The End of AMP?
lafoo.com
lafoo.com
AMP launched in 2015. At that time, Larry Page was CEO, Eric Schmidt was Chairman and David Drummond was the Chief Legal Officer. I don't know how any of them signed off on AMP.
Giving advantageous ranking (including, but not limited to, showing AMP content in Top Stories) is the very definition of using your market power in one area (search) to force publishers to adopt something else you created.
For a company that is (or should be) very careful about attracting antitrust attention from the US/EU, this seems completely reckless.
And what is the upside for Google here? Fast pages? That seems like one hell of a gamble.
And on the user side? I, as a user, absolutely hate AMP and wish I could opt out.
I have poor vision so use larger fonts on iOS. AMP pages routinely render with the edge of pages cropped off and you can't scroll so here so this happens to me way too often:
1. Click on a link
2. Get an AMP page
3. Can't read the page, click back
4. Force touch the link to bring up the preview of the page, which is the non-AMP version
5. Click on that preview to get the non-AMP page.
Why is this still an issue? Why can't I just get this version of the page always? It's ludicrous.
But I honestly don't know why Google pushed so hard on AMP.
My only real issue with this workaround is image search where I have to turn my phone sideways to see the images.
Pessimistically, seems to be yet another case of a huge company doing something awful with absolutely no consequence.
What has the last time Google was punished? The biggest threat to its business model was the gdrp and it has achieved next to nothing.
Quite honestly, the concept never really made any sense to me, and often the load speeds were just as poor as the original website itself.
One thing that really bugged me was how disorientating it was for the user; on some AMP pages, you literally could not tell if you were on a legitimate website or some spoofed version. The AMP bar at the top was about as trashy a UI design as I've seen since 1997. And, God help you if you wanted to share the link - it was hard to know if the AMP link would be "persistent" or "sticky" enough to share, or whether you should go to the actual underlying web page (which the AMP bar did a great job of hiding) and share that.
But what is most annoying of all is that smart people at Google were kidding themselves that their botched up implementation of AMP was a great thing to have launched upon the world. It really was not. I bet it didn't really do much for their bottom line either.
Add that to how untrustworthy the link looks, it's just terrible execution on an idea that might have merit.
Disc: Googler.
I usually just go to google.com in Chrome to search for things, and the amp link in the address bar is really rather confusing.
But why is using the Google AMP service through the Google widget so terrible? I would expect a more fluid experience, but you end up with this really terrible experience trying to leverage the Google ecosystem. Theres nothing from stopping me from opening FireFox and searching the article title in Bing or DDG at that point.
> Both sites are measurably worse on their AMP versions. The amount of data transferred is higher, the script execution is worse, and the page takes longer to load. So much for the promise of speed and a better user experience.
AMP pages render instantly because Google is able to pre-cache their data. So you're not paying the same price at click as you do when you go to a non-amp page. Yes, AMP doesn't make sense as a tech to render your main site when a user directly visits it, but for embedding your content into another's experience, it does allow for very fast load times.
> they didn’t go to BBC’s website; they view a copy of the BBC’s content on google.com
That's exactly what a CDN is. The only difference between AMP cache and CloudFlare is that the latter uses some DNS trickery to show BBC's domain name, whereas Google can't. Google solved this in Chrome, and there are other ways to solve this with Web Packages [1] and Signed HTTP exchanges which allows the content to be served from Google but with the original url, without DNS changes.
[1] https://blog.amp.dev/2018/05/08/a-first-look-at-using-web-pa...
Doing a direct comparison is kind of missing the point.
Hackers / developers seem to insist on the non-AMP sites to bloat sites to absolute trash and back. If I saw the amp logo, I knew that was the one to click.
Abusing users with bloat is supposedly some great freedom worth fighting for, but AMP forcing these developers into a much smaller box that they can't screw with users as much on I just loved AMP.
I think a lot of the HN hate comes because AMP I think prohibits random third party javascript ad libraries. So obviously it hits folks where it hurts, but the web is much nicer to browse without all the junk.
