I don’t even understand how AMP is legal or how they could possibly argue it gives any benefit to the user. Every single AMP page i’ve used has broken CSS and often broken content.<p>The speed boost is negligible if it even exists and all it serves to do is add an additional stupid pop up I have to click out of to read a web page.<p>I really wish I could switch to DDG but queries related to anything technical, like biology or programming, usually fail to turn up any relevant results.
Another take - and why I like AMP.<p>Web developers deliver crap, ad laden slow websites - with HORRIBLE users experiences.<p>Content jumps around as the web page loads, sometimes even 5 - 10 seconds later!<p>Autoplaying content is hell - but they seem to love it.<p>Popovers, unders and sliders gallore.<p>And for some reasons the GIANT mess of analytics trackers and CSS and dynamic content means the thing is<p>a) slow
b) nightmare on a phone<p>Google has basically FORCED a bunch of these idiots to actually make a page usable and pleasant. It loads fast.<p>What I don't get - all these high and mighty web devs can't seem to deliver clean websites that work without Google beating them over the heads with AMP. Is it that they cower in front of management with demands for YET ANOTHER tracker (seriously, websites ship with like 10 trackers - why?) They feel the need to show off their lazy loading dynamic content skills that jank the page?<p>So thank goodness for AMP, perhaps developers will actually try to compete with it a bit by building usable non-amp websites.<p>And trust me, users are getting trained that the lighting bolt actually does mean FAST.<p>Anyways - when I read another rant from a web developer about AMP - a small request. How about de-jankifying and de-scamifying the web first a bit more? Then we won't need to turn to google for all this stuff.
The answer is Core Web Vitals.<p>I have spent countless hours doing real optimizations on a website with real traffic, without submitting to AMP (which I view as a disgusting move by Google and everyone involved with it.) (Traffic which, by the way, does not reflect the device profiles of what Google considers the "average" user, based on real vs. lab results in Lighthouse -- which is forcing us to work on issues that are not proportionally relevant to our business, though I will concede it's a positive improvement for us overall. But still an unwanted Google influence, like most SEO.. but moreso.)<p>Core Web Vitals should render AMP irrelevant, and thus seeing as both projects are being pushed by Google, it's time Google takes AMP behind the barn. Unfortunately Core Web Vitals takes a pretty hard stance against bleeding edge technology like (Vue/React) server side rendering with client hydration. Anything beyond a todo app starts to see considerable main thread time during hydration which obliterates the Core Web Vitals scores. I predict with continued focus on CWV we will see: much greater focus on startup times for client side apps (including better hydration strategies), and maybe even some server-side only JS front end frameworks -- more aligned with the JAMstack idea (everything old is new again, yay.)<p>As much grief as CWV has caused me, it is the correct solution to the problem of slow websites and its impending inclusion in Google's page rankings should have a positive impact on the overall health of the web.<p>Why people who aren't being paid by Google continue to defend AMP absolutely baffles me.
Thought experiment: Divorce AMP from Google. Google withdraws from being the AMP standards author and "prototype" AMP cache provider. The project becomes truly non-commercial and is handed over to a non-profit that users trust, let's say, hypothetically, the Internet Archive. The IA adopts as AMP's goal: making web pages less expensive to crawl (ideally, by parties other than Google) as well as making pages faster on mobile. In addition, the AMP standard is revised to require that AMP pages must allow equal access by all clients, whether "browsers", "bots" or otherwise. No preferential treatment for certain browsers, e.g., Chrome, or certain search engines, e.g., Googlebot.<p>Bias disclosure: I use a text-only browser and AMP pages look great in links. For a links user, the AMP version can be useful on some sites that have a large amount of cruft, e.g., excessive number of same site URLs, at the top of the page, with the content buried below it, and yet more cruft at the bottom. AMP eliminates the necessary scrolling on such sites.
Before I even knew what AMP was, I saw the lightning bolt next to search results and I quickly realized that the lightning bolt mean that the site would load fast on my phone. I loved it. So many modern sites are hostile to usability, that part of me is sad to see AMP being attacked. I don't want Google to own the web, but I don't trust that web developers can stand up for sane usability practices anymore.
It’s easy to say “Ignore Amp” when you’re not a content site that depends on Google Search to stay alive. Sadly, for a large amount of sites on the web, what Google says is what goes. Chasing a #1 keyword ranking puts food on people’s tables. It’s not always as easy as “this thing is bad: stop using it!”<p>There will need to be a bigger driving force to get amp out of popularity. As long as AMP pages unlock preferential treatment in search results (mobile carousels), sites that want to compete will be forced to use them.
> In 2018, my recommendation was to avoid AMP, to use AMP for the most relevant pages, or to use AMP only.<p>What an excellent, mediocre, and/or bad recommendation!
I switched to DuckDuckGo on mobile a while back because it meant that I didn’t have to see constant AMP results. Even though I think DuckDuckGo’s search results aren’t nearly as good I have no intention of going back.
