Summary of My Meeting with Google AMP Team
alexkras.com
alexkras.com
> I think I did a decent job communicating this concern to Paul and Malte. They both assured me that they understood the problem and are confident that they can find a solution that will work for everyone.
How about instead they just order some delivery and delete the fucking part where URL's were rendered unusable instead of this vague "is there anything else I can reassure you about" non-answer?
From time to time I still check if they included a way to go the real site.
My fear is that websites start using AMP as their CMS, because I can see it looking like a good idea "Hey guys, we don't need to manage how content! Let's just publish directly to Google and be done with it. So long, node.js"
I wish they'd just be up-front about it. Then it's just a straightforward decision: Are you willing to give Google control over your URLs in exchange for access to a speedy Google CDN with preferential search ranking? For some people, that is an extremely good trade-off (at least in the short term. It seems naive and dangerous to me in the long-term, but I'm not a CEO.)
Malte and Paul, what you are doing is a huge abuse of power with a random specced product and you are not listening to your users, especially those on Github. Many people made lists of drawbacks and the previous thread is full of many more. AMP is heavily flawed and beside the abuse of power, AMP is from a UI perspective an utter disaster: You wan't to improve the mobile user experience but steal so much screen real estate from am already limited mobile view, I made a quick mockup: http://imgur.com/a/VYVZ0 Just tell us, why does the bottom navbar on iOS stays now with AMP sites forever and does steal even more real estate? Who made this design decision and all the other random design decisions?
It would be great if Malte and Paul finally take the responsibility and comment to all the previous attacks here. I guess that Sundar has been informed about the dissatisfaction with AMP and it's now really time to give a statement. It seems that Google has already slightly deranked AMP sites in their SERPs which shows that even internally Malte and Paul are not fully backed up.
I think you have been drinking the punch at your lunch. That bar is still a problem and should not be there. There is a native back button, and there is no need for that bar. It lowers conversions for everyone's websites and increases everyones bounce rates.
Of course there is no problem with everything else google has implemented in terms of speeding up the web. That's great. It was a major achievement and a lot of good engineering went in to the caching, rendering, and image resizing and optimization machine that is AMP.
Your meeting should have been about hijacking the ui.
> I did not bring this up because I forgot :)
You punked out [0]
[0] https://www.google.com/webhp?sourceid=chrome-instant&ion=1&e...
I've set DuckDuckGo as my search engine on my phone for the time being as Google Search is practically unusable to me.
Sure, I could bookmark it and use it that way but I like the convenience of being able to search from the address bar.
I visited a subreddit with AMP and couldn't see nested comments or do sorting and other things. It literally hijacks the functionality.
Why reinvent the wheel to add those features when I am happy visiting the site to get it natively.
> For now if you want to vote or comment and you end up on an AMP page click the “View more comments” button below the first set of comments. This will take you to our regular mobile web experience where you can vote and comment to your heart's content.
Either way I'm not sure how to feel about AMP. I think it may force web devs to actually do something about bloat which is a very good thing or make bloat worse by offsetting mobile users to use AMP pages only and serving bloat pages to desktops.
Sure you could argue that companies can create a similar experience with their web sites but let's be honest that only few would invest the time for it, and then they'd still have a discoverability problem.
AMP might still be somewhat flawed, but it's a centralized step in the right direction, and it does sound like the team is open to feedback, so I'll give them points there even if they haven't made it perfect yet.