Android App Devs now require 20 people to test before publishing to Play Store | Hacker News Reader

Android App Devs now require 20 people to test before publishing to Play Store

techcrunch.com

284 pointsrobertwt7194 comments

As the article says, it’s likely that the devs who previously would test their apps prior to release had higher quality apps just because they were experienced and skilled devs, not because of an arbitrary number of people testing the app for an arbitrary length of time.

If Google would want to get rid of shovelware on the Play Store, they know what to do - (edit: better) automated + human reviews, like Apple. That would be what would nip shovelware in the bud.

But they’re just trying to outsource the cost of that. It feels like one of those Google initiatives that was started by someone who cared, then got watered down to the point of counter-productivity by MBA logic.

To get rid of shovelware, Google should just add a search filter that allows filtering for "No In-App Purchases" and "Doesnt show ads" plus a "Costs money (no subscription)". But something tells me thats only good for the user, not the business.

That will filter out a lot of high quality apps made by single person dev teams that include the option for users to support them through in-app purchases.

That would effectively make it impossible for single/small dev teams to justify making apps for the Play store.

And I don't see why filter paid apps would solve anything. The aversion of the average Play store user to spending money on anything is in part the reason we are in this mess.

12 continuous years working as an Android developer and I've never bothered to personally release an app on the Play store. There is nothing to be gained from it except needing to support it without any hope of making back the cost of developing it.

Ads, I however agree with, are the bane of our time. Except, due to the aversion to paying for apps, it has become the only way small dev teams can make a living.

The change needs to happen on the consumer side for this one. Buy the apps that you enjoy using.

I'm not bothering to make open source apps for Android because they'll be inevitably flooded out by 1000 ad- and microtransaction-infested variants, and I'm not even allowed to distinguish myself from them with my primary feature, "free of that garbage" (IIRC you can't even mention it in the app description, definitely not in the name).

There are enough people happy to just make and maintain a cool thing, but I can't find their work.

F-Droid is what you're looking for. Both as a developer and as a user.

I hate "this!" comments, but F-Droid is the thing that keeps me an Android user and makes me a die-hard proponent of alternative app stores (and why I'd use an iPhone if I could get something like F-Droid (with Termux) on it). Even phones with no other app store can install F-Droid, and it has enough good software that I can do most everything I need with only F-Droid.

    - FFUpdater for installing browsers (Mull on my phone, Brave on my tablet because keyboard support)
    - Termux for all my CLI work
    - KeepassDX for passwords
    - Syncthing for syncing
    - Etar for calendaring
    - VLC for media
    - Kore to control my media center
    - RCX for running local webdav servers (good for hosting Tiddlywiki and FeatherWiki locally on-device with WebDAV-based saving)
    - Tiddloid for notetaking when I'm not using RCX instead
    - Conversations for XMPP chat with family/friends
    - KOReader for ebooks
    - MullvadVPN for ... VPN
    - OsmAnd~ for mapping, including offline maps
    - "Puzzles" and Unciv for gaming
    - Trail Sense for navigation/survival

I know this is a laundry list, but my point is that it's a trove of exactly the sort software I like to develop and use. Thanks for mentioning it!

You might like to know that RCX has a new, maintained fork in RoundSync [0].

If you're just running it as a local webdav server, you can just stick with the current version. But if you also use it for the more typical rclone tasks (keeping cloud storage synchronised), RoundSync is updated and fixes some things that were missing in RCX, like cronjobs.

[0] https://github.com/newhinton/Round-Sync

I wish there was a desktop equivalent of this app, time to fork RcloneBrowser ;)

I've considered it, but I'm not sure if the audience size is worth it and the store used to be so broken that I don't know if it wouldn't be even more frustrating to work with them than it is to work with Google Play.

In particular, I've consistently heard that it takes them forever to update app versions. At least https://gitlab.com/fdroid/fdroidclient/-/issues/2316 seems to be fixed now - it used to require user interaction for each update, which doesn't work when you have dozens of apps (and updating from the notification also had some kind of bug requiring extra manual steps).

Maybe I should give them another try, how is the developer experience nowadays?

I think you just argued that we should go back to shareware. Free to download and then a one-time payment to unlock all functionality. That's also easy to support officially in the store and it's easy to add a filter for.

The problem with allowing IAP to be used for a shareware-like unlock is that there's nothing stopping shady developers from forcing a subscription on you later even though you already paid now. Forgot which one that was, but this kind of double dipping recently happened to a popular baby monitor.

Quiz Duel (MAG Interactive) did this.

Partner wanted to play all the time. I discovered the videos ads were using up all of my prepaid data before the end of each month. I bought the premium version, which was a one-time purchase "back then".

They launched an app revision that had a handful of very noticeable bugs. Then they changed to a new app (rather than upgrade), pushed all the users to the new app, and made the ad-free version a subscription. What do you get for your monthly payment? Well, nothing significant, like you'd expect from a subscription.

