I think it's not correct to conflate the App Store and the Play Store like this. Apple is very hostile to PWAs in the App Store, whereas Google wrote a tool to easily package your PWA for the Play Store https://github.com/GoogleChromeLabs/bubblewrap
So for me, I publish to the Play Store to get some free users who find my app that way. App Store, not worth the headache of trying to skirt Apple's rules.
Ideally I think app stores should allow listing PWAs directly, rather than banning them (like Apple) or forcing them to use a wrapper (like Google). Whatever complaints people have about that (app can change without review, etc) also apply to the current situation in the Play Store. Once your wrapper PWA is accepted, you can change the website however you want. So what's the point of the wrapper?
> Ideally I think app stores should allow listing PWAs directly, rather than banning them (like Apple) or forcing them to use a wrapper (like Google).
I think it’d be a bit of a challenge — in both ecosystems, corporate MDM and parental controls currently define policies to control app installation by store-specific metadata. You’d need equivalent PWA standards these OSes could grab ahold of, to determine which PWAs to allow installation of, or to surface in app discovery views in the store. (Probably it’s be something akin to the old https://en.m.wikipedia.org/wiki/Platform_for_Internet_Conten... system, but probably reworked now that we actually know what things people care about preventing their kids — or employees — from accessing online.)
Which in turn, to support the spirit of PWAs as a vendor-neutral “web thing” — means that the OSes would need to agree on what exactly the set of content rating controls are that they’re allowing administrators to define policies against. I.e. the stores would need a more-unified data model; perhaps even an externally standardized one.
Why? Remember, current status quo in the Google ecosystem is that I can write a PWA, put a wrapper around it, list it in the Play Store, and then update the PWA however I want with no review. If that's good enough for whatever corporate/parental controls Google has to obey, then just listing the website directly should be too.
They could ask for whatever extra metadata they need upon submission of the URL to the app store. Or standardize it so we could just put it in some JSON file.
We are currently in this mess off app store monopolies issue because of Google and Apple. PWA is trying to be the solution to this. Why would you upload it and repeat the same mistakes all over again?
It's like those people who use 2FA TOTP functionality along with the password managers. My friend, that defeats the whole purpose of 2FA. It's not just supposed to be an extra step, it is supposed to be isolated in another device from where your passwords are not. And 2FA's are NOT supposed to be in the cloud.
But this will happen. Because VC's will tell them to do this. And thus people it will eventually become a norm. sighs
> It's like those people who use 2FA TOTP functionality along with the password managers. My friend, that defeats the whole purpose of 2FA.
It defeats _one_ of the purposes, and if that's not _your_ purpose, than it's acceptable to you. One of the purposes of 2FA is to stop randos in non-extraditable-country-x hacking your account by attacking just your password.
What it doesn't cover is protection from having your machine compromised, since as you've pointed out, everything is in once place (in memory if the password manager is unlocked). For me that's an acceptable risk-to-convenience trade, since I'm in much bigger trouble if my machine is compromised anyway.
I think the advice makes sense for some apps depending on your goals. If your competitors are in the App Store, having to explain the PWA install process will slow your word of mouth growth, and you won’t look as legitimate. If the point of your PWA is just to make a website or tool used by a small or captive audience (something like Worday), then no need.
Like, if you were considering buying an office product, would you buy the one that’s available in the App Store and online, or the one that’s not on the App Store?
Yeah, his example of ChatGPT being a web app seems off. I have the app on my phone? Is it just a thin wrapper over a web view? Probably. Still an app and sort of runs counter to Kevin’s argument
I wish the native ChatGPT app on Android had all the functionality of the web app. I dearly miss the ability to navigate conversations as a tree, going back and editing any prior turns to try out different ideaa.
At least 62% of visitors on my PWA come from Play Store, with 20% being labeled "unknown".
It really depends on your app, but many mobile users just search for stuff in Play Store first before searching on Google, and even Googling it will help as you will get a widget about your app (sometimes). The web version is (for the most popular queries) on average 1.6th in Google results.
Although the dataset may not be very large (10k+ installs shown on Play Store page).
Everything is an ad these days, check any new content similar to this blogpost, YouTube videos, etc. That said, the article is brief, direct, deliver good points then wraps up with an ad.
I don't think there are any good points at all in this. On the first 5 (I gave up caring to type up more):
> 1. This won’t get new users
App Stores are still a discovery mechanism and it's a pain teaching users how to add them to their home screen vs. just linking to an App Store entry.
> 3. It defeats the whole purpose
There's times to take on the world and times to join it, and the current trend is to put them in the App Store. I don't know how many companies starting out can afford to wait for the world to change on this.
> 4. PWAs don’t belong there
Incredibly subjective.
> 5. App stores aren’t more convenient
...huh? Users are already familiar with the flow to open an app store and search for a company. Users are generally not familiar with adding a bookmark of a PWA to their home screens.
I agree with the idealistic sentiment, but disagree with most of the points made in this post and if I had a product or business I wanted people to use, I'd absolutely have it on a store (disclaimer, I now work for Google Play, but base this on my previous experience at a retail startup).
1. Organic app store traffic is non-zero, particularly in markets with less tech-savvy users. People do search app stores for apps to solve problems they have. Also it's a marketing channel (like any other performance marketing channel) that you can bid on and spread marketing across, more channels typically means lower business risk.
2. Yes there are horror stories, but apart from the HEY story (which comes with a dose of DHH caveats), many policies are things like needing a privacy policy, or needing account deletion functionality. If you can't meet these sorts of requirements for an app store I think there's often a deeper problem for users (although I recognise this is not always the case, legitimate issues do slip through cracks in my non-Google experience).
3. This is web idealism. I agree in principle, but the world doesn't work this way.
4. I don't think the sort of modern, good PWA is what Apple envisioned when they said "don't repackage your website". If an average user can't tell the difference, Apple won't care. And that statement from the MS open source project does not appear to be from a user with any actual association with MS or the project, it's a random GitHub comment.
5. App stores are more convenient, anyone making this claim is either being dishonest, or has no idea of the average level of technical ability of a mobile user.
6. This is a legitimate point and a blog post structured around this as the main point would be a much stronger argument.
7. Cool. But why not both?
8. Sure it's more work but like anything, it needs trading-off with the business opportunity of being on an app store.
9. ChatGPT is on the App Store and has been very successful there. Wordle famously got cloned by a number of app store devs who made a ton of money before getting called out for it.
The recent HEY example is where I checked out. Calling Apple a bully is fair, but DHH definitely set out to antagonize Apple, baiting the rejection and whining about it. An "everybody sucks here" situation.
I've been rejected on the App Store, only to have a discussion over email with the reviewers and come to an understanding that with minimal work resulted in approval. I've read many instances of this. Those discussions require good faith, going in and assuming bad intent or gaslighting the other party with regards to their motivations will go nowhere.
From his public persona, I have little confidence that DHH can play the necessary part in those discussions, and therefore I assume that he gets worse overall outcomes and his experiences are therefore not representative or useful to most.
So for me, I publish to the Play Store to get some free users who find my app that way. App Store, not worth the headache of trying to skirt Apple's rules.
Ideally I think app stores should allow listing PWAs directly, rather than banning them (like Apple) or forcing them to use a wrapper (like Google). Whatever complaints people have about that (app can change without review, etc) also apply to the current situation in the Play Store. Once your wrapper PWA is accepted, you can change the website however you want. So what's the point of the wrapper?