It depends on your requirements and what features you want to provide to your users.
Do you want to provide only an app for mobile devices or also a desktop version?
At the moment I don't use many PWA features on this PWA, like notifications for example, but in another PWA I built I used camera access to take photos and scan barcodes and it works well.
I use GoLang, HTMX, W3CSS, and Handlebars. The codebase is the same for all the environments: phone, tablet, desktop.
I considered Ionic Framework for the front end, but it uses only web components with shadow DOM and HTMX doesn't work well with this technology on forms submit.
W3CSS is fast and small, easy to understand, change and manipulate.
I never did bother finishing the Android Chrome PWA part, because as others noticed it doesn’t get you much, and users really don’t seem to bother with the iOS one which does work. (iOS has navigator.standalone for detecting this).
iOS is arguably not “real” PWA but these days has equivalent functionality for much of it that is even supposed to work. It is on the todo list this week to get iOS push happening as proof of concept.
By far the biggest problem is lack of user familiarity. The app stores are understood, while the process for installing these is not. You can trivially package such things for the app stores, but then you have to jump through the increasing amount of hoops about that.
I have built a web app for streaming music (for personal use/satisfaction) that i primarily use as a mobile PWA.
PWA is sort of like if you tied an electron app to the user's installed browser, so you are still going to have browser specific quirks. Not all browsers support installing a PWA (most notably for me, safari on macos).
On mobile (especially iOS) PWA can have considerable limitations compared to a traditional app, but less limited than running the app "as a websute". Local storage limits come to mind. I believe a PWA on iOS can only store 50mb?
Distributing it is in theory VERY easy. but its not a common way that users install apps, so it comes with more explanation than simply saying "download from the App store".
React native is a whole other thing entirely, and is not limited to simply being an embedded website in a browser wrapper.
If there already exists a web version of the application, then the saved effort might be worth these quirks, but if its a greenfield project I'm not sure the "Ease of development" actually beats something like React Native.
I built a 3D scene creation tool for visual artists as a PWA about 6 years ago. It works great, supports all devices, etc. The only issue I've had with it is that iOS hides the functionality to install it in the "Share" button -> "Add to home screen".
Most regular users aren't aware of PWAs either, and use the App Store as a way of finding apps. We've since released a simple wrapper of the PWA on the App Store as well but had to get really messy to conform to Apple's payment requirements.
> Most regular users aren't aware of PWAs either, and use the App Store as a way of finding apps.
x1000
Normal people have got to the point they trust app stores a lot more than the web. I know of several startups that were seen as not credible by potential customers due to a lack of app store presence.
Thanks for the comment!!
Can you elaborate on "multimedia heavy"? Do you mean photo/video contents?
WDYT about using PWA-s for something like chatgpt mobile clone?
Can you share the thought process a bit deeper?
It depends on your requirements and what features you want to provide to your users. Do you want to provide only an app for mobile devices or also a desktop version?
I am working on my own PWA, source code here: https://gitea.speedtech.it/roberto/BrainMinder.
At the moment I don't use many PWA features on this PWA, like notifications for example, but in another PWA I built I used camera access to take photos and scan barcodes and it works well.
I use GoLang, HTMX, W3CSS, and Handlebars. The codebase is the same for all the environments: phone, tablet, desktop.
I considered Ionic Framework for the front end, but it uses only web components with shadow DOM and HTMX doesn't work well with this technology on forms submit.
W3CSS is fast and small, easy to understand, change and manipulate.
This was one of the experiments with the games I did like https://luduxia.com/reversi/
I never did bother finishing the Android Chrome PWA part, because as others noticed it doesn’t get you much, and users really don’t seem to bother with the iOS one which does work. (iOS has navigator.standalone for detecting this).
iOS is arguably not “real” PWA but these days has equivalent functionality for much of it that is even supposed to work. It is on the todo list this week to get iOS push happening as proof of concept.
By far the biggest problem is lack of user familiarity. The app stores are understood, while the process for installing these is not. You can trivially package such things for the app stores, but then you have to jump through the increasing amount of hoops about that.
I have built a web app for streaming music (for personal use/satisfaction) that i primarily use as a mobile PWA.
PWA is sort of like if you tied an electron app to the user's installed browser, so you are still going to have browser specific quirks. Not all browsers support installing a PWA (most notably for me, safari on macos).
On mobile (especially iOS) PWA can have considerable limitations compared to a traditional app, but less limited than running the app "as a websute". Local storage limits come to mind. I believe a PWA on iOS can only store 50mb?
Distributing it is in theory VERY easy. but its not a common way that users install apps, so it comes with more explanation than simply saying "download from the App store".
React native is a whole other thing entirely, and is not limited to simply being an embedded website in a browser wrapper.
If there already exists a web version of the application, then the saved effort might be worth these quirks, but if its a greenfield project I'm not sure the "Ease of development" actually beats something like React Native.
I built a 3D scene creation tool for visual artists as a PWA about 6 years ago. It works great, supports all devices, etc. The only issue I've had with it is that iOS hides the functionality to install it in the "Share" button -> "Add to home screen".
Most regular users aren't aware of PWAs either, and use the App Store as a way of finding apps. We've since released a simple wrapper of the PWA on the App Store as well but had to get really messy to conform to Apple's payment requirements.
If you're curious, here's the tool: https://justsketch.me
> Most regular users aren't aware of PWAs either, and use the App Store as a way of finding apps.
x1000
Normal people have got to the point they trust app stores a lot more than the web. I know of several startups that were seen as not credible by potential customers due to a lack of app store presence.
Fantastic tool btw!
I believe there were issues with ios support of some features.. is don't remember which one..
Reliable Install, native share, background tasks, etc
Notifications were also a huge issue for iOS PWA. I haven't followed so I'm not sure if anything changed.
They were (and are) a huge issue everywhere, see a link to the thread below. Some active fans made us to believe it’s an ios problem when it wasn’t.
Depends on what kind of app you have, if it is multimedia heavy or needs a lot of native capabilities, it's better to use react-native.
Thanks for the comment!! Can you elaborate on "multimedia heavy"? Do you mean photo/video contents? WDYT about using PWA-s for something like chatgpt mobile clone? Can you share the thought process a bit deeper?
What native capabilities do PWAs not support?
I’d test (and extensively read on) push first.
Found it: https://news.ycombinator.com/item?id=38402368
reading files from sdcard or a million other things
As a reference: https://whatPWAcando.today/
best first move imo