4 ms·
that makes me smile and I'll have to shift the goalpost toward what I actually think, which is the reality where the theoretical performance of optimizations i
by vmception 4y ago
that makes me smile
and I'll have to shift the goalpost toward what I actually think, which is the reality where the theoretical performance of optimizations is undermined by the 100 megabyte analytics package that your users are going to have to load anyway
so while we all played around with page scores to optimize how fast things load to theoretically reduce strain on the server or have less users that bounce, its just so far divorced from the user experience because we'll wind up putting so much extra stuff on our pages anyway
- darepublic 4y agoI won't. And for analytics I will proxy via my server
- andai 4y agoFor most applications on most consumer devices, benchmarks comparing the time it takes to update thousands of rows are not really relevant, other considerations will have a stronger impact. However, the lighter frameworks (where most of the work is done at compile-time and the result is highly optimized vanilla JS) really shine in specialized usecases like underpowered embedded devices, where the more popular frameworks introduce too much latency (>0.5s) for a smooth user experience.[0] [0] - https://twitter.com/sveltejs/status/1088500539640418304 https://twitter.com/sveltejs/status/1088500539640418304 (That being said, it is kind of hilarious to me the world we live in where we're running an entire web browser on an embedded device just so we can use JavaScript to program it.)