In fact, they essentially crowd-sourced their Q&A content by letting users submit feedback when a pair was inaccurate, so users are fixing problems in the content for free.

Server usage?

Not to minimize the daily work of countless devs, but a quiz game is essentially a handful of databases, a dollop of quite straightforward logic, some static UI visuals, and then content scraping to fill the main Q&A database. The logic and UI visuals almost never need a rewrite. DynamoDB is pennies per month per user, and if you're a real POS you can use long advertisements to push away users temporarily when your RCUs/WCUs are getting close to a scale-up threshold, which has the nice side effect of increasing revenue if everyone watches the long adverts anyway.

I guess they expected a percentage of users to bail switching to subscription, and that doesn't cost them a cent. You said it best: shady

To be fair, even if the payment is IAP, the publisher can still attempt double dipping, by first unpublishing the original app (which to be fair, while user-hostile, has the benefit that they don't have to bother supporting different/new hardware)

See this 8y old thread about what EA does on Android: https://old.reddit.com/r/OutOfTheLoop/comments/3mikbc/whats_...

s/even if the payment is IAP/even if the payment is not IAP/

(Too late to edit)

That's how drawing apps like Concept and Infinite Painter are priced. Free download, no sign up, several good features in the basic version.

There are IAPs to unlock brushes and other things, and a small fee to buy everything. I believe I paid $15 one-time to get everything for Infinite Painter.

We are already there, that is basically what open core is all about, a new marketing name for a generation born in FOSS.

> That will filter out a lot of high quality apps made by single person dev teams

Maybe.

On the other hand, when I see an app that includes ads or in-app purchases, I assume it's a low-quality app by default. Such a filter would be useful to me and wouldn't change my selection process. It would make it much less onerous over having to "manually" filter those apps out (which would make the app store much more useful to me).

I prefer apps that just charge me up front rather than use in-app purchases or ads.

Integrating ads with your app isn't merely trying to manipulate your users in a way that is liable to exploit, harm, disadvantage, or oppose their interests. It is virtually certainly compromising their privacy in a fashion that isn't immediately obvious and opening up a security hole that may be used to harm them further.

The first and most obvious clue that what you are considering is wrong is the fact that you are admitting that if people were presented with a choice they would overwhelmingly opt out of such a system so pervasively that there wouldn't be any money in it at all.

> That would effectively make it impossible for single/small dev teams to justify making apps for the Play store.

This is an inversion of priorities. The dev's profession exits to create useful things for users. The users don't exist to enable the devs profession. If the only way an app can come into existence is through the exploitation and disadvantage of its users then the world would be better off if it didn't exist.

And lo there is another way an app can come into existence: through paying users. Which they don't. The users prefer to pay with their "exploitation" so we have today what we all (as a society) requested.

To me it’s kind of wild we call such products “free”. Like YouTube claims to be free, but it’s ToS prohibit use unless the user pays with their headspace and attention to ads (and really awful ones to boot). If you use an ad blocker and refuse to pay, it will now tell you you’re not allowed to use it. So it’s “free”, but you need to pay to use it. And they will try to enforce it. But don’t overthink it, and repeat after me - “it’s free™” (so long as the dues are paid)

Convincing everyone that content online is free while taking payment for it is one of the greatest sleights of hand of the century. That kind of thing used to be called a con.

> This is an inversion of priorities. The dev's profession exits to create useful things for users. The users don't exist to enable the devs profession. If the only way an app can come into existence is through the exploitation and disadvantage of its users then the world would be better off if it didn't exist.

That statement just oozes entitlement...

If that seems hard to chew on try this on for size. Food doesn't exist to demonstrate the art of the chef. Without someone to eat it becomes rot. This doesn't degrade the chef nor his art it defines it.

You just need to change your filter rule such that a single iap to upgrade from free to premium version, can be considered similar to a paid app with no iap.

> single person dev teams

not to put too fine a point on it, but does a single person really constitute a team?

“One person team” is a fairly common phrased to describe such a situation I believe. The “team” is the group of developers responsible for something, which may sometimes only contain one person.

yea, sets with one or zero members exist.

thou, one person does not constitute a team, sorry, not sorry.

You're technically correct, which, as we know, is the best (/s) kind of correct.

What matters is actual usage, and while "one person team" does grate on me somewhat, it's in wide usage, and it's understood what it means. So, "sorry, not sorry", but you're wrong.

Words mean whatever the standard usage of them is. You can’t just declare a word to have a certain meaning. This is not my opinion - this is what scholars of language believe about language. The meaning of words changes over time as society finds other uses for them.

https://ludwig.guru/s/one+person+team

> And I don't see why filter paid apps would solve anything. The aversion of the average Play store user to spending money on anything is in part the reason we are in this mess

If people can find apps that meet their needs which are also free, and free of ads and in-app purchases they should absolutely use those apps. Making it harder for users to find those apps is a much bigger problem than "but I wanted someone to pay me for giving them the same thing!"

