Dear Twitter, Please Make AMP Optional | Hacker News Reader

Dear Twitter, Please Make AMP Optional

alexkras.com

95 pointsakras1466 comments

I feel like this constant anti-AMP rhetoric has extraordinarily little to do with users, and primarily are web developers who don't like that AMP rolls back a lot of abuses. The argument isn't just "I don't like AMP", it's "No one should like AMP!" and it grows tiresome.

> At the same time I understand that a lot of users DO like AMP content. I completely respect their right to enjoy it. But it would be nice if Twitter provided an option to opt-out from that experience for those who don’t.

- TFA

Right. This person is saying "The last time I bitched on AMP it got hyped, so here it is again". I'm not talking about the author. I'm talking about the crowds on here that vigorously vote up everything anti-AMP.

You're arguing against an imaginary group of mindless AMP haters instead of the actual group of people very concerned with the far reaching implications of AMP on the open web. If it's not your concern, move on.

No, I'm against the actual group on here, having seen this argument play out multiple times. I am not concerned about the implications of AMP, have a long history with the web, and think it is an arguably correct reaction to help protect the "open web". It is impossible to have a rational discussion on here, however, as so many are so viciously anti-AMP.

I bitch about AMP every time it finds an annoying way into my life. I have been bitching about it a lot less, since I disabled it on my site.

It was actually a pro AMP article that was posted here and my response to it that "triggered" me to bitch again.

https://www.alexkras.com/some-thoughts-on-amp-in-conde-nast/

It plays to the crowd. I get it, and I'd do the same.

Your original argument was that you purposefully enabled and published AMP on your site and then complained that Google `stole' your content, which normally would elicit groans and criticism on a technology site, but it played to the crowd. I'd keep playing that tune as long as people will listen.

Don't worry, I got my fair share of criticism. I had no clue what AMP was when I first enabled it, all I heard was that it might help my site rank better and I was sold.

I should have RTFM and never enabled it, but it's not how it played out.

So, your problems were of your own making. Got it.

Yes, and I am very explicit about it from the beginning. Mainly to make sure others don't make the same mistakes.

Instead of arguing against the actual positions claimed, you're arguing against what you imagine are people's motivations.

+1. As a user, I love AMP and don't understand the fuss.

How would you feel if Twitter turned off the support for format that you liked, AMP in your case, even though that format was readily available, and instead forced you to use some other format you don't like?

Regular web pages is the format that I personally prefer to interact with.

Does the format I like tend to be (even if not all) full of weird table layouts and spammy ads?

Most sites that end up peacking my interest are credible sites with regular ads. I only see benefits of AMP when I end up on crappy gossip type sites etc.

I dislike it because it breaks scrolling, sometimes breaks completely (Reddit pages never load under AMP for me, so there's always an extra tap needed, for example), and provides essentially no benefit. Yes, they load faster, you can make pages load quickly without it.

Heh, I’d have thought you of all people would be on iOS 11 beta by now, where scrolling speed is now consistent between normal pages and scrollable iframes (which is what AMP uses). Or are you referring to a different scrolling problem?

I avoid OS betas. I don't like crazy bugs on stuff I'm trying to use. Even the initial .0 releases are iffy.

Scrolling speed is one problem. The lack of scrolling to the top by tapping the scroll bar is another. Is that fixed too?

Yes this is my biggest frustration with AMP. Tap to the top doesn't work on iPhones.

> The lack of scrolling to the top by tapping the scroll bar is another. Is that fixed too?

Nope. (I forgot to test that.)

Yuck. If AMP just uses frames then it seems like WebKit should be handling this properly.

You are right, it has more to do with it not being inline with the ideas of the open web, and not liking the prospects of where this can take us.

An open standard, used by multiple parties (e.g. many services now consume AMP content), based upon HTML, consumable by just about everyone. That's about as open web as it gets.

Secondly, I think the biggest strawman is this notion that it's AMP vs some conceptual "open web". That open web you like has much more formidable competitors and was the reason AMP appeared.

Hosted on google servers?

Scripts pulled from cdn.ampproject.org. Should they be long cached on the browser? Should the browser hot replace references? Etc etc etc. There are a lot of interesting, beneficial discussions that can and should take place about this notion of a "lite" web, but instead it all gets short circuited with a losing defiance. The most tiresome is the recurring proclamation that it isn't really faster because you could technically abide by the same limitations, which completely misses the point.

There are a million amazing discussions that could be had, but there’s little point in starting that when the existing implementation is so rotten.

The cache architecture is open, so that anybody (eg, Twitter) can run their own AMP cache.

Is it though? Spent the last hours trying to find this open architecture but all I can find are API reference but no actual standard or examples...

It's a lot less open than html; and the development is clearly steered by platform companies, and thus is unsurprisingly biased in their favor, over that of lots of other stakeholders.

I don't think AMP would have a snowflakes chance in hell to succeed on it's own merits. It's success now is due to SERP-based strongarming by Google.

Given how simple it is load quickly, I think its inexplicable that AMP is so complex and its development so closed.