The thing people were offended by wasn't the lack of bloat, it was the fact that it was ceding even more control to Google.
Billing on line is the same thing. Unless you are on a trusted site (Amazon / Apple) you can just get hosed trying to cancel subscriptions with jack up pricing etc.
Same thing with spam - at some point users stopped trusting the hackers and developers and migrated to more controlled worlds (ie, gmail etc) which blocks HUGE quantities of the junk.
I just wish hackers spent some time trying to make users experience and lives better rather than the constant push to spam them (EFF wanted to pass laws banning anti-spam measures), now AMP (which was amazing) is being attacked endlessly and the list goes on.
Google has a commercial reason for pushing this stuff - the web was turning into a trash bit (flyover ads with audio etc etc).
Besides, does blocking ads and trackers not at least partially alleviate the issue? Works well for me and, in combination with a reader view, is a significantly better experience than AMP. I'd argue that those "hackers" you very much hate are the exact ones fighting against corporate bloat on the web.
I'm blocking a good percentage of ads in Safari on my iPhone via third-party apps Hush[0] and Magic Lasso[1]. I typically read articles in Reader mode. Because of this, I can't give a great accounting of how many ads actually get through. Certainly, the number is small enough that I don't think about it.
[0] https://apps.apple.com/us/app/hush-nag-blocker/id1544743900 [1] https://apps.apple.com/us/app/ad-blocker-by-magic-lasso/id12...
I completely understand why open web advocates would not greet AMP with a warm welcome. That doesn't mean it wasn't necessary to fix the problems. The problems were and are real.
Google could just make page speed more important in their rankings. They probably don't because page speed isn't as important as content quality in regards to what people are looking for in search.
I still struggle to see how AMP has anything to do with improving the search experience and everything to do with Google abusing their market dominance.
So a ton of the stuff that HN folks demand (being able to do dynamic third party javacript ads and tracker loads, DOM modification after render - gone - you just are totally blocked.
If you try targeting AMP (or just consuming AMP) you'll pretty quickly get what they are going for.
For example - AMP websites really do not reflow that often (or at all). I've never I don't think had something move when I was going to click it even early in page load.
I've found a lot of complaints about AMP are developed / hacker focused complaints. If you switch your mindset to users, including non-tech types, the value becomes more obvious.
I think you might be strawmanning why some HN folks don't like AMP. To me, at least, it looks like someone abusing their position as the dominant search engine.
AMP lightning icons didn't get attached to pages with just one tracker and fast renders. They were attached to pages served by Google that only supported Google tracking.
There was a win with the AMP user experience. The fact that it had to come from google is embarrassing.
You can tell people don't use AMP in the real world with mainstream websites when they say this.
Most people I've seen who I would consider hackers have web sites that don't do all the things you're complaining about.
I hate the "throw on another library, and stack some ads" attitude as much as anyone else, but AMP isn't the way. I would be less offended if Google had just down-ranked pages with heavy JS load, and boosted pages that didn't move content around.
And frankly, if the argument is that news websites should be text only, then maybe Google shouldn’t have killed off Google Reader, because that was a much faster way of loading news articles.
HIGHLY predictable behavior - some users like that. The apple walled garden, the apple billing flow, the AMP subset of HTML etc - all hated here.
Too many users have been burned by developers flying interstitial ads over the content (still happens) with a tiny unhittable x in a corner that comes in 10 seconds later. It's not the size of the content (that was already ranked), it's the absolute crap that hackers and developers shove down users throats.
I get it, you don't see or chose not to see the value in this. But there is value, at least for me and I'm sure many others started learning that the AMP icons meant less crap.
The number one user first request was "let users turn off receiving AMP pages." It was never implemented. The AMP advisory board item on it just... disappeared, without explanation or resolution.
AMP was not user first in any way.
> If you try targeting AMP (or just consuming AMP) you'll pretty quickly get what they are going for.
So if I go to Reddit, then the AMP site is basically unusable and pollutes search results. The AMP site removes all the useful comment information in place of application ads. Multiple clicks are required to restore the content.
