Keynote: On Page Experience - Malte Ubl, Principal Engineer, Google

hey everyone my name is melta and i'm a member of the technical steering committee for the amp project and the tech lead for the team at google that's contributing to app now i want to recognize that there are worse problems in the world right now than javascript frameworks and web performance i want to thank the protesters in the streets and unequivocally state that black lives matter in today's talk i'll talk a little bit around the original thinking about amp how it evolved over time and how we think it will develop in the future

now thinking back to like the summer of 2015 this post by the wire even though it's a little bit older it was kind of capturing the sentiment at the time like the web is that native is taking over and also at this time platforms like facebook instant articles and apple news are kind of coming up and we were having all these conversations publishers who were talking to us and were worried about the situation and wanted to find a solution that worked for them and work for users and kind of made the web um become um viable for

users again and obviously for us at google this was also really important because we're extraordinarily successful on the web and kind of out of this sentiment um the amp project was born with the vision of creating a strong user first open web forever now we highlight user first because it's really about rethinking how we can do things in a way that they really think about what the users want and make it great for them instead of for example putting a you know pop-up app in front of everyone however

for it to stick around forever it also must work for publishers and merchants because if their business aren't successful you know that website will just go away now one good way i think to think about how we wanted amp to be was to kind of think about that performance is this race different race cars right and so certainly in 2015 i'm not sure this actually changed was that the average mobile site you know just wasn't doing quite uh so well and was kind of there in the

back of that race right and but we also realized that there wasn't actually this is really important that was not a technology problem you could make a website with great ux great performance who was right there at the top right if you just hand tune it if you were an expert now our goal with amp was to get up there in the front we were not aiming to win against every hand tuned website in the world we just wanted to be up there and and

this is really important we wanted everyone to be up there right all amp documents and um this actually brings me to a point i don't think it's talked about enough which is that web performance is biased right roughly the smaller a website is the less well funded it is the slower it is and i'm really really proud that amp does not have this bias very roughly you

know obviously there's some variation but like i'm documents have a quite uniform speed even though there's huge diversity in how they look and how their experiences and you know i think it's it's fair to say that this worked in terms of adoption there's now tens of millions of websites using and billions of pages you can't count them and you also like it doesn't matter right like it's a very large number so that's that's kind of how how we got started and then as it does the internet got mad at us

all right and i actually want to acknowledge that like many you know there's complaints and criticism about amp and i actually personally agree with many of them and we've taken that to heart and and worked of them right i want to broadly summarize the known issues that amp was facing as one of them being awkward urls on google that was largely based because of the iframe based viewer which created that privacy preserving instant loading and on the other hand there was this feature called top stories