What exactly is the benefit that these platform companies derive from AMP? Last time i asked the claim was that google gets some "tracking" benefit by serving the amp.js file the pages have to include, which seems unlikely but whatever. But for Twitter, or Bing, or any other platform that is preferring to link to AMP pages over full-featured HTML, what is the benefit? It can't be tracking for them, they aren't hosting the amp.js file.

The only possible benefit I can see for Twitter to link to AMP is that their testing proves users prefer AMP pages, meaning a better user experience is result when "development is clearly steered by platform companies". When you say AMP is biased in the platform companies favour, what you really mean is it is biased in the user's favour.

As we've seen with the state of the non-amp mobile web, it's not simple to load quickly. Maybe it's simple to implement if that is your goal, but convincing all the involved stakeholders that it's a worthwhile goal is hard, and so the mobile web is slow as hell. The open web had their chance to make fast sites, and failed. So now an industry-led carrot-and-stick style approach is being used, and it's actually working. pages load fast now. And the HN Community, who is responsible for making a lot of those slow sites ("oh! let's load another three analytics platforms, it's only a couple megs") is angry.

Note also that due to browser caching, Google wouldn't even get some sort of "tracking benefit" if they wanted to. Most of these documents do not trigger an actual browser request for these javascript files.

Also, if the webmaster wants, they can simply add `rel=noreferrer` to the script tags. This is still considered valid by the AMP Validator.

Benefit for Google, locking people (making it easy to return) into their "platform" thanks to AMP bar and carousel and control over AMP and by extensions large portion of the web.

Twitter, I can't see benefits, other than what you mention i.e. tests show users preferring it. My premise is that there is a large enough group of people who would prefer to opt out. I have no idea how large or small this group is.

One way this is beneficial is that it allows them to steer e.g. the evolution of stuff like https://www.ampproject.org/learn/who-uses-amp/amp-ads/

Some ways of implementing fast (mobile) loading have the potential to be quite disruptive to Google's business model; AMP is a way to ensure that doesn't happen. Note that AMP is open-ish now, not when it was initially developed. The general model is thus a fait-accompli.

AMP is harmless enough nobody is likely to get all to worked up about it (despite these comments, neither am I!), but it does fit the mold of classic monopoly abuse rather neatly: using dominance in one sector to influence others in your favor.

The real crime against the open web is twitter (and google, and facebook...) replacing all their links with t.co redirects.

Agreed

The biggest frustration comes when Twitter forces me onto t.co which redirects to analytics.twitter.com, which is rendered dysfunctional by my network-level adblocker.

My biggest problem with AMP, as a user with low vision, is that it's impossible to zoom in on images in AMP pages. I have of course checked the box in Chrome (Brave) to force enable zoom, but somehow AMP overrides that because AMP knows best that a detailed infographic or photo of a document should be no wider than my phone's display.

AMP does not do anything in this direction, and certainly not on purpose. If it does please file a bug. We'll treat it as high priority.

Note, that at the beginning of the developer preview 2 years ago, AMP did do something like this, but it was quickly fixed.

No, that’s never the argument I’ve seen. It’s always been fairly well accepted that fast pages are good and AMP helps to make pages fast. The common argument - one which I unreservedly agree with - is that pages could be made fast without the use of AMP and especially the web-breaking nature of Google’s implementation.

The fact that users like AMP is not material - the problems that it causes are not ones they are going to see in the short term.

But the common argument is nonsensical. Yes of course sites could make non-abusive, readable content. But they don't. The non-stickiness of the web, coupled with a low attention span of users, has yielded a sort of tragedy of the commons. AMP is a standard to say "this can't do that".

We should have an HTMLite of the sort like AMP. Like how Google Gears became a variety of accepted web standards, it should build into something rather than the nonsensical claim that it does nothing.

That’s fine. I agree; “HTML-lite” is fine, and most of AMP is even an acceptable implementation of that concept.

But in practice, the way AMP is used has fundamental problems. Like the fact that content is hosted by a third party, or that links are broken, or that scrolling is wrecked. I like the speed of AMP; the implementation is catastrophic and I wish dearly that I could disable its appearance in Google search results.

+1 would love to see HTML light, and that's why I initially assumed AMP to be.

Edit:

Now that I think about it, Twitter does try to use AMP as HTML light, since they don't use the cache.

Yet I think AMP is does not make a great HTML light, likely because it has to support the cache use case.

> ... I wish dearly that I could disable its appearance in Google search results.

You can if you use https://encrypted.google.com

Amp doesn't really break the web any more does the trend of writing sites that generate some HTML boilerplate and then load in all the real content through JavaScript: if anything, AMP is closer to the original model of the web as a collection of interlinked documents than any SPA is.

That sort of thing is also incredibly shit though.

Considrng AMP fucks with sharing links, deep linking etc it's a huge step back to wap era.

It rolls back a decade of responsive innovation.

Enjoy your trojan horse. I agree it is pretty.

Edit: If you can't see the progression, look up this history of froogle. For starters, publishers are accepting a banner that takes up the most valuable space and features an [X] button that goes straight to Google. Worse if your page is in the carousel. And, it gets worse from here. They will tweak the UI over time...it's not your site. It's theirs. Eventually, this turns into an AOLesque private interweb.