The Guardian also strips comment content from it's AMP pages.
That's not uncommon.
If your marketing department want one more tracking script and your execs want one more pop up banner etc and each change is bad but not egregiously so, then your average "web guy" at most non-big-tech companies will just do what they're asked to.
With AMP they get to say "it's not possible"
People might have had more room to negotiate on how to hit a performance target but the thing which would have mattered was a number in the webmasters tools console.
[1] Stats for the US: https://www.emarketer.com/content/the-majority-of-americans-... (I wish I could find stats for India or other growing economies)
I get some happiness about it but hopefully we can have nice things without such massive sacrifices.
From a pure performance perspective AMP had one secret weapon: Google could safetly prerender content before the user clicked on it. This was offset by the fact that AMP made lots of other page speed optimisations impossible.
Would also improve the web dramatically.
https://chrome.google.com/webstore/detail/js-toggle/bnhjfamo...
Social logins are user-initiated. That's fine.
What's not fine is that as soon as you visit the site, a giant in-frame "GIVE US YOUR E-MAIL ADDRESS" newsletter popup shows up, covering the content.
Instead of going that route, Google could have said on day 1 that they would just rank based on the site's experience score (as they are doing now), and the end result would be the exact same.
As a user I just expect things to be fast and working out of the box.
You know what the biggest ads and tracking provider is on the internet? Google DFP and Google Analytics.
AMP is a solution to a problem of their own making, a clever solution that gives them even more control over search results and data collection (since it never leaves Google domains) while pressuring publishers to spend limited dev resources on a limited subset of viewers while making no impact on the existing website and apps they also run.
There's nothing good about AMP beyond technical experimentation. If incentives are a problem, then fix the incentives.
It's ads/tracking with their complex network payloads and rendering that create the slowdown.
However, it is impossible to beat AMP’s speed when doing something like loading an article from a Google search result.
The AMP spec makes it possible for a search engine to preload an article safely, and as a result, the article can be shown instantly when the user clicks on it.
Using the AMP libraries has never been the issue. Google hijacking the URL's has.
For some reason I almost always want to read them, and most of the time the AMP version does not have them, and most of the time there is no way to get to the non-AMP version.
Example from my Google feed: https://www-tomshardware-com.cdn.ampproject.org/v/s/www.toms...
>>Part of the Google update is that all pages with high Page Experience scores are eligible to be in the featured top news carousel. This effectively means that publishers will no longer be forced to use AMP and can instead provide fast, rich experiences on their own domains
Disc: Googler but not associated with AMP whatsoever.
All pages with high Page Experience scores are eligible to be in the featured top news carousel.
This effectively means that publishers will no longer be forced to use AMP and can instead provide fast, rich experiences on their own domains.
His call to action is pretty straightforward:
"Get to work right now to ensure that your site has a great user experience AND has a great user experience score. Avoid unnecessary JavaScript, plugins and bloat, and make your site easy-to-use". https://www.lafoo.com/javascript-is-driving-your-visitors-cr...
Great joke. Visit any non-amp news site and tell me that one again.
The only, major benefit AMP brought to the internet is that it was legitimately light and fast.
It's absurd, how could regulators and authorities let this happen? Let's also mention how hard and counterintuitive is getting to the original site from the amp version.
I stopped using Chrome as my primary browser and moved to Firefox for exactly same reason - AMP.
because they have been asleep at the wheel for almost two decades now, letting tech companies privatize and eat away at the open nature of the internet, so now we have to live in tech giant fiefdoms where they can determine the very form and content of everything displayed on the web because they're the ones who control every access point and standard.
Google was saying they were trying to solve a real problem, and it isn't easy (for me at least) to have known their true goals. Now if you read the the Texas anti trust suit -- and I recommend it to anyone who is interested-- it seems pretty clear that their motive for AMP was to improve their ad revenue.
It can be both. AMP legitimately sped up a lot of crappy news sites. That put pressure on non-AMP sites to get their game together. It was probably a net positive at first. But then Google abused their position.
