If you’re ever in the position where you have to market an app you’ll soon discover actually just getting users to commit to installing something which they’ve never heard of is a massive hurdle - compared to clicking a link in a browser.
lol — have you tried to engage, in a serious way, with the JavaScript ecosystem? It is a _dumpster fire_ to produce anything that is reasonably as interactive and complex as an iOS Application.
I do agree, however, with the idea that some things ought _not_ be apps, especially things that are dead simple, for which it _truly is_ easier to develop a web application.
Lol, seems like you haven't tried to engage with the web ecosystem in a serious way. Web apps with TypeScript and React/Vue/Svelte are the easiest, most productive way of making software that runs immediately anywhere, and the ecosystem has everything (don't forget about Wasm).
SwiftUI is nice, but needs a lot more years until it reaches the level of React.
Android libs are a joke.
(I write enterprise React applications last 6 years, and used to work with WinForms before that.)
>Yeah, now I can see that you have indeed never really tried.
This is totally untrue!
I am open to the possibility, however, that I have been working with a challenging front-end codebase and that, on that basis, I may not have seen the best that JS has to offer. But in my experience — intermittently building toy iOS apps the last few years and having also built some production react apps in a large codebase (perhaps not a fair apples-to-apples comparison) — iOS has felt much more straightforward for creating apps with idiomatic UI/UX.
Forgive me if I came off as a bit glib. But I do tentatively, at least, still hold my original position; though am open to the possibility of being wrong!
You could flip it around and ask - given that phone chips are approaching desktop level performance, how native do you need to go? We're already seeing native apps become less "native". Using high level frameworks that provide a common core between platforms is pretty standard nowadays.
IMO the biggest driver already is and will be app store/platform policies. Fundamental hardware constraints are secondary.
Unfortunately PWA seems to have almost no adoption at all. I'm not sure why that is. Native features missing? Developers want to fingerprint you? Distribution mechanisms insufficient? All I know is PWA seems to have stalled.
One issue with PWAs is that Apple has hobbled what they can do on iOS. I think this prompts a lot of would be adopters to skip PWA and go straight to packaging their app with electron.
Not an issue to do and hence not an issue for public at large.
But apps come with plenty of their own issues which is gaining public recognition. It is early days yet though.
Personally:
- I find it odd to need an app for a service entirely reliant on off-phone data.
- I like my phone battery life. I avoid apps if I can help it. Web page links all over my home screen.
Isn't your anecdotal statement that which should be replaced by a citation, as it's an of-no-use example-of-one? Whereas what I said might not have been accompanied by a citation, but it's a general statement, easily verifiable, including with citations:
Not to mention there's no "grandmother test" as some kind of ultimate marketing gatekeeper, unless you market adult pads or something. I'm pretty sure lots of billion dollar industries fail the "grandmother test" too (youth-oriented ones, self-selectively so).
Same for the CEO of my company. Each additional app is a potential security risk they want to minimize. Web browsers they don’t seem to understand the risks as well. They think of webpages as being a potential attacker on the street that they can just avoid where as an app is a potential burglar of their home with their prized possessions
It's true that apple hobbles PWAs on their mobile devices. But not everyone uses those devices and most apps don't make money off the initial purchase.
ARM has a huge problem with compatibility and secrecy.
Buy a random intel machine .... will it run Windows and Linux? YES.
Buy a random ARM device..,, will it run Linux? Maybe, probably not, even if it does, probably there’s problems and issues caused by the CPU vendor keeping aspects of its design secret.
I do agree, however, with the idea that some things ought _not_ be apps, especially things that are dead simple, for which it _truly is_ easier to develop a web application.
SwiftUI is nice, but needs a lot more years until it reaches the level of React.
Android libs are a joke.
(I write enterprise React applications last 6 years, and used to work with WinForms before that.)
This is totally untrue!
I am open to the possibility, however, that I have been working with a challenging front-end codebase and that, on that basis, I may not have seen the best that JS has to offer. But in my experience — intermittently building toy iOS apps the last few years and having also built some production react apps in a large codebase (perhaps not a fair apples-to-apples comparison) — iOS has felt much more straightforward for creating apps with idiomatic UI/UX.
Forgive me if I came off as a bit glib. But I do tentatively, at least, still hold my original position; though am open to the possibility of being wrong!
IMO the biggest driver already is and will be app store/platform policies. Fundamental hardware constraints are secondary.
PWA trumps mobile apps — until or unless it can be articulated why a PWA/hybrid app won’t work, but that is the small minority of app cases.
In what sense? Most people use mobile apps than browse the web through mobile -- and most surf from mobile vs PC.
(The fact that said mobile apps might be web-based is an implementation detail).
Personally:
- I find it odd to need an app for a service entirely reliant on off-phone data.
- I like my phone battery life. I avoid apps if I can help it. Web page links all over my home screen.
Well, a browser is an app for services entirely reliant on off-phone data :-)
Just a generic one, slower, more battery hungry, with less platform conveniences and access to native APIs.
Which is what I was trying to say though I did it poorly I guess. :-) Why do I need an app when the website will do just fine?
"slower, more battery hungry, with less platform conveniences and access to native APIs"
I think the efficiency loss is traded for the lack of always on services and privacy invasion common in native apps these days.
The important function of the web browser is the sandbox effect.
My mum wont install an app unless she has to. In my eyes this has failed the grandmother test.
https://www.mobiloud.com/blog/mobile-apps-vs-the-mobile-web#....
Not to mention there's no "grandmother test" as some kind of ultimate marketing gatekeeper, unless you market adult pads or something. I'm pretty sure lots of billion dollar industries fail the "grandmother test" too (youth-oriented ones, self-selectively so).
ARM has a huge problem with compatibility and secrecy.
Buy a random intel machine .... will it run Windows and Linux? YES.
Buy a random ARM device..,, will it run Linux? Maybe, probably not, even if it does, probably there’s problems and issues caused by the CPU vendor keeping aspects of its design secret.
ARM is a very very long way from replacing Intel.
Chromebooks run android apps today.
Windows computers will run android apps with BlueStacks or a similar emulator.
Until mobile apps can do EVERYTHING that desktops apps do, desktop apps are here to stay.