to me it'd have to come from serialization perf, I'm just not sure we can outdo sharedarraybuffers
-
-
Replying to @_developit @blasten and
I doubt we can.
@wycats's theory is that avoiding the IDL layer for many mutations helps.1 reply 0 retweets 1 like -
haven't yet read, but seems to pair well with MutationRecord.
1 reply 0 retweets 1 like -
Replying to @_developit @annevk and
another part of my theory is that having a built-in story w/o gotchas would 1/
1 reply 0 retweets 1 like -
Replying to @wycats @_developit and
help, but indeed that could be done in userspace. I'll be working on it in glimmer 2/
1 reply 0 retweets 1 like -
-
Replying to @wycats
a big thing I’m curious about is performance of inserting 1000s of custom elements that then get constructed.
1 reply 0 retweets 0 likes -
-
Replying to @wycats
either, doubt the difference ends up mattering much since that is all “user” code.
1 reply 0 retweets 0 likes -
Replying to @annevk
depends whether we could tweak the timing of construction in that case since we know it's one whole transaction.
1 reply 0 retweets 0 likes
as well as whether we can reduce the number of mutation observer calls.
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.
3/3