Absolute market dominance corrupts absolutely. Even, and in spite of, efforts to the contrary from within the company.
Lobbying.
The AMP mess is not the death knell of the open internet, perhaps the sick idea of a rotting company
The minute these publishers weren't getting as much search/news traffic anymore, they would have capitulated, making their sites run faster immediately, the same way they quickly capitulated to AMP.
As proven by AMP, when Google says jump, the entire interest responds by saying "How high?"
Can anybody give me a single non-nefarious reason for why AMP was necessary instead of simply enforcing page weight/loading-speed factors via the algorithm?
Your marketing department wants another tracking script? no can do. Your exec wants a javascript "click the monkey" promo? not on the AMP site. The UX folks want parallax webGL spinning logos? Sorry AMP doesn't allow it.
It puts people who were previously not in a position where they could push back against decisions which effect load speed in a position where they can push back.
The tech team could have just as easily have said, "sorry, that would put us over our X ms page speed limit."
The fact Google spent 100X the resources of a simple algorithm update by creating AMP, tells us there were other motives behind this.
They actually did this a few years ago: https://developers.google.com/search/blog/2018/01/using-page...
Wooow slow down buddy. Nothing like that is hapening. I can still hook up my computer to the internet and start serving websites right away. Google can't do anything about that. Neither can Facebook, or any GAFAM.
However Google did change your behavior is entirely on you. You can live without them just fine. Until that is changed I can't see any need for regulation.
Republicans want deregulation everywhere, no matter the cost, the second you talk about limiting big business or monopoly busting they run away.
Democrats on the other hand rely heavily on favors from the tech establishment to keep their agendas front and center and hide any downsides, this means letting BigTech(tm) have free reign.
Thus both parties are incentivized to look the other way. Only recently did the Republicans wake up to TOO BIG = bad; but their solutions are all terrible and thankfully were never entertained by any meaningful majority. Meanwhile the democrats are working very hard to cement the current big tech conglomerates into having control and power so you can expect more of the same.
Unless big business is de-platforming right-wing content, in which case this principle is instantly forgotten and we need regulation and monopolies need to be broken up...
Covered that here
Gop politicians (for Republicans call themselves this) don't care about deregulation anymore -- they just want to funnel money and power to the wealthy.
Dem politicians now seem very excited about breaking up Facebook and Google.
Both parties want to hurt Google and Facebook for perceived political bias. Gop politicians want this because Google and Facebook sometimes allow some left-of-center voices to occasionally penetrate their newsfeeds. Dem politicians want this because Facebook has been a vector for QAnon and other wingnut-terrorist radicalization.
It almost makes one wonder if that's what deregulation was for all along....
Is GOP, it's an acronym for Grand Old Party.
But I was just describing the current climate, the Repubs, never cared about dereg until recently when they realized that FB and Google are abusing their power, but all the GOP solutions involved draconian backwards thinking that would only make things worse.
> Both parties want to hurt Google and Facebook for perceived political bias.
Not quite, GOP wants to hurt 'em, Dems are very happy with them, and all their "dereg" is basically aimed at giving them more power.
Rather than link to those campaign promises, I'll point out that the current legislative strategy for restoring competition to the tech sector is being spearheaded by Democratic Representative David Cicilline. [0]
Additionally, Professor Tim Wu's appointment to the National Economic Council by the Biden administration speaks volumes of the urgency for antitrust action on the part of the Democratic party. [1]
[0] https://www.reuters.com/article/us-usa-big-tech-antitrust/u-...
[1] https://www.nytimes.com/2021/03/05/technology/tim-wu-white-h...
It's much easier to get companies to move your way if you threaten them with broad investigations into their activities. They're much more likely to behave the way you want later.
That doesn't change the real danger these companies represent to the continued freedoms and viability of society, given the current social media feedback loop and the death of journalism as a result of tech.
We're in uncharted territory.
It's also much easier to enforce antitrust by initiating antitrust legislation.
