3 ms·
This argument I don't understand https://twitter.com/fasterthanlime/status/1452066487913664512 https://twitter.com/fasterthanlime/status/145206648791366451...
by kzrdude 5y ago
This argument I don't understand https://twitter.com/fasterthanlime/status/1452066487913664512 https://twitter.com/fasterthanlime/status/145206648791366451...
Using the median page load time is one of the best tools for analyzing effective load speedups. Sounds legit to me, and they're just trying to speak to the jury.
- lostdog 5y agoYeah, for this whole case we have to choose whether Google or the Texas AG is being more honest.
- tgsovlerkhgsel 5y agoIn my subjective perception, most of the improvements from AMP came from the fact that all the abusive technology that publishers had amassed on their primary web sites over the past decade or so got temporarily dropped in the forced rewrite. A non-AMP site could be just as fast as an AMP site (minus the initial prefetching) or even faster by not requiring all the AMP-specific stuff, but in practice, the AMP sites were a minimal rewrite of the full site, leaving behind the popups, autoplaying videos, 40 different trackers, ads from 5 different companies, etc. AMP overcame the inertia of just leaving all the crap in and incrementally adding more by forcing the rewrite. Over time, as the publishers had more time to crappify the AMP versions too, they became worse.