@bz_moz @YuryStrozhevsky @sleevi_ discrete test execution time should be part of all conformance tests.
-
-
Replying to @rmhrisk
@rmhrisk@YuryStrozhevsky@sleevi_ The W3C conf. test harness doesn't measure this. But also, the numbers would be misleading is my point.1 reply 0 retweets 0 likes -
Replying to @really_bz
@rmhrisk@YuryStrozhevsky@sleevi_ Consider an impl that's slow in 90% of normall unexercised edge cases, but fast in cases that matter.2 replies 0 retweets 0 likes -
Replying to @really_bz
@rmhrisk@YuryStrozhevsky@sleevi_ Again, measuring perf is important. Measuring it in half-assed ways often does more harm than good.1 reply 0 retweets 0 likes -
Replying to @really_bz
@rmhrisk@YuryStrozhevsky@sleevi_ Because we don't want browsers overoptimizing edge cases at the expense of what people actually use.1 reply 0 retweets 0 likes -
Replying to @really_bz
@bz_moz@YuryStrozhevsky@sleevi_ if your going to spend time automating test execution time of discrete tests is no brainier.1 reply 0 retweets 0 likes -
Replying to @rmhrisk
@bz_moz@YuryStrozhevsky@sleevi_ what you do with the data is a separate issue.1 reply 0 retweets 0 likes -
Replying to @rmhrisk
@rmhrisk@YuryStrozhevsky@sleevi_ What people do with the data in practice in cases like this is misuse it.2 replies 0 retweets 0 likes -
Replying to @really_bz
@rmhrisk@YuryStrozhevsky@sleevi_ In particular, if you publish numbers, people immediately start using them in news coverage.2 replies 0 retweets 0 likes -
Replying to @really_bz
@rmhrisk@YuryStrozhevsky@sleevi_ So it becomes good PR for browsers to optimize useless things, which diverts resources from useful ones.1 reply 0 retweets 0 likes
@rmhrisk @YuryStrozhevsky @sleevi_ And the upshot is that you should only publish numbers for things that really need optimizing. :(
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.