Skip to content

Google tried to fix the web – by taking it over (AMP)

theverge.com
49 pointsthrowaway20567 comments
On HN

Comments

The problem was that some (news) sites were/are designed by myspace profile artists. Elements jump around the screen 15 seconds after the content loaded.

How many display ads do we want to show on a page? Yes.

…and Google’s now solving this in the way they should have from the start: by penalising (via CLS metrics) sites with this behaviour in the search results.

AMP was an attempt to link hijack the whole net, funnelling everyone through a demostarably worse google implementation. Dystopian is to soft a word.

AMP was a cool piece of technology that was actually faster, they (Google) just had to infect it with bad intentions.

AMP was a shitshow from day one: when every instance of the product had to load a piece of JS from Google’s servers[1] unless you wanted a 8sec delay[2] then you knew that you were in for a bad time.

[1] <script async src="https://cdn.ampproject.org/v0.js"></script> is required on every AMP page

[2] https://news.ycombinator.com/item?id=21714264

I don't recall AMP ever having a "cool, uninfected" form — it was pretty clear what Google's intentions were from the start, using the promise of "rank higher in Google" to strong-arm adoption.

IMHO Matle Ubl will in the future be known as 'The Web Thomas Midgley Jr" for the damage done with AMP - and his relentless push to make it a thing.

Now he's left Google, and they are abandoning it - that's another decade of web content that will just die without the chance to be archived.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.