There are so many other things you could do that fixes your problem WITHOUT handing over the internet to a single company.
Also, the parent said there was many things the GP could do, while the GP was talking about as an end user, not a developer. As an end user if my choice is between a fast smooth sleek amp page or a 10 second blank screen and a page that hides the content, has autoplaying videos, and required me to dismiss a 1/3 cookie banner, I'm going to pick the amp version.
One could just as well portray "downranking slow pages" as "boosting fast pages". With the latter system developers would be encouraged to reduce filesizes and JS bloat on their existing pages. As a mobile, and desktop user, I would appreciate that.
It's Google's arbitrary proprietary subset of web technologies.
It's only fast because Google preloads all of JS while you search.
It aggressively hides the origin source, depriving sites of user traffic.
Google prioritizes AMP content over any other content even if that other content may be better and/or load faster.
What does this mean? If I click a link on an amp page, I may end up on the full site, and if not, analytics and ads are supported. What do I lose other than the cost of rendering a page view?
You don't land on the original page. You land on a Google page. Even the URL says google.com. To get to the actual original page, you need to click on the tiny (i) in the corner. And even there the link is styled as non-clickable link.
If you swipe, for example, you end up on the next Google property ^W^W that is a news item that Google displays on its own domain.
And AMP may provably drive traffic away: https://kinsta.com/blog/disable-google-amp/
Yes. So what? If I, from the AMP page, click on a link, I'll go to the full site (I just checked this on CNN's amp pages). Ads and analytics are still supported.
> If you swipe, for example, you end up on the next Google property ^W^W that is a news item that Google displays on its own domain.
This didn't happen when I just attempted it.
> And AMP may provably drive traffic away:
If you read the article, they say that they got more clicks and pageviews, but fewer conversions, likely because they hadn't spent time optimizing the AMP site for conversions.
I realise the irony as this still relies on Google, but: I'm surprised that such user-hostile designs are still around, considering Google presumably takes steps to punish such designs in their search rankings.
You will note that Wikipedia, for example, doesn’t have the type of bloat that an online newspaper article does. Why is that? The question is one of incentives.
It seems pointless to bring up a volunteer run website where people do charity work in a discussion about business practices of businesses that need to have cash flow to pay employees.
True and false. It forced companies to pay attention to page performance... and then create an entirely separate version of every page they have, using up valuable developer time that could have been spent making the original pages faster.
Google should (and now/soon are?) assess the performance of the page and rank it accordingly, not force everyone to make two versions of their site. But this is where the sneaky part of AMP comes in: they wanted that side-swiping carousel that meant you never engaged deeply with any one site and instead stayed on google.com (and kept those google.com ad cookies, natch). There was no way to force people into that without AMP.
If the reading I've done is accurate, Google tried this for quite a long time before AMP. The issue seems to be that there are just too many other factors with higher ROI for speed to be prioritized consistently.
IMO, AMP put companies in a position where speed mattered and most other things didn't. It focused minds in ways that using speed as a factor in ranking did not.
That argument makes zero sense to me. Either they rank faster and lighter sites higher or they don't. AMP is completely orthogonal to that.
With that said, as far as I can tell, the difference between two sites is almost never a simple as a single factor. There are hundreds of things that influence rankings. So companies can and do worry more about other concerns, including squeezing more ad income out of each pageview and getting into more search results in the first place.
AMP took away a lot of the other things that companies would invest developer time in.
Having to edit url would matter little on PC, but massively sux at mobile. And it is overall slower then however slow original page was.
I think the point you're actually making is that without AMP forcing these sites to be faster, they would still be slow. Which I agree with. But as I said in my original post, Google ranking sites according to their performance would have the exact same effect as forcing AMP on people, except that it would let sites choose their own implementation rather than be strong-armed into one controlled by Google.
Again, no. The argument was that time spent focused on page performance would have the same effect. If Google punished slow pages in rankings that is all the incentive needed, not the creation of an entirely new platform.
The m.foo.com usage has been around for ages in Internet time, it could do with some standardization and tooling to ensure it gives the quickest page loads possible.
They do not, and they will not.
I don't like amp. I don't like that unrelated amp results get promoted over the actual content i was looking for (that can't be amped because it's dynamic content) in search results, I don't like that have to use to inferior search engines on mobile to get rid of it, and I don't like how watered down amp sites are.
Maybe Google should just build something like this if usability is in their interest. Someone pitch this idea to them!
Whenever I'm on an AMP website I make an effort to use the "real" version instead by following the link in the top right corner.
Nine times out of ten that non-AMP page content is harder to read – the notifications more intrusive, the content jumps around while loading.
It looks like Google are now taking a more-refined approach to prioritising pages in mobile results, but I like that AMP forced publishing companies to take a hard look at page performance, for the first time in a while.