The solution for people who want money is to make apps that offer valuable features the free apps don't have, not to make the free apps impossible to find because they're scared they can't compete.

> 12 continuous years working as an Android developer and I've never bothered to personally release an app on the Play store. There is nothing to be gained from it except needing to support it without any hope of making back the cost of developing it.

As a high schooler it was pretty cool to publish a small game onto the play store and have friends download it from there.

Easily solveable by offering the same app for free with IAP (or as shareware), and the same app for a price as a one-time purchase.

Thanks to sandboxing, the paid version would not be able to access data created by the free version, when trialing.

To avoid this problem in the past, authors offered the free version, and then paid unlock as a separate app package. So you could continue to use the original application you trialed, and the unlock package would turn it into the full version.

Unfortunately, this stopped when in-app purchases appeared.

It feels like an IAP would be more fitting than an entirely separate app, though -- it'd be a shame to require a new listing just to work around this filtering.

Maybe static analysis could determine whether the IAP unlocked a new feature/functionality or not, feeling into the labelling/filtering?

What does this solve?

The developer now has two apps on the play store, still with no paying customers for either of them. The aversion to paying for quality is still present.

You need F-Droid. It's not a silver bullet but the high ranked apps on there are great.

Users hate paying upfront for apps precisely because the quality is unknown. It's the Akerloff "market for lemons" paper in real life.

App stores offer refunds.

It was better when apps had a fixed price like in the early 2010s. At some point it was decided that subscription apps were where the money was at, both for devs and for Apple who took a silent cut every month.

I would challenge that. iOS Appstore is way more known to charge for apps and have less IAP/Ad financed App. Also, good examples are Steam or other Game Stores - but in order for that to work, you need a refund policy (i.e. have a 30minutes refund period). For Apps with one-time use that can't have a return policy, there could be separate indicators/warnings during purchase.

iOS apps are better because you have to pay Apple for a developer license.

Do these apps even exist anymore? I feel like they never really did on Android and are a dieing breed on iOS.

Of course them being hard to find on the store, is part of why they don't exist in the first place. Fixing that would help.

Don't try to find that needle in the haystack of a million-app Play Store. F-droid is all you need.

I'd love a filter like that, I'd totally pay once for apps if they just worked and didn't have extra bloat, subscriptions, IAPs, etc.

They could also start ordering search results by relevancy or quality instead of propensity to make money, but that won't happen either

It is so bad that in the official play store app ads potentially for competitors or irrelevant apps will appear before the item even when you search for the entire name of the app. This means that devs are incentivized to buy ads for their own name lest their competitor be listed first and if they don't you might well download malware.

On net the play store is 100% useless because of ads and poor search. The only way to adequately search for apps is by using an actual search engine, even googles, in your mobile browser to find apps and then using the play store solely for installation.

The play store is a product so bad if it wasn't required to install software it would have no users. It is the windows ME of google's product lines. It is so bad that Google ought to be embarrassed to show it off in public.

I don't really understand how this could stop shovelware. Shovelware devs have the bandwidth to get something tested on 20 devices. This would only stop one man passion projects.

I wonder if its just a way to sell a cloud build and testing service.

They do have automated + manual reviews. I think the $99/yr developer account is what separates Google from Apple here, along with much stricter policy enforcement on the Apple side.

I have to think that $99 a year keeps away a lot of people who aren’t serious, and maybe even discourages various scams and schemes that depend on flooding platforms with apps scammers know will get rejected eventually in a whack-a-mole fashion.

If your account gets banned and you lose $99 before the scam paid off it’s not worth your time.

There are multimillion dollar scams on the App Store[1]:

> That man’s name is Kosta Eleftheriou, and over the past few months, he’s made a convincing case that Apple is either uninterested or incompetent at stopping multimillion-dollar scams in its own App Store. He’s repeatedly found scam apps that prey on ordinary iPhone and iPad owners by luring them into a “free trial” of an app with seemingly thousands of fake 5-star reviews, only to charge them outrageous sums of money for a recurring subscription that many don’t understand how to cancel. “It’s a situation that most communities are blind to because of how Apple is essentially brainwashing people into believing the App Store is a trusted place,” he tells The Verge.

The fee is nothing.

[1] https://www.theverge.com/2021/2/8/22272849/apple-app-store-s...

I think you underestimate how much scamming pays. It’s much more than $99 per day, so unless Apple blocks the app immediately they’re still making money.

How about $99 to register plus $10,000 fine + 100% of proceeds of perfidy + termination for certain misbehavior + mandatory presence of someone company or individual present in a country willing to enforce such a provision.