Why AMP exists?<p>A website I consult in a popular sports niche and has a slow, broken, ad-infested main website grew its traffic 500% with AMP.<p>On the main website, it's still broken. On AMP it's... AMP. So Google thinks it's fast enough/good enough.<p>On AMP we implemented a lot of annoying CTAs to go to the main website. ”Read Full Article here” ”Read more” ”Details at...”<p>In the past this website would have needed to optimize its real website to gain this much visibility in the search engines. Now they just AMP, then they optimize their AMP to real-world-website CTR, and can continue to have a... sub ideal.... website.<p>AMP is whitelisted cloaking for slow websites. And a burden on webmasters and developers.<p>I always say AMP is the internet if germans (most AMP leads were at one point germans) would have invented it (I am Austrian, we always joke about Germans ): Efficient, mostly boring and long term innovation harmful.<p>I am rooting for AMP to die. Sadly it will still be around for about 5 years until the ”what a great journey” blogpost.
AMP is an anti-competitive abortion that should just die.<p>It's Google throwing its weight to force websites into dubious practices all in favor of an alleged performance. Users also get the short end by being served low quality pages instead of the full experiences they expect.<p>I have this extension [1] to make sure I never visit an AMP page again.<p>[1] <a href="https://addons.mozilla.org/en-US/firefox/addon/amp2html/" rel="nofollow">https://addons.mozilla.org/en-US/firefox/addon/amp2html/</a>
Oh, besides the back-end technical mess that is AMP, just from the standpoint of a user, one can encounter catastrophic faults with its design:<p>- can't share articles using android's builtin share widgets because it points to the amp pages<p>- Navigating away using "Open in $BROWSER" option from the Google app in Android opens up the amp page again instead of the source page.<p>- can't see embedded article widgets like tweet blocks, maps, overlays, animations<p>- Attempting to do things like Comment on an article triggers navigation away from the AMP page to the actual site, forcing you to then scroll down once more<p>I've become accustomed to opening AMP pages and looking for "View article on actual site" link as a matter of course. It's just so horrible.
Interesting he mentions both Baidu and Yandex have competing products to AMP:<p><a href="https://www.mipengine.org/" rel="nofollow">https://www.mipengine.org/</a><p><a href="https://yandex.com/dev/turbo/doc/concepts/index.html/" rel="nofollow">https://yandex.com/dev/turbo/doc/concepts/index.html/</a><p>They seem to work on similar principles.<p>China and Russia will still have this issue as well.
I looked into AMP while working on Post Office Map (a.k.a. post office near me) [1], after a lot of looking into structured data and something called JSON-LD [2] which I'd never heard of, ultimately I read that preparing my site for AMP might actually _break_ my site, and there is no guarantee of any benefit. A lot of people are using AMP for SEO, but I suspect that's a fool's game - Google is smart enough to know that just being AMP-ready doesn't mean anything for relevance.<p>[1] <a href="https://www.postofficemap.com/" rel="nofollow">https://www.postofficemap.com/</a><p>[2] <a href="https://json-ld.org/" rel="nofollow">https://json-ld.org/</a>
I can pick an AMP site out almost immediately. I click on a link somewhere, and a moment later as I start to the read the article I'm like: "this looks weird, like a crappy AMP site not a regular website!" I inspect the URL and sure enough it is. I always switch to the real site at the earliest opportunity. So not only is AMP terrible regarding the open web as others have commented, it's also freaking annoying on a pure UX level.
AMP is here to stay. Google shows a special lightning bolt Mark on amp pages and prioritizes those results. As long as google favors Amp, it makes business sense to use it.<p>Google needs an anti-trust slap for Amp. Until it gets it, it’s here to stay.
3 days ago on a different thread I asked HN for a way to avoid AMP. [ <a href="https://news.ycombinator.com/item?id=25467112" rel="nofollow">https://news.ycombinator.com/item?id=25467112</a> ]<p>User coldpie recommended [ <a href="https://news.ycombinator.com/item?id=25467438" rel="nofollow">https://news.ycombinator.com/item?id=25467438</a> ] the firefox extension <a href="https://addons.mozilla.org/en-US/firefox/addon/amp2html/" rel="nofollow">https://addons.mozilla.org/en-US/firefox/addon/amp2html/</a>
One thing i find irritating about AMP pages is that, even Google chrome's page translation feature doesn't work unless you navigate to the non-AMP page
Google could use dns-prefetch to tell the browser to to pre-fetch the DNS records for the domains in the top search results... That would decrease page load time up to one second! (the caching being the main advantage of AMP - it would make AMP unnecessary) Really, why are not browsers pre-fetching DNS records automatically for all links visible on the screen!?
I get 99% of my news from sites like these<p>text.npr.org and lite.cnn.com<p>Just plain text, if I want more, then I go look at the ad revenue site and I'm good to go. More news should be available to the public. And since you asked, I do send an annual donation to NPR to cover the usage of the text site.
Wildcard blocking the following domains will break/disable AMP pretty much everywhere, leaving only the header bar (from which you can click to the original page):<p>- google.com/amp<p>- ampproject.org<p>I've been including AMP in my blocklist for quite a while now, and while I've occasionally felt like I'm tilting at windmills, it's honestly not much more inconvenient.<p>Before blocking AMP, I would get confused for a few seconds by a broken page that looked like a real page. Now I just see an empty page immediately, prompting me to get to the real page more quickly.