4 ms·
WASM makes browsers simpler, not more complex. It's much easier to get the implementation of a 20 page spec correct, than the combinatorial monster that is 200
by j-pb 4y ago
WASM makes browsers simpler, not more complex.
It's much easier to get the implementation of a 20 page spec correct, than the combinatorial monster that is 200 highlevel language specs and APIs.
The more we can push into a small formal core, the better.
Formal verification tools call this the de Bruijn criterion. You create a small formal core for your proof system, and everything highlevel just compiles to that.
- RjQoLCOSwiIKfpm 4y agoSurely that is a noble goal! But isn't it a case of XKCD 927? https://xkcd.com/927/ https://xkcd.com/927/ I.e. I would say the probability is zero that because WASM exists other existing complexity of browsers will be removed. Because the web is so vast, if you once add a feature to browsers you can never remove it because that would break an unknown amount of websites, or even intranet sites. So now we have wasm + N other subsystems, so N+1, and the security of N+1 systems is less than that of N.
- j-pb 4y agoWe break/deprecate the web all the time. Web backwards compatibility is a myth, it's just that wo don't get any calls from the 90s complaining that their dogs website broke.
- genjii931 4y agoHow long have you been using the Web? ActiveX? DYNSRC? Frames? Flash? How about Gopher support? Capabilities being removed is a time-honored tradition in the browser world.
- illiarian 4y ago> It's much easier to get the implementation of a 20 page spec correct JS spec was also 20 pages. Until we got modern Javascript. wasm spec will grow. Just look at the roadmap: https://webassembly.org/roadmap/ https://webassembly.org/roadmap/
- j-pb 4y agoFeature creep is an orthogonal issue that I 100% agree with. We need to stay vigilant and push back against bloat wherever possible. Exceptions are a mistake imho, memory64 is an absolute must have otoh.