1. 53
  1. 14

    tbh i’m surprised if anyone really believes Google’s intentions with AMP are anything less than “control the entire web”

    1. 6

      Does anyone not think that? I though everyone knows that but wants to get higher rankings on search.

      1. 3

        I don’t know for sure. I get the impression from articles like this that web-based businesses don’t see how easily they’re playing into Google’s hand.

        1. 5

          I think its more that they dont care. In the short term getting higher rankings means more than protecting a free and healthy web.

          1. 5

            There’s more to this than apathy. It’s a classic collective action problem.

            Think like a business person; suppose your competitor gets on AMP and their page loads faster and (maybe) ranks higher in Google search. If their advantage is enough that you go broke in the short term, there will be no long term for your business.

          2. 2

            The biggest concern for is was that AMP lets Google host your content, which makes it difficult to understand what users are doing :(

            The dynamic changes quite a bit with the combination of scripts being restricted AND content not being hosted on your servers. Measuring traffic is tough.

          3. 2

            Honestly, I’ve seen where an AMP implementation doesnt even help that much with ranking. I think people using it are really on about the mobile performance

            1. 4

              it Helps a ton with ranking if you get into Google News rankings, but otherwise only a small bit. -.-

          4. 3

            I think the AMP team have good intentions. They allow you to submit scripts to their repo for your projects.

            I do think that them requiring you hosting it on their servers is a requirement that came from another team to use the requests to track users, though.

            AMP could work fine self-hosted if it wasn’t for that.

            1. 2

              What is the point of self hosting amp? You could just not add the scripts that amp blocks and your website is now just as fast as an amp page.

              I dont understand why we are celebrating this new tech that is essentially Google banning you from shooting yourself in the foot. Self hosting is just not shooting yourself in the foot.

              1. 2

                You can’t self-host AMP. That’s the problem I’m mentioning.

                Nobody is celebrating AMP, as it is poorly executed. People implement it as an SEO measure, and it’s honestly potentially harmful depending on your business goals. Sometimes being on Google News isn’t worth the trade-offs.

                That said, there are enough developers out there who will take efficient frameworks and produce inefficient, janky, buggy code. AMP honestly is a solution to this particular problem.

                I just wish people would be more disciplined in their coding before it got this bad. Then people wouldn’t have to have resorted to disabling their ability to write JavaScript without a third party reviewing it. Since businesses don’t prioritize quality before it’s conveniently “too late” for them to revisit, this is essentially the only way to give performance guarantees at this point with the web.

                Not defending AMP, of course, but only explaining the benefits that some people see with it. I personally would rather people just implement proper continuous unit, acceptance, and integration testing with performance as automated necessary-to-meet criteria - but that’s just me.

          5. 7

            The amp solution is pretty ridiculous. Makes your web pages faster by not allowing you to add scripts slowing it down. Or you could just not make a slow website to begin with.

            1. 8

              The irony is even better.

              With a content blocker and a reasonably extensive list, including the Google-hosted AMP JS URL, the ‘regular’ URL for sites loads quickly, while the AMP pages suffer a 10-15 second white page delay.

              1. 1

                As someone who has implemented AMP on a site with a huge amount of users, I can tell you that whomever implemented it with a 10 second white page is doing it wrong.

                Time to first interactive frame should never exceed 1200ms.

                1. 3

                  The people who implemented it wrong are the amp people.

                      body {animation:-amp-start 8s steps(1,end) 0s 1 normal both}
                      @keyframes -amp-start{from{visibility:hidden} to{visibility:visible}}
                  1. 1

                    Is this new or is it doing something that I’m missing? What is this?!

                      1. 3

                        That seems like a literal dark pattern to try to get people to turn on JavaScript when it’s off… Like, it seems like they want users w/out JavaScript to think JavaScript makes things faster. That’s new to AMP since I last used it, but seems terrible :(

                  2. 1

                    https://www.ampproject.org behaves exactly the same way. Block any script resources from cdn.ampproject.org and you get a nice delay.. while it does nothing.

              2. 4

                Avg first meaningful paint of this person’s clients webpages is 4 seconds.