Currently, the strategy of broad investigations is yielding to multiple focused pieces of legislation spearheaded by David Cicilline (link in GP).
Sure, but I don't think thats the case.
>Currently, the strategy of broad investigations is yielding to multiple focused pieces of legislation spearheaded by David Cicilline (link in GP).
I think you have it backwards.
From what I can see David Cicilline will go after social media for pushing 'misinformation' and call for even harsher corporate censorship in the name of information security, and will likely want to alter section 230 to make it easier to punish them for algorithmic 'misinformation' rather than to push more freedom for normal people.
Don't expect more freedoms, expect more control. They aren't going after these companies in your defense, they're doing it to control them and use their power against their enemies. They'll use misinformation and election security and whatever else just as a cover for making things worse but being in control.
This is unnecessary and also incorrect comparison
>Some Dems are corporate-friendly and want contributions from lobbyists; others are social democrats who are distressed about corporate dominance.
True, but not relevant. There's a reason Warren stayed in the primaries long enough to make sure Bernie couldn't proceed - and Kamala Harris who didn't go anywhere is now Vice President. There are functional powerblocks that are making decisions, whether the party is unified or not.
>So it depends on who you're talking about: AOC and Joe Manchin aren't going to have the same perspective.
That has nothing to do with what the effective leadership and powerblocks are doing, and both can be used to push a position that does what the powerblock wants by propandazing it in their perspective.
Tech regulation is good! AOC can push it because it will protect against misinformation, protect minorities against hate, and guard against russian intrusion in elections similar to what allowed Trump to be elected. See how easy it is? It's just spin. spin is not what the actual politics going on are.
I wouldn't say Cicilline is spearheading anything. However, it's good that after 26 years as a politician, Cicilline now "is preparing to come out with 10 or more pieces of legislation targeting Big Tech companies" [0]
But, don't forget that Ciccilline went to prison after he "pleaded guilty to conspiracy, obstruction of justice and making false statements for his role in the courthouse corruption scheme" [1]
Regarding fiscal competance, Fitch Ratings also downgraded Providence's ratings, and Cicilline was accused of "hiding the scope of the city's fiscal woes through 'illusory revenues, borrowing and other tricks.'"
Cicilline may do something positive today, but there are better people to be the "face of anti-trust." I'd still be wary, due to his past criminal activities, that he isn't continuing to engage in the same practices one more time.
[0] https://www.reuters.com/article/us-usa-big-tech-antitrust/u-...
[1] https://en.wikipedia.org/wiki/David_Cicilline#Controversies
This is mostly about forcing BigTech(tm) to censor more free speech and calling that legislation "breaking up big tech" it's anything but.
I will admit that Warren has been beating the monopoly busting drum for 10 years now, unfortunately she's kinda made herself irrelevant with her other policies and claims.
The failure here is more that regulation covering this stuff is very unclear. Is it anti-trust? Telecommunications?
And if it is antitrust, what is anti competitive? How is the consumer harmed? How are others being prevented from competing?
Basically you can't just say "Google bad" here. Regulators need to go from that to "Google bad, and illegal." That last thing is not clear right now
Also - has anyone ever been happy to land on an AMP page, and think, "wow, this is really useful, I'm glad I'm on this weird page, instead of the original one" ?
I tend to click on AMP results in searches because they always load so much faster.
I don't have a strong opinion either way, but I do think that the stated purpose of having an easily cacheable and loadable webpage is a good one.
Reddit is already fast, Reddit does not benefit from AMP - except by putting that "This page looks better in the app!" modal back in the face of people that have otherwise disabled it via their user settings.
If the AMP page is showing everyone the same thing, then of course I won't see Reddit with my user settings (old mobileweb, do not ask me about the app, etc), and of course trying to interact with the page will redirect me somewhere else. If I could opt out of AMP to avoid this issue, I would.
Especially in the early days I was really happy about the AMP pages, because the publishers haven't yet managed to cram so much garbage into them as they have into their regular pages. I don't need e.g. a sticky video player playing some irrelevant video just to create space for a video ad.
