1. 44

  2. 11

    I don’t like amp, but a page speed score of 80 to 86 isn’t really what I’d consider evidence that it’s slower. It’s also very easy to build a slower site with a higher score.

    1. 1

      Author here, you’re right about the evidence.

      I added quite a lot more in the article, from page speed insights but also from a far more in depth analysis site recommended by a (former) Google SRE.

      To summarise, AMP in this case is consistently slower, sometimes considerably.

      Here’s the info: From page speed insights (the numbers are worse for AMP after running it a bunch more time) I ran it until I got bored of running it, alternating between amp/non-amp, chrome on Windows 10 (normally I’m Chromium on Fedora, but let’s try mainstream)

      AMP: 75 90 91 91 91 89 80 91 (avg 87.25)

      Non-AMP: 95 95 96 96 95 94 96 (avg 95.29)

      Here’s data from a performance test that ran 9 times for each version of the page: AMP, non-AMP

      The mean time to first byte for the AMP page is 1005ms, and for non-AMP it’s 989ms. So the server renders the AMP page 16ms slower, or 1.6%. This is a tiny amount, but is it enough to explain the discrepancy?

      The time until the page is visually complete for AMP is 2166ms, and for non-AMP it’s 1955ms, which is a difference of 211ms or 9.7%, a much larger discrepancy than that on the server-side. Also, I suspect the server difference might be because the AMP version needs to inline 50kB of CSS in the HTML, while the non-AMP version just links to an external file. This is an AMP requirement.

    2. 10

      Serve up AMP page to Google bots, and non-AMP to everyone else.

      1. 7

        When you visit an AMP page from Google’s results page, it’ll have a google.com url. You can’t get the higher ranking without serving the AMP version.

        1. 4

          Is this really doable? I.e. do you have experience / data that shows that this is something you can do without getting penalized?

          1. 3

            It’s hard to do, because Google bots for crawling AMP won’t tell you whether they’re Google bots or regular users.

            1. 2

              Agreed. It’s hard, but if we don’t fight back Google’s going to hoover up everything.

              I, for one, will not sit idly by and let the free and open Internet die. I remember the walled gardens of the 80’s, or the siloed access of the 90’s with AOL.

              1. 1

                I dunno. I still miss getting free frisbees in the mail every month.

          2. 4

            That won’t work — Google search users will get the AMP page anyway. Part of Google’s AMP implementation is that you no longer host the site yourself.

            1. 2

              That part would work fine. People are talking about giving AMP where AMP is not necessary, not when it is expected.

          3. 16

            There’s always a choice.

            1. 3

              Amen. And I will choose to fight.

            2. 9

              God… Where are we going? What would be end of that? And, if Google somehow gains the “ultimate power” where no one else could do anything significant on the web, what they’ll do?

              1. 4

                AMP is important as long as Google dominates the web search. Hopefully it won’t be for too long.

                1. 2

                  i actually have no problem with the technical aspects mentioned in the article - 50kb of css is more than enough, and limiting javascript can only be a good thing.

                  are there any negative technical aspects which give google more control? seems like they could still cache your page if it’s a sane static site.

                  1. 2

                    limiting javascript can only be a good thing

                    AMP requires Javascript (specifically, their Javascript) to render ‘properly’. This should be the first sign that google’s claimed goals with AMP are unrelated to their actual goals.