Catching up on AMP News

Avatar of Chris Coyier
Chris Coyier on

The big news since we last talked about AMP is that the AMP team announced that there will be a way for non-AMP sites to make their way into the coveted Google search results carousel. Malte Ubl:

Based on what we learned from AMP, we now feel ready to take the next step and work to support more instant-loading content not based on AMP technology in areas of Google Search designed for this, like the Top Stories carousel.

You can’t do it now, nor is it clear exactly how you’ll be whitelisted, but apparently Google is open to it. The ticket in will be utilizing hopefully-someday standardized things like Feature Policies, which, like AMP, disallow certain features in the name of… well that’s the thing. It used be be in the name of speed, but the messaging seems to be more like in the name of privacy and security now. Also a good goal, just funny that “Pages” is about the last thing left in the AMPcroymn. Sorry.

This is the checklist of related technologies. Makes my eyes glaze over a bit as I know precious little about any of them. Much of it needs to progress through standards processes and browser implementation first, and we’ll just have to see how practical it all is. If it means people stop creating amp-dot version of their websites (the new m-dot), that’d be pretty swell.

Tim Kadlec and Yoav Weiss got shout-outs in that AMP announcement for their idea of a “Content Performance Policy“. As in, if a website promises and upholds the promises of being fast and not doing nefarious crap, it gets the same benefits as an AMP site.

I mention that as Tim has some new AMP data to share, directly related to AMP’s speed. Tim found 50 random news article pages that were using AMP and tested them various ways. I’ll summarize the results in best to worst order:

  1. The non-AMP version: not great
  2. AMP version standalone: a bit better
  3. AMP version with cached copy of AMP library: a good bit better
  4. AMP version from Google Search: instant

Tim mentions that it isn’t very useful to test against those instant results:

evaluating AMP’s performance based on how those pages load in search results tells us nothing about the effectiveness of AMP itself, but rather the effectiveness of preloading content.

If we can adhere to some standards-based stuff and get Google doing that same preloading for those of us who’d rather not amp-dot, that’d be great. Show me the money. Did I do that movie reference right? I never saw that movie.