All of these functions are already served by my adblocker, which does not hijack the URL to do so.
Tech companies were ridiculously popular until recently.
This is no longer accurate. Tech companies as a whole have seen a sharp decline in net favorability since 2015 [1]. Within the cohort of tech companies, Google is favourably views [2]. But even then, there is thin popular support for breaking it up.
[1] https://www.pewresearch.org/fact-tank/2019/07/29/americans-h...
[2] https://www.theverge.com/2020/3/2/21144680/verge-tech-survey...
Google, and tech, is still popular. It's just not ridiculously, untouchably popular.
When it comes to Teflon in politics, what matters is (a) having a vocal minority who love you and (b) not having a vocal minority who hate you. The former protect you. The latter attack. Google went from having everyone on their side, strongly, to having most people on their side and a vocal minority against. That's the difference between a Congressional investigation or DOJ complaint gaining traction and getting sniped.
Also, nitpick, universities were expressed as having a positive impact by two thirds of Democrats and one third of Republicans in 2019; the comparable numbers were 54 and 44% for tech companies. In 2010 it was 71% and 66%. You don't lose twenty percent of the country costlessly, even if people start disliking other stuff a bit more, too.
What's the benefit of this (except caching = faster page loads and navigation) for Google? To my knowledge, the ad revenue goes to the publisher, and any other company that wanted to serve the content under their domain could do so.
b) can they not do that on the publisher domain too, because one of the 300 trackers that every publisher embeds tends to be Google Analytics?
Sadly it is at this point that Google will also decide that this neutral and open technology is not actually worth pursuing for the good of humanity after all, and will remove support for it from Chrome.
This is a pity, as it would be nice if sites could sign their content with an offline key, to avoid the risks of compromised web hosts. Even better would be if browsers let you pin a version of a web app and only accept signed upgrades if you opted-in, to further protect against hacked or coerced use of the signing key.
“ In one given scenario, if an attacker is able to obtain the private key used to sign a particular website for Signed Exchanges, that attacker can then deliver false websites that appear in almost every way to be legitimate. Considering this false page in no way connects to the valid server, it would be difficult for admins to even detect.
Additionally, Mozilla has publicly shared its stance that Signed Exchanges are “harmful” and will not be implemented in Firefox as currently specified. In particular, they take issue with a browser serving a particular website without connecting to its original source (or “authoritative server”).
Mozilla has concerns about the shift in the web security model required for handling web-packaged information. Specifically, the ability for an origin to act on behalf of another without a client ever contacting the authoritative server is worrisome, as is the removal of a guarantee of confidentiality from the web security model (the host serving the web package has access to plain text). We recognise that the use cases satisfied by web packaging are useful, and would be likely to support an approach that enabled such use cases so long as the foregoing concerns could be addressed.
A leader of Apple’s Safari development team, Maciej Stachowiak, recently backed Mozilla’s stance, adding an almost humorously phrased explanation of what Signed Exchanges for AMP really is.
But even so, I’d say we are pretty uncomfortable with this approach, for similar reasons to Mozilla. We can see some advantages to Google re-serving the whole web from their own servers and getting browsers to present it as if it comes from the origin, but it also seems like a worrisome change to the web security model. ”
https://9to5google.com/2019/04/18/apple-mozilla-google-amp-s...
for those of you who want to your support the middleman: http://www.google.com/amp/s/9to5google.com/2019/04/18/apple-...
Firstly, there are many use-cases where the content signer does not care whether an attacker has managed to steal their private key. For example, sites that just host static content (like cat pictures). There are even some hosts who don't see the point in TLS at all, as they don't believe they are hosting anything that an attacker would want to interfere with.
Secondly, this is an opt-in feature and not something that would apply to all TLS certificates, so entities who considered it a genuine risk could just not use the technology. (They might also have to keep an eye on certificate transparency logs to make sure that an attacker hasn't requested such a certificate for their domain by controlling their DNS, but hosts should be doing these checks anyway, and if an attacker is in control of your DNS and has received a TLS certificate for your domain then you have bigger problems than Signed Exchanges. In any case, such certificates can be revoked once the legitimate owner has control over their domain again).
