinistic" was certainly a decision https://wingolog.org/archives/2024/01/08/missing-the-point-of-webassembly
missing the point of webassembly — wingolog

wingolog: article: missing the point of webassembly

Show threadsean wattersApr 22, 2025

@hipsterelectron i felt like mozilla's intro-to-wasm blog posts were a lot better about explaining the what/why (more in a browser setting), and seemed a lot better oriented toward supporting c/c++ and rust (and beyond) https://hacks.mozilla.org/2017/02/a-cartoon-intro-to-webassembly/

i personally see WASM primarily as a useful sandboxing mechanism for multi-tenant hosting scenarios similar to the way that cloudflare (ik, bad) workers and [i assume] deno deploy are using v8 isolates for server-side javascript.

A cartoon intro to WebAssembly – Mozilla Hacks - the Web developer blog

WebAssembly is a way of taking code written in programming languages other than JavaScript and running that code in the browser. So when people say that WebAssembly is fast, what ...

Mozilla Hacks – the Web developer blog
Show threadd@nny "disc@" mcClanahanApr 22, 2025
@swatters i am fully aware of what wasm is and is good for. i am critiquing andy wingo's portrayal of it here (significant since he is a very clouty implementor) and in general i'm concerned about how people want to (mis)use it to play political games to enclose instead of expand
Show threadd@nny "disc@" mcClanahan
@swatters i am not looking for documentation about it i'm building c++-rust bridges in wasm and the c abi and the fact that the tooling for this does not already exist is a really huge problem and indicative that in fact a lot of people are not using wasm as a useful translation layer between languages but as an exotic compile target in which toolchain support is being used to preclude linguistic interop
Apr 22, 2025 at 9:56amWeb
Show threadd@nny "disc@" mcClanahanApr 22, 2025
@swatters addressing this is not a matter of advertising what it's good for or what's interesting about it. there are a lot of interesting technologies and wasm is not the only sandbox layer. it would be interesting if it was interested in being a cross-language compile target. its continued disinterest in supporting arbitrary CFGs directly because v8 doesn't want to really pokes a hole in the argument that it's supposed to be about more than the v8/browser/chrome environment
Show threadd@nny "disc@" mcClanahanApr 22, 2025
@swatters sandboxed code is possible and cool. if posts like this from andy wingo weren't describing wasm as a force for exclusion and if there were much better tooling for integrating wasm across languages (not having to use webpack for rust for example) i would be a lot more interested in it as an ecosystem
Show threadsean wattersApr 22, 2025

@hipsterelectron sorry, wasn't trying to link to docs or assuming confusion on your part—providing the link to the mozilla posts was more to highlight that my experience of wasm/intro to it didn't include a lot of the "just for chrome" and chrome i think (memory could be bad) was actually far behind firefox back in 2019 when i first started looking at wasm.

if it has since been coopted more strongly for chrome/v8 monopolization that's a bummer. tbh i wasn't even aware of this wingolog person.

Show threadsean wattersApr 22, 2025
@hipsterelectron far behind on wasm support**
Show threadd@nny "disc@" mcClanahanApr 22, 2025
@swatters right now i am personally fine with using the lack of goto support as an excuse not to dabble in wasm yet because i have other priorities