Hacker News Reader

Comment by cal85 | Hacker News Reader

Unless I’ve misunderstood, this a quite a big event in the history of the web platform. Apple’s quiet refusal to support this part of the PWA standard has for years been quietly maintaining a significant situation: that if you wanted to make a viable digital product that involved notifications, you basically had no choice but to make a dedicated iOS app, along with either a PWA or an Android app. If iOS Safari fully supports PWA notifications, with no funny business, that is no longer the case. Suddenly it is viable to make a single version of your app for every phone.

This also has big implications for the hegemony of the App Store and Apple’s control over the web platform. I don’t fully understand why they are finally giving up this power they have held onto for about 8 years, but I’m glad they are.

View on HN

Isn’t there a push by some legislators in some countries for Apple to allow alternative app stores on iOS devices in the near future? That might have something to do with this perhaps ...

Exactly, I think that's what this is about. either no need to allow sideloading cause PWAs, or it's an olive branch to iterate closer to prevent the big hammer from dropping. Also of course possible that Apple just supports it, but given then long history of evidence against that I don't consider it likely.

> This also has big implications for the hegemony of the App Store and Apple’s control over the web platform

No it doesn't. It will make zero difference.

Because developers have been building web based mobile apps for nearly 15 years now. And they never took off not because it lacks push notifications or some other minor feature. It's because the end user experience is terrible.

And I've yet to see anything that addresses the difference is latency, responsiveness, feeling etc that you get between native and web apps.

I build a PWA currently in use by around 130 people, no one can tell its not native.

I assume you have a native app as well and have benchmarked both with users.

Because when you compare native/PWA with mainstream apps the experience is significant.

> I assume you have a native app as well and have benchmarked both with users.

Yes, I can speak for them. They compared their native implementation with the one that exists. The users enjoy using the web app compared to the current native app.

Why? Because it doesn't exist. Because creating a native app doesn't solve the problem. Because when you say "native", do you mean iOS or Android? Or Mac? Or Windows? Or Linux? Or whatever else there is?

Yes, eventually, there might be a desire to build out a native app if it's needed. But the assumption that everyone needs a native app is absurd. Not everything needs a native app, and can work just fine without.

As for this persons users? Yes, they enjoy the PWA, because without it, the "native" app wouldn't work for some of them.

If it’s a choice between PWA and no app then of course PWA has a use.

But if you think it will upend the industry then that is frankly delusional. PWA apps are worse than native in every single way for the user. Great for developers. Terrible for users.

Developers trying to satisfy their users are not trying to upend the industry.

And as users can't tell, why is it terrible for users? All kinds of apps can suck and be user unfriendly. That's a different dimension than native vs web.

Users can tell.

That's the whole reason for why the last 15 years of trying web based apps have failed.

Is isn't noticeable at all if the app is built correctly. My app is used by health care professionals to log their hours and patient charting. Everything is instant when clicked and the app also has limited offline functionality. It's snappier than most mainstream native apps. Everything is native HTML 5 & CSS with very minimal javascript to handle logic and validate inputs.

> And they never took off not because it lacks push notifications or some other minor feature.

Actually, that is precisely why they didn't take off: Without push notifications a lot of mobile applications are simply not feasible. If something is not feasible, most people will have no reason to invest in ux. It's self-fulfilling.

However, with push api, cache api, being installable (and maybe some additional apis that elude me right now), I don't see any specific thing that would not allow carefully crated web apps to have great ux on mobile.

I am genuinely curious about this. I prefer native apps to web apps, but I have notifications disabled for every app I install except those that my wife and parents use to contact me. I’m sure I’m not alone in this behaviour - I hate notifications and any workflow model based around them.

Is my viewpoint very uncommon? Are notifications generally used and loved by users? Do app developers have good stats about this?

Thinking as a fellow notification disabler, and having no data, just intuition: People want their attention to be tapped. I am not exactly sure why that is (although I am sure that it can't be good) but notifications are a tap on peoples attention.

Building certain apps without notifications seems like it would simply not be acceptable to must users. Most phone apps still vibrate or ring – that's basically a notification. I would go as far as saying its uniquely distracting notification is the primary feature of the phone app but, at the very least, a phone without one would be something very different. And since messengers basically compete with phone apps now, for most people that would probably not work.

It's not an uncommon viewpoint. But the fact of the matter is that without notifications it is hard to build a habitual trigger loop to re-engage users (a lot of things just fall out of mind and sometimes you'd like a reminder or a notification of something special). If you are careful with what type of, how many, and how you send notifications, then not as many users will disable them.

Yes absolutely. People will learn a new way of interacting with webapps. Currently nobody installs a pwa. But maybe that'll change. Going through the web browser is inconvenient for some apps (unnecessary extra menu on top). After installing however, the experience is very similar.

Download the Twitter and PWA apps. Compare the two.

If you think the only difference between them is push notifications and one click install then of course you will think PWA will be a success. But a lot of us can easily see the difference.

And based on the last 15 years of real world experience there are enough users like us to sink the hopes of PWA apps.

When you have a strong reason to not optimize one of two ways of building your mobile app, you are not going to optimize one of two ways of building your mobile app. Not having push notifications – for years on end, with no inkling if that might ever change, because the platform makers have strong reason to sabotage it – is a pretty strong reason, specially when your measure of success is literally how often users get distracted pulled into your app.

In 2023 there is absolutely nothing that would technically not allow us to build a buttery smooth 90fps experience in any mobile browser running on a recent-ish phone. Of course that is not to say it's particularly easy. But then again, so isn't building a performant native app.

The twitter PWA is actually a very good PWA. But it doesn't implement some gestures like swiping left to see more feeds on the home screen, and has little to no animation support (retweet dialog pops up, instead of animating in like it does in the native app). It also does not do paint holding, so you get a temporary white screen when swiping back to the previous screen. But all of those problems could be fixed, with current web tech, afaict.

Is there any other lackings (which might be fundamental and unfixable for a PWA) that you had in mind?

And what about the 80%+ of app revenue that are from games?

> No it doesn't. It will make zero difference

I'm pretty sure it does, at least with purchases. PWAs won't need to pay for apple's app store 30% cut.

I once made a chat app that had significant traction on Android, and many users wanted to try it on iPhone but there was no way to do so without web push, and it was fundamentally banned by the app store guidelines and without side load there were no alternatives. This is a game changer for those kinds of apps!

I don't know what the percentage of "native" apps that are really web views and/or javascript underneath, but am certain that it's a lot. In fact, I wouldn't be surprised if it's more than 50%.

This 100%. Amazon web app is awful. Same goes for any ecommerce pwa I've used on my phone.

Things have changed a lot in the last few years. This is a React Native vs. Ionic React+Capacitor responsiveness demo: https://twitter.com/magnemg/status/1404073426193690634?s=20

You can achieve 60 fps with webapps now: https://ionic.io/blog/ionic-vs-react-native-performance-comp...

With the Shared Element Transitions API you even get smooth page transitions (even for MPA's with full page reload): https://www.youtube.com/watch?v=6mv0_jsWhoE

It came out in the Epic Trial that 80% of all app store revenue comes from games and in app purchases.

The difference that web push notifications will make is infinitesimal.

Reply on news.ycombinator.com