3 ms·
> The same way it reliable detects the content of the site and everything else. Take a snapshot and keep the latest history. Wouldn't benchmarking page speed a
by seanwilson 9y ago
> The same way it reliable detects the content of the site and everything else. Take a snapshot and keep the latest history.
Wouldn't benchmarking page speed accurately for all pages on a website for a variety of screen sizes on a regular basis cost significant resources? If it's so easy, why are they not doing this already?
What I meant about AMP is that when a page validates as an AMP page you know there's certain things it can and can't do. This makes the task of identifying performant pages significantly easier.
- manigandham 9y agoBenchmarking is just logging the time taken during the process of waiting for the content to load, which is already happening to index the page. No special framework necessary. They aren't doing it because AMP gives far more data and keeps first party cookies because of the cached domain.
- seanwilson 9y ago> Benchmarking is just logging the time taken during the process of waiting for the content to load I don't think it's as simple as you make out at all. Compared to just downloading the HTML and checking its contents, this requires executing JavaScript and rendering content in an actual browser which is orders of magnitude slower and more resource intensive.
- kmonsen 9y agoAnd can be gamed by detecting Google and serving different content to users that if slower.
- manigandham 9y agoThat's exactly what happens already because javascript can also change content, including all the React/SPA sites now. They've been doing it for years: https://webmasters.googleblog.com/2014/05/understanding-web-pages-better.html https://webmasters.googleblog.com/2014/05/understanding-web-...