-
-
Replying to @slightlylate @michaelbenin
Here's video of the difference: https://www.webpagetest.org/video/view.php?id=180811_7aef36019aeadf3cd2fa8e86850537958bebf842 … And the timelines: https://www.webpagetest.org/video/compare.php?tests=180811_W9_9225fd126256bd1721ce5b4d8d8834d4-r:1-c:0,180811_7P_df566597ec85868e288a2867ca0dac4c-r:1-c:0# …
1 reply 0 retweets 0 likes -
Replying to @slightlylate @michaelbenin
Without getting into the first-party content, we can pretty definitively say at this point that the third-parties are sinking this experience. It's gonna be a bad time without an ad blocker *no matter what*.
1 reply 0 retweets 1 like -
Replying to @slightlylate @michaelbenin
Lets dive into the 1p-only trace (4G, Moto G4): https://www.webpagetest.org/result/180811_7P_df566597ec85868e288a2867ca0dac4c/ …
1 reply 0 retweets 0 likes -
Replying to @slightlylate @michaelbenin
I'm guessing this looks a lot like what the developer sees with devtools emulation turned on: https://www.webpagetest.org/video/compare.php?tests=180811_7P_df566597ec85868e288a2867ca0dac4c-r:1-c:0 …
1 reply 0 retweets 0 likes -
Replying to @slightlylate @michaelbenin
A few things jump out: 1.) SSR is indeed getting some content on screen relatively quickly, but at the cost of a pretty bloated initial payload; HTML is ~90K/575K (wire/uncompressed). 2.) The JS bundles are huge; more than 200K/1MB for two main 1p packages.
3 replies 0 retweets 0 likes -
Replying to @slightlylate
You're forgetting the site is usable without js to begin with.
1 reply 0 retweets 0 likes -
Replying to @michaelbenin
Not forgetting. I just don't give much credit to things you can scroll but not tap. SSR + heavy JS bundles = perceived brokenness.
@kevinpschaaf illustrated this brilliantly:https://www.youtube.com/watch?v=0A-2BhEZiM4 …3 replies 0 retweets 0 likes -
Replying to @slightlylate @kevinpschaaf
The links work when tapped. It's a full page reload when no js is available. Bit yes slideshows are probably broken. The v1 of this had slides on their own urls.
1 reply 0 retweets 0 likes -
Replying to @michaelbenin @kevinpschaaf
That's the thing tho; links don't work when tapped when the main thread is busy. So this much JS breaks your site until it's done running.
1 reply 0 retweets 0 likes
Need to run out for a bit; will pick this up later. Until then, suggest watching Kevin's talk in full.
-
-
It's important to understand that JS is a run-to-completion language. Work you do on the main thread blocks the browser's ability to do *anything else* on that thread, including dispatching taps/clicks
1 reply 0 retweets 0 likes -
Scrolling only works because browsers got so exasperated at pages blocking scroll with script that we implemented exotic workarounds ("off thread scrolling" + heuristics around touch listeners)
1 reply 0 retweets 0 likes - 3 more replies
New conversation -
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.
& Web Standards TL; Blink API OWNER
Named PWAs w/
DMs open. Tweets my own; press@google.com for official comms.