[08:48:16] <phedenskog>	 Krinkle: Have you seen https://developers.google.com/web/updates/2019/05/paint-holding ? 
[08:49:30] <phedenskog>	 That come as reason why metrics started to glide between the one reported by our synthetic tools and the directly from Chrome.
[08:49:44] <phedenskog>	 And then Chrome mimics the behaviour of Safari? :D
[09:00:56] <phedenskog>	 But that seems wrong since we test in first view. 
[13:06:51] <Krinkle>	 phedenskog: btw I did see the chrome update on print for lazy load. Nice progress indeed. We can run some tests and see how it goes also in other browsers, eg PDF from Firefox and mobile safari probably most common
[13:07:29] <Krinkle>	 And then maybe start experimenting with more img lazy load for thumbs
[13:07:51] <Krinkle>	 Wordpress released some posts as well how they evolved their approach.
[13:08:11] <Krinkle>	 I believe in the end they set it for the third and later images only
[13:08:43] <Krinkle>	 Parsoid is not going to like that global state in the parser though..
[13:10:45] <Krinkle>	 https://duckduckgo.com/?q=site%3Amake.wordpress.org+lazy+load+images&t=fpas&ia=web
[13:11:21] <Krinkle>	 "to achieve better LCP values out of the box."