... or they could just create a PWA and publish it on the free and open web and have the app be available globally in 40s w/o any regulations or artificial restrictions.
* You can use all the device APIs, not just the ones that have been exposed as web APIs.
* They're way faster and less janky.
* You don't to use endless workarounds to disable typical web behaviour when you don't want it.
PWAs are probably suitable for basic apps that just display information, e.g. an app for an event, or a pub or whatever. You would be insane to try and make a "real" app like Snapchat or Google Maps as a PWA though.
Google's Android UI toolkit is slower than Chrome building UI from HTML, sadly. So it's not a clear win for native apps (unless the native app is using HTML for UI, I guess).
That said, I don't think PWAs have a good way to get background data updates, which is unfortunate for apps that could work offline. You can't do a silent push to update data, last I checked, anyway. (Use case: I want a weather app that has current/recent data without loading when I start the app, because I often start the app in dead zones; also I want to have it background fetch the data when my location changes).
You mean slower to write? You're probably right for native Android. But I didn't say anything about development effort. Also it's definitely easier to get polished apps with Flutter than PWAs, so I think they'd still lose on that front.
No, I mean slower to run. It takes way too long for widget objects to be created, and you can only do it on the UI Thread, and you have to make most of your widgets when the app opens (otherwise there's nothing to see).
It was also slower to write for me, but I've never had a good time understanding UI toolkits, so I assume that's my own character flaw.
Erm. There's no way the web is faster than Android's native widgets or Flutter's. Your complaint about the UI thread makes no sense for a start - the web only has a UI thread.
I wish you were right. But for a reasonably simple (in my mind) weather UI, constructing the native Widgets on a cold start takes long enough that you get warnings for too much activity on the UI thread. But you can't build Widgets off the UI thread to get parallelism like you would do for other things that are slow, because Widgets have to be built on the UI thread if you want to show them.
A similar HTML layout shows up a ton faster in Chrome or a webview.
There is no 'free and open web,' that's long in the past now, to the extent it ever existed (for about 10-15 years in a limited form).
Which countries partake in this supposed free and open web? None.
There are dozens of separately regulated prominent webs now and that's going to get a lot worse this decade. We're never going back, nations are not going to give up the control they're assuming, it will splinter further and further; complexity will grow, accessing foreign markets will become more onerous. It'll gradually fully mirror the physical world, where if you want to travel to another nation, or do business in another nation (access their market), you will have to comply with all of their myriad of local regulations (which will continue to grow rapidly in number and complexity).
Nations will fence off their citizens online, and you will have to comply to access them. We're already seeing this directional shift in many nations, it will spread comprehensively.
Turkey's web is not regulated the same as India's web which is not regulated the same as China's web which is not regulated the same as the EU's web which is not regulated the same as Brazil's web which is not regulated the same as Australia's web which is not regulated the same as the US web which is not regulated the same as Britain's web and so on.
That’s not the problem, it’s having to manage a bunch of sets of credentials, and the logistics of pushing updates through multiple stores. Both are obviously solvable problems but definitely annoying barriers to someone who really just wants to make their app.
Or more realistically, it means developers will have to start paying for a middle man saas thing that distributes apps in all the stores and aggregates reviews etc.
Have you tried publishing apps for the App/Play Store? Last time I made a game with Unity then I had to export it to different platforms and for some of them, such as iOS, then I had to make additional code modifications with xcode to get e.g. ads to work properly. I’d easily end up spending a day on making sure each store listing looked good (required files in different formats, and tons of settings to setup).
I also struggle to think of a scenario in which you wouldn’t want your app published on all app stores (unless you for moral reasons wouldn’t want your app on Chinese-owned stores or something like that).
Yes, I do have applications on Android, iOS, macOS.
I want more competition in app stores, I want places where adult applications can be distributed which is completely impossible at the moment. I want app stores that offer different models than what Apple and google do. I want Apple to have to improve their developer experience because of competition (the google play store is way simpler to deal with in my experience). Etc.
You are used to Apple-style bureaucracy. It won't be even close. If 10 countries roll out their stores - most of them probably won't even be in English for a long time. Custom SDKs will add a lot of overhead on the development side. And app review madness will be like never before.
There are 193 countries (footnote[0]), so that might mean 193 app stores, but then some countries might delegate the power to require their own app store to their own regions/"states".
i guess there's a right number. I love the idea of having an alternate solution to reach your market whenever you don't want to comply with apple or google guidelines for some reason
