The main benefit of super apps, and why it is so popular in those countries, is that it doesn't require users to download anything to use new feature. In some use cases I also found that so convenient.
Just as an example, WeChat has an ecosystem of mini-apps. When you go to a restaurant in China, you can often find QR code on each table. If you can that QR code, it would open a mini-apps that lets you place orders and make payments, without waiting for the waiter to serve you. A mini app is like a website that doesn't require download and storage on the phone, but the user experience feels more native and is tightly integrated with features available in WeChat. It would be much harder to get users if the same thing requires you download an app.
As someone who's never been to China and has never used a super-app: just curious, what are some of the obstacles faced installing a mobile app? Is it that most people don't trust un-curated apps?
A super app that brings existing user bases solves both the deployment and discoverability problems.
Actually learning every new app is an obstacle, but they managed to learn to a few apps like WeChat because it's super useful. They won't spend time to learn about something that won't bring immediate reward. Once they have a few apps installed on their phone, they'd feel that's enough.
A nice, smooth, mobile-optimized website, I mean.
I can now go to a vendor's website (say jrwatkins.com, etc.) on my phone, checkout, do a fingerprint/face scan on Apple Pay and the transaction is done in under 5 seconds.
I assume the experience with Google Pay is similar.
There's no longer a need to create an account on the website. The delivery address defaults to the one on Apple Pay (and can be modified at time of payment). Delivery method can also be selected on the Apple Pay screen. My credit card number is never transmitted. This type of frictionless checkout that was once the domain of companies like Hoteltonight is now becoming more common on ecommerce sites.
I don't know what payment methods these services require now, but in the beginning at least, they both supports just credit cards, which is basically suicidal. Most users won't be able to use it at all.
From merchants' perspective, they don't need to pay for transactions on WeChat Pay or AliPay, whereas both Apple Pay and credit card companies would charge them for usage. For this reason most small business wouldn't even think of using them.
According to the Apple developer docs[1], websites don't have to integrate with Apple Pay specifically; Safari also supports the (fairly new) platform-neutral Payment Request API[2]. Google's docs[3] suggest that something similar is available for Chrome on Android.
1. https://developer.apple.com/documentation/apple_pay_on_the_w...
It supports Apple Cash, which in turn can funded via a bank debit card. (kind of like Venmo). There's no transaction fee with Apple Cash.
I believe there are currently no additional fees on the merchant side to accept Apple Pay [1], unless the customer uses a credit card via Apple Pay in which case the standard credit card fees apply.
We've never been big on mobile payments in the U.S. (because credit cards), but I'm finding myself using Apple Pay more and more. I read somewhere than ~50% of payments in the U.S. are now contactless.
I'm kind of thinking of markets outside China though.
I don't know about iOS but the two copy features from each other all the time so Apple probably has it too (if they weren't the first to launch them).
Honestly, if you're working from a QR code you might as well just serve a mobile web app. If you add PWA support your customers might even install it if they come around often.
This has to be prove by market. If the OS-ish super apps are winning, means the customers are approving it.
If people are not eating meat in North Korea, it doesn't necessarily mean they're supporting vegetarianism.
It is funny, that GMS itself is yet another layer built upon Android.
How is it different that Tencent distributing this layer via WeChat, other than Google bundles this layer through GMS?
Tencent does have its counterpart of GMS, like piece by piece, but that approach isn't winning against WeChat either.
- there are multiple play store equivalent, i.e. Mi Store, Baidu store, and several others.
There is actually more competition at app store level in China, as opposed to Google's monopoly on that front, which might be the reason why user behaviors are different.
- not sure how you define OS-like abstraction layer, but Chrome is also basically a super app. In fact, developing for Wechat is more similar to developing website than native app.
What do developers do for push messaging? In older versions of Android, you could maintain a connection to your server in the background, but recent versions make it really hard to stay active, even if you're big enough to get whitelisted by phone OEMs --- if you're small, is it even possible?
Looking at f-droid: hobbyists seem to do just fine without google play services. I feel like this gives you a business advantage rather than a technical one.
F-Droid is doing fine but mostly with a certain kind of apps that either run entirely on the device, or connect to third-party networks via more or less established protocols like IMAP, XMPP, OpenDAV, etc.
Google provides Cloud Messaging, Firebase, etc. free of charge. Everything can be worked around with enough money thrown at it of course but choices like this tip the scale in favor of going for a mini-app instead of a standalone one.
I have set up Android phones with Lineage for several relatives, and if the user is comfortable doing online banking through the browser instead of a bank-provided app, then there is really no need for Google Play services or the MicroG substitute. Whatsapp runs fine without Play services, and in my region OSM-based maps are a viable competitor to Google Maps.
I go searching and I can't figure out what some of the apps even "DO", others are "Here is this app, you need to install these other things too to make it work, also it only works with X, so maybe it will work for you I duno....".
Now I get why it is the way it is, I don't expect much more from folks putting out software for me like that, but it's a terrible user experience.
It's not good for app discovery. You already need to know what you are looking for. Other than that, most of the time it works fine.
In other words, it's a bit like Amazon.
This has very little to do with what I said (arguing some library doesn’t make things easier) but ok.
>I go searching and I can't figure out what some of the apps even "DO",
It’s been a couple years but I don’t remember the play store being different.
>Here is this app, you need to install these other things too to make it work
Like the Termux API app needing termux? That’s the only one I remember being like that. Most of the apps I used were self contained. Do you have an example?
From memory F-droid was much better than google play except for the lower selection but it has been a while.
What are you installing to make it a bad user experience?
I don't have google services (no google account) at all, only use F-Droid and Aurora store and haven't any problems. Even convinced a friend to go the same route with minimal handholding and they aren't technically inclined - no problems.
Google themselves has struggled in this market: see the Nexus line.
- There is no single Play Store equivalent, so the barrier to make people install a relatively unknown app is higher.
- Phones do not come with Google Services Framework, so the barrier to develop a networked app is higher.
- Just like everywhere else, people will use whatever is available because a suboptimal solution is still better than opting out completely from the benefits these solutions bring.
I don't think in China specifically it's about low-end devices or data cost at all. South East Asia - perhaps, but all in all the situation reflects the structure of the economies (the prevalence of monopolistic conglomerates with a finger in every pie).
Most of the time the most widespread solution is far from excellent because it only has to be better than the competition. Even more so when competition is limited by entry barriers to the point it hardly exists. So while the article is a good summary of the situation, it ends up being a bit like wondering why people use Windows.
Android is already one abstraction layer on top of another, there is no intrinsic technological benefit of adding yet another one OS-like super app on top of it, or at the very least any such benefits remain to be demonstrated.