AMP is such an egregious attack on open standards it shouldn't be possible. The fact that it is shows the system is broken.

The whole point of having a system is to prevent short term interests from trumping long term health. Tomorrow any monopolist can offer a short term fix for a fraction with the public lapping it up only to run down competitors and capture the market. This is well known monopolist behavior with strong systemic safeguards which curiously are not working.

People keep on complaining about heavy web pages but what happened to faith in markets and letting users decide? Surely if a website is too heavy people will automatically seek alternatives. So there is a contradiction here, does the market work or does it encourage a race to the bottom? Perhaps without Google's monopolist intervention a more open less self interested solution could have emerged.

Google is becoming a monster with an obsessive insatiable appetite for information and knowledge. A push back is long overdue or we will pay with private self serving entities controlling access to our own information and knowledge.

Put this on the list of dozens of settings Twitter should be offering users, like opting out of seeing likes from people you follow, or permanently turning off "in case you missed it".

Yeah, it's a long shot. Yet they took the time to get the AMP in.

The website owner has added markup to a document that states:

"Hey useragent, if you understand this markup, please send mobile users to this other URL that I have provided."

Twitter then respects the website owner's markup request.

If the markup was "here's a mobile optimized page", we wouldn't get this kind of strong reaction in HN. If the markup was "here's a the language X variant of the document", we wouldn't get this kind of reaction in HN.

May be you are right, I'll have to meditate on that one.

Edit:

It's a complicated situation. Google offers a great carrot/stick to get people to adopt AMP. Publishers adopt it. Other platforms start using it. A lot of users like this format. Some users hate it and the idea of it.

As I've mentioned before, I do think that Twitter was trying to do the right thing for their users given the situation we are in.

It would be nice to have "request Desktop version" option for AMP, but not sure if it's too much to ask.

If the publisher provides both AMP and non-AMP versions of a document, it is possible to write a browser extension that will redirect to the non-AMP URL when provided.

You would want to look for <html amp> to detect this case, and then <link rel=canonical> to determine the possible non-AMP URL. If that URL differs from the current URL, redirect to it.

If you personally want a per-user option here to override the publisher, this seems like the way to go, but there is some similarity to ad-blocking here.

Yes, except it's not so easy to do in Safari on iOS.

> Twitter then respects the website owner's markup request.

Since when is Twitter the user agent?

Replace 'user agent' with crawler if you'd like.

Twitter isn't a crawler, either. They have no business changing the target of links that their users tweet.

If Twitter is redirecting to the original URL, what's the problem? It muddies the claim of those that oppose AMP so fervently to mix AMP the framework with the use that Google makes of AMP. The publisher has decided to use this technology to make their mobile page. It's perfectly valid (and well engineered) HTML and javascript. Asking for a non AMP version of a page is like asking for the non React page or the non Bootstrap page.

Google recommends that the "canonical" link always be the one that gets shared. https://news.ycombinator.com/item?id=15086192 Implementations have been ignoring this, but it does seem more predictable, and also less likely to break if a publisher removes AMP later.

And deep linking be damned right? Now we have to mess with the canonical url to include the deep linked resources so that sharing works and a page that could have been static need dinamic generation of their meta

It's not, it's redirecting to the AMP version, even if non AMP link was shared.

See: https://www.alexkras.com/i-decided-to-disable-amp-on-my-site...

Twitter for iOS has an option to open all links in reader mode. It's awesome!

https://medium.com/@pixeldetective/open-twitter-links-direct...

I make a somewhat popular Chrome extension [1] designed to aid website readability. I think this article is spot-on, and I'm going to look into adding a feature to my Chrome extension that does precisely this article suggests (changes "amp=1" to "amp=0" on twitter links).

Ping me if you want to try this feature in beta: nick@beelinereader.com.

1: https://chrome.google.com/webstore/detail/beeline-reader/ifj...

OP does not explain WHY he wants AMP to be optional. In the other article the author says "I don’t know why I do it, but for some reason it just doesn’t feel “right” to me to consume the content through the AMP. It feels slightly off, and I want the real deal even if it takes a few seconds extra to load."

That is a weak argument to persuade a major corporation to change the way something behaves or give the option to change the way something behaves.

OP should put himself in the shoes of Twitter: "What do I get by making AMP optional?"

...now you know what their response will be.

Edit: fair rebuttal, thanks for the response

Well, why AMP, and not plain HTML? It's pretty clear at this point that AMP isn't any faster than simple HTML, and some of the limitations are clearly shift control (and advantage) to platforms like google, away from the hosting site.

I mean; stripped down pages are way overdue, but it's mixed in with some unnecessary powergrabbing and arbitrary restrictions. And it's not like it's hard to strip down pages.

It's a valid criticism of my earlier post, I did have a follow up where I tried to do a better job explaining why:

https://www.alexkras.com/i-decided-to-disable-amp-on-my-site...

Also check out:

https://www.alexkras.com/some-thoughts-on-amp-in-conde-nast/

For why I think it is bad for the publishers.

Reply on news.ycombinator.com