If you want it to really stick make them buy a bond against in such a jurisdiction. After one party has been burned for their 10k they'll find their former partner exceedingly aggressive in recovering their money and none willing to do business.

Furthermore most of such abusive behavior is trivially recognizable based on dollar figures involved, behavior, organic reviews, complaints. In this day and age there isn't much excuse for a trillion dollar company not to have automated 99% of the situation and employ labor to handle the rest.

Can companies actually enforce "fines"? The closest I can think of would be a "breach of contract" penalty, but you'd still have to go to court of file small claims for that. I don't see Apple getting into any of that to improve the quality of the App Store.

But not all crummy behavior, right? Things like ad-infested basic apps might be discouraged even with the $99. I am talking about apps like tip calculators, daily quote apps, and very basic games like slots that have no chance of being monetized well beyond ads.

What is discouraged is the solo devs making their first apps and the solo devs in poorer countries (but Apple lost those markets anyways). The scams and the useless apps have no problem paying a single 99$ bill.

Will Apple stay pay out the scam money if they ban you immediately?

They don’t ban you immediately.

They could hold funds for new unknown developers for 60 days for good behavior. During this time if you get banned they should seize the entire proceeds to refund to their users.

This would be great but in practice they'll just use third-party ad networks, which pay regardless of Apple's decision. It'll definitely break those pesky inflated subscriptions though.

I didn’t know Google reviewed apps manually as well. I thought it was all automated as I’ve not heard of an app not passing. There are even some people who have spoken about automating making and publishing shovelware apps for a time.

It takes a bit of effort even to get an app on public TestFlight on Apple’s side. Plus the fee, which is reasonable for a thorough review.

Anyways, I edited my comment to not misrepresent the Play Store process, thanks.

I can assure you, Google does have people review apps.

Our app gets held up fairly regularly, I'm not sure why us in particular, but we have an app that requires our licensed service to use, quite often they'll muck up the registration process with the test accounts we provide (they're one-time) registration, and then hold up our app until we provide more test accounts for them to use.

They also recently took issue with some functionality of our app that they claim existed (but doesn't) and forced us to update our privacy policy to mention it so we can get it through. I know for a fact, said functionality doesn't exist, because I lead the team that builds the backend, but there was just no arguing with them.

It's mostly automated but then you will get someone every so often that holds up your release, can only speak by quoting the guidelines, make you decipher their cryptic quote, just to find out that it's about some rule change from six months ago you missed in an email.

It's even more fun when it's a purely administrative thing like updating information in a link on your developer profile, but they make you upload a near identical binary, since the automation has kicked in again and they can't rereview an already submitted and rejected app.

Google charges developers fees, as well, when it comes to publishing.

It's a one-time fee for the developer account and it's $25, which is quite different from a $99/yr charge to keep your apps live on the App Store.

> If Google would want to get rid of shovelware on the Play Store, they know what to do - (edit: better) automated + human reviews

They could also do something like a progressive tax, like per year (USD), based on apps submitted:

1 free; 2-5 apps $20 each; 6-10 apps $40 each; 10+ apps $100 each

Nobody is putting out 10 high quality apps per year ;)

Couldn't I just create two accounts?

At the moment, you can. I have two accounts (by mistake). However, I'm not sure what the purpose of keeping 2 accounts active simultaneously is ... if one is banned, the other will be too (so there is no point in safety by numbers).

In order for this progressive taxing scheme to work they would have to make sure each personal account is unique per person, and they are already increasing the requirements to verify your identification (all Android devs have to verify this information soon).

Two, maybe. 100, with much more trouble.

Google's attachment to "low touch" approach to support is maniacal at this point. There's no use confronting them about it, they'll never change

What a stupid requirement. I run an app builder and offer my users plenty of ways to test their app beforehand, but outside of Google Play, since their system is just cumbersome.

This will make it nearly impossible to get an app published as a small business, since who the hell has 20 users just waiting for your app? It's already a huge struggle to help people with publishing their app.

The road to hell is truly paved with good intentions. All the scam apps will have no problem cheating this system, but honest devs will struggle. I can only hope Google reconsiders this requirement, otherwise it might be worth it to push my users more towards alternative app stores. We already support the Amazon App Store and the Huawei App Gallery (which are a lot easier to publish apps in), but the user base is just not there unfortunately.

>since who the hell has 20 users just waiting for your app

And that's how test-as-a-service is born.

Already here. I work in that industry.

> who the hell has 20 users just waiting for your app?

I'd wager that the majority of apps published to app stores struggle to find 20 users total, period. Let alone before public release.

It really seems like an obviously counterproductive measure. I think you're totally right in that the people who will have the least issues passing these new requirements, are those who don't shy away from dishonesty.

Thought that i would atleast create an app that is useful, so spent 1 month developing it, their was more to it but i wanted to publish atleast 1