Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Done.

amp.dev/documentation/guides-and-tutorials/learn/spec/amphtml/



An open standard that includes rules like "your page must include this Google controlled JS url which will contain whatever we want".

That pretty much kills the "open" bit.


Agree, but technically if the spec is well-defined browsers can recognize that JS URL as a magic string instead of actually loading that JS, sorta like how browsers don't actually go to w3c.org when they see an HTML 4.01 doctype.

Perhaps the right place is to lobby the WHATWG for an explicit exception in the definition of the script tag to handle these ampproject.org URLs in an API-compatible way.


That appears to be how it works. Here's an AMP page served from Bing's cache. No resources are loaded from Google. view-source:https://www.bing.com/amp/s/amp.azcentral.com/amp/1912438001


If you don't load the js, the amp specific tags don't work. So, for example, no images.


I mean that browsers should implement support for the AMP-specific tags directly, without needing to load the JS to do so. If the AMP spec is complete, then this should theoretically be possible.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: