Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

While I hate Apple's anti-consumer practices as much as anyone, the PWA platform is a system set up by Google first and foremost. Take-up has been limited outside of Google Chrome. I wouldn't say Apple's PWA approach is necessarily an example of Apple's fuckery.

This wouldn't be much of an issue, of course, if Chrome would just run on iOS like it does on any other OS, so Google can implement PWAs themselves.



Mobile web apps that can be installed on device were invented by Apple.

This was the way developers were supposed to develop apps for the iPhone when it was released, before Apple introduced the App Store.


I don’t think that’s true. Apple said web sites were the way to add functionality to the first iPhone, but “can be installed on device”?

Jobs framed it that way, but IIRC, all you could do is create bookmarks. Creating an icon on the Home Screen? Impossible. Reliably storing data on-device? Impossible. Backing up your on-device data? Impossible. Accessing your on-device contacts, photos? Impossible.

Also, Jobs made a vision statement about web apps in June 2007, but Apple announced a SDK only four months later (in October 2007) and shipped it in March 2008.

⇒ I’m fairly sure he knew about that SDK when he made that statement.


The ability to install web apps that open as standalone apps, and not in Safari, was introduced by Apple with iOS 2.1 in 2008. Well before this ability was added to Android.

Apple invented installable mobile web apps.

Link about the needed metatag: https://www.mobilejoomla.com/forum/4-feature-requests/330-ip...

Steve Jobs introducing web apps as the way to develop apps for the iPhone in 2007: https://williamkennedy.ninja/apple/2024/01/30/steve-jobs-int...


Mobile web apps were what Apple wanted developers to use, but they weren't new, let alone invented by Apple.


I didn't say Apple invented mobile web apps. I said Apple invented the ability to install mobile web apps on device.

I'm not 100% sure no other mobile OS allowed this before to be honest, but I'm pretty iOS is the one that popularized it.


Another Apple myth, Symbian had a Web runtime before anyone at Apple came up with the idea.

Also that was precisely the idea behind Windows 9x Active Desktop apps.


IMHO. Apple were the first to make it useful. Because the iPhone was always online and the browser window was limited. Active Desktop aimed for the technological stars and was just buggy and slow as a result, it was cool but too flaky to be used.

Symbian I just never had an Phone expensive enough to use like that.

In the end none of them really worked out I guess.


> This wouldn't be much of an issue, of course, if Chrome would just run on iOS like it does on any other OS, so Google can implement PWAs themselves.

You do understand that the reason it doesn't is because Apple won't let it, not that Google don't want to?


Of course, Apple is sabotaging Chrome and has been for years, and that's a much bigger problem than PWAs. The Safari team shouldn't need to implement PWAs against their will, Apple should instead let Google bring out a browser that does PWAs and then let the users decide if they want to use PWAs or not.

Google does something quite similar, though; Chrome can install applications into Android's app drawer, but that requires privileges other browsers can't attain, needing to resort to things like widgets instead. Firefox doesn't care about PWAs and Apple doesn't care about any platform but their own, so it's not as obvious a problem, but Android is full of "you must be the manufacturer or Google to compete" permissions. Android is just a lot better at fair competition than iOS, to the point you'd barely notice.


This is a fantasy. No customer wants PWAs. They exist to make developers' lives easier, not consumers' lives.


The consumer doesn't care which method is used to serve an application. PWAs could easily be presented to the end user like a native App.

The problem is rather that PWAs would prove a viable path for universal cross-platform applications, taking away the gatekeeper role the OS-vendors have.

Paradoxically PWA-support is also part of the "we're no gatekeeper" narrative, so it's in the OS-vendor interest to keep it maintained as a hampered alternative to native apps.


> The consumer doesn't care which method is used to serve an application. PWAs could easily be presented to the end user like a native App.

No it can't. The web will never support what's necessary for parity with native apps. Imagine trying to implement Liquid Glass in CSS.


First, you're mixing up capabilities of PWA vs native apps (no one stated they're equal) and how an OS presents Apps differently from PWAs (which was my point).

Second (even though it's completely beside the point), especially Liquid Glass could be implemented in PWA, because it's a rendering effect the OS could put on top of appropriately tagged elements of the application. And voila, the same webapp could render in Liquid Glass in IOS26 and in less-gaudy Liquid Glass in IOS28, and meanwhile in no Liquid Glass at all on devices that don't have it...


PWAs are the primary way for small busineses to have internal private apps for running staff services on local devices. Apples App Store has way too many hoops to jump through and has far too high a wait time to publish for businesses to move fast and update internal apps with bugfixes and new services etc.

Android accomplishes this by allowing devices to connect to private app stores and repos, which enable companies to issue their own apps on their own terms. As Apple plays hard ball on this front, the only way is to use a PWA.


Custom apps published for internal use by companies with fewer than 100 employees who aren't eligible for enterprise app distribution sounds like a niche of a niche use case, so it's pretty consistent with my view that they're more developer catnip and not a serious technology.


Thats a lot of assumptions to back up your own point.


Your point was PWAs are necessary for small businesses to distribute apps. I just spelled out what that meant, since businesses with >100 employees can just use enterprise app distribution on iOS.


I would personally say that the barrier to entry for Apple enterprise app distribution is very very high, and requires so many hoops to be jumped through that it would take a much larger company than 100 staff to warrant that

https://support.apple.com/en-gb/guide/deployment/depce7cefc4...


Developer efficiencies can be translated to customer wins.


Then allowing Apple the efficiency of not implementing yet another way to build a GUI also is a customer win.


Apple already implement everything needed. They just decided that they can clear client-side storage for PWAs whenever they like (deleting user data), making them useless for anything that needs to store data and isn't synced to the cloud.


The goalposts move every time Apple resolves some bug that PWA advocates promise is the one issue holding them back from taking over the world with crappy web apps.


Nope, it's been push notifications and persistent storage for a decade.


I would also add: - Install Prompts / Discoverability - Bugs (like https://github.com/web-platform-tests/interop/issues/84) which likely would only get fixed if Safari has heavy competition on iOS (it's a funding issue)


Apple hasn’t resolved any of the main issues.

Install and discoverability is still hidden. Push is gated behind install. Safari’s scroll bugs haven’t been fixed despite us extensively documenting them, emailing to Safari’s leadership and raising them every year as the number one bug.

The number one thing we’ve asked for is third party browser engines on iOS.

What goalposts do you think have moved?


Certainly in theory, almost never in practice. The enterprise slop shop that chooses web technologies because the consultants are cheaper is not trying to make anything lasting or delightful.


Microsoft promoted PWAs pretty heavily back in Win8 days. That was the reason why HTML/JS was one of the "first class" (alongside with C++ and .NET) development stacks introduced there - the idea being that people would write cross-plat PWAs to target iOS and Android, but then they would already have something that's 90% of its way to a native Win8 app.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: