Yeah, they seem to have agreed with this and immediately closed the issue:https://github.com/facebook/react/issues/17154#issuecomment-544635255 …
-
-
Replying to @CiPHPerCoder
Remko Tronçon Retweeted Henrik Joreteg
To illustrate the mindset: https://twitter.com/HenrikJoreteg/status/1186885531692457986 … (Compressed) LibSodium-Plus takes you half way to an 'atrocious monstrosity'

Remko Tronçon added,
1 reply 0 retweets 0 likes -
Replying to @remko
I'd like to be able to use the built-in crypto module as an alternative backend for people who obsess over filesize, buthttps://github.com/nodejs/node/issues/30097 …
2 replies 0 retweets 0 likes -
Replying to @CiPHPerCoder
TBH, I don't think anyone cares about size on backend. From your comments, I take it SubtleCrypto from the Web Crypto API is not good enough? E.g. the spec only mentions "P-*" curves.
1 reply 0 retweets 0 likes -
Replying to @remko
First: I have a very negative opinion of WebCrypto as a spec. But let's put that aside for now. I wanted Ed25519 / X25519 as a basis for an early "fewer dependencies" sodium-plus. P-* curves aren't compatible with libsodium.
1 reply 0 retweets 0 likes -
Replying to @CiPHPerCoder
Yeah, disappointing! I expected someone at least asking for Curve25519 in the Web Crypto API (even just as a browser-specific extension), but can't find that either. I would expect having it as a web API would be the way to get it 'accepted'. I guess browser crypto is too niche?
1 reply 0 retweets 0 likes -
Replying to @remko
The problem is that WebCrypto is a potluck standard, designed by committee, with the initial goal of implementing DRM in browsers, not offering secure cryptography for general use.
1 reply 0 retweets 0 likes -
Replying to @CiPHPerCoder
That doesn’t mean it can’t be (made) to serve a useful purpose. But that would probably be a long term investment: would take very long to get implemented, if it gets accepted at all. If the curve is all that is missing, maybe it would be doable, and a POC can be easily created?
1 reply 0 retweets 0 likes -
Replying to @remko
Better idea than trying to fix a broken, bloated standard by increasing the bloat: Replace it with a superior standard.
1 reply 0 retweets 0 likes -
Replying to @CiPHPerCoder
Sure, but that would be more work, and would require a very convincing case I guess. Not sure if this is realistic, maybe someone like
@agl__ knows.1 reply 0 retweets 0 likes
The WebAssembly standard interface doesn’t have a crypto module yet. It will be badly needed, but no one is working on it. It’s a good opportunity to design something better than WebCrypto.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.