Thirdly, hosts which do opt in to using Signed Exchanges can choose to include code in their signed content which sends a background HTTP request to their servers, as a way to "phone home". That wouldn't be able to detect when a key had been stolen, and it probably wouldn't reveal where the content had actually been served from (unless the specification were altered to allow signers to require that) but it would at least give signers visibility of which content was still being served.
Finally, the biggest risk to websites is not that "the host serving the web package has access to plain text", because these are publicly served files which everyone has access to. The biggest risk is phishing, where a victim believes they are visiting a certain site but they're actually visiting a site with a similar-looking name, which TLS doesn't prevent and Signed Exchanges doesn't make any easier.
Then they started cramming more and more of that into the AMP version, despite AMP making that hard.
Giving better placement for the main page version won't be as effective, because it doesn't force them to start from scratch. At best, we'll see some incremental improvements.
I've seen an ad that managed to consume so many resources that Chrome blocked it. Reducing the threshold for that blocking would be much more effective.
Good to see it being out in the open and included in the lawsuit no less. I certainly share the sentiment of the author and hope this is the start of the end for AMP.
But maybe that's the point?
Besides AMP pages break horizontal scrolling in some page as well. For example: https://www.google.ca/amp/s/ca.finance.yahoo.com/amphtml/new... You can scroll left and right but it is not supposed to.
Some app will also render wrong thumbnail for AMP links as well.
Altogether AMP makes the UX much worse :-/
The fact that they didn't do that, tells us that Google is either:
A) Stupid and inefficient, because creating AMP cost 100X more than a simple search algorithm change
Or
B) Trying to achieve something more nefarious with AMP, something that is worth the extra cost, and that they weren't telling us about.
After seeing the Texas lawsuit against them, it's pretty clear to my eyes that the answer is B.
As for why they didn't just "dial it up to 11", I'm guessing that would cause the quality of their search results to go down.
How would announcing preferential treatment for sites under X kb in weight have been any different than preferential treatment for sites using AMP?
Both would have required similar action from the dev teams at these publishers.
In addition, the carousel was a much more desirable placement for publishers to chase. It made them pay attention more than a small ranking factor bump would have.
I do however disagree with Google's decision to implement this. I feel it's one of the legitimate criticisms of AMP, and it took them far too long to fix it.
I could agree with the performance sentiment if it were directed at the real issue:
One-click-and-you-now-have-amp solutions, which is like 99% of AMP pages using WordPress.
I don't see how this gets mixed up with AMP.
Without Google, AMP alone doesn't have the speed or the technical advantage to make it worth it. Google's neatest trick is their attempt at framing AMP as something akin to an open-source project that has nothing to do with them when, in reality, it has everything to do with them.
P.S. AMP does give engineers an out by allowing them to build pages from scratch without the cruft imposed upon the OG pages by their employers, but this could be easily avoided by product owners making better(and rather obvious) decisions rather than always making the more exciting ones that cause a lot of dilly dallying and technical debt. Still, at the end of the day, on its own is just busy-work.
And that's exactly why I can't hate AMP. Publishers don't give a fuck about performance, and user experience in general as long as they get clicks that can be monetized. They don't want to keep users for their quality content, they want to game search engine results and trap users when they get in.
Things like AMP seems to be the only way to make publishers care. Seen the other way, Google have seen the mess, and took advantage of it and made AMP, which both addresses the problem and get them a comfortable position. If sites were fast the problem wouldn't be there in the first place, and Google would have had to find something else for their monopoly.
And sure, what you said in your PS is true. Unfortunately, that's not how things are. I don't know why, greed maybe? But if it is the case, it is the stupid kind of greed, because now, publishers are struggling and Google is making shittons of money.
AMP is the symptom of the problem, which is Google and its monopoly on search.