Before I jump in to some of the points, I hope it is clear that I do understand the frustration, I understand how things look, and I understand that users shouldn’t care about our org charts etc. But saying all that, here goes.
You define Google == Android and Google Play Services == Android, both of which aren’t accurate, so I just want to run through a few more points.
However, we shouldn’t forget we are talking about an operating system with two billion active users monthly. Two billion.
Later you mention Android Oreo. This is the exact point of Google Play Services, to help us deliver a consistent set of experiences across all of those users and not have to rely on Android versions being deployed — we all know how that story plays out.
Arguing that Oreo should have Web APKs etc is such minor piece in all this when you consider what Google Play Services can enable for the entire ecosystem across all devices and users. It is this service which enables Web APK’s for Chrome and will enable it for every other browser too. It’s simply not feasible to rely on operating system updates.
I can’t believe there was no time or resources or enough priority to ship a new OS version without harming PWAs if the solution is around the corner.
I understand the intent of what you are saying, but you are looking at this from a bookmark world where PWA’s were only faking it as a native integration. Yes our roll out of Web APK’s is slow, but it’s such a fundamental game changer that I think it’s ok to get it right slowly rather than rush it and ruin our chance.
Yes, WebAPK might ship in a couple of months and solve most of the issues stated here, but it feels like it should be here today with Android Oreo launch.
Android OS has a permission in the manifest that allows you’re app to install Apps. I believe that outside of pre-installed OEM apps, only one other app has this permission — Google Play Services. I’m not sure that anyone would want to open that permission up to the browser given that the web is an actively hostile environment for users, which is why we delegate through Google Play Services. At the same time you want to bake this capability in across the ecosystem so that every browser across every version of the OS can use this and not open the user up vulnerabilities.
When the Android team decided to change the shortcut behavior, it’s clear they didn’t think on PWAs. They didn’t care about it. I might be wrong, but that’s the public image that we see from the outside. Several browsers with a broken behavior? Automatic Badging and Shaping for PWAs?
I get frustrated with many teams, but don’t take this as malfeasance. User trust is at the centre of the badging story. When an app is installed it gets the ability to have a presence on the homescreen through the defined actions and components. That icon is clearly tied to an APK and an app-id so it is entirely clear of the intent of that icon and is discoverable. Bookmarks from the web, which is what they are, now give the user the ability to understand that app that controls, it helps prevent spoofing and it helps the user understand which browser will be opened from the homescreen icon, it’s already a mess with the versions of Chrome, but it’s completely unclear to users that Opera, Firefox, Samsung Browser might be the app that made the short-cut. Badging makes it a lot clearer to users and it helps users manage their profiles a lot more effectively, i.e, Chrome might be the work browser, and Opera the at-home browser.
With a WebAPK, the intent is entirely different, it’s like an app, it is in all the expected places and could have extra privileges too.
And of course Google can set its priorities, but don’t lie to the community.
Huh. Thanks, I guess.
