4 ms·
Right, but the core reason they’re doing it that way is that it is a hard problem to solve practically speaking (rather than theoretically) without tanking perf
by abritinthebay 8y ago
Right, but the core reason they’re doing it that way is that it is a hard problem to solve practically speaking (rather than theoretically) without tanking perf.
Which it will anyway so WASM devs will end up having to be just as careful as JS devs about DOM & UI manipulations or run into the exact same problems.
At which point (the the same problems and issues) there’s really no reason to use anything other than JS for that portion of the application.
WASM is great, but it’s going to be for the parts of the web that JS is not good at dealing with - and that’s less than the crowd here on HN would like to believe.
- mpweiher 8y ago> hard problem practically (rather than theoretically) Err...actually, the theory is hard, and that is a practical problem, because the WASM people actually wanted to have a formal semantics for WASM from the get-go. That's why they're moving slowly, only releasing what they can specify formally.
- armitron 8y agoJavascript is a horrible language, this by itself is a good enough reason to use something else to a lot of people out there. I think the number of programmers who are currently doing JS but hate it and would jump to something else the moment it becomes viable is enormous. Then there is also the issue of the proliferation of substandard JS programmers, a lot of them knowing nothing about basic CS principles, that are viewed as "necessary evil" today. Opening up that space to better programming languages will be a net gain on many fronts.
- abritinthebay 8y ago> Javascript is a horrible languag Thank you for putting that first so everyone can very quickly realize you have nothing to contribute to this discussion and very little knowledge in the first place. It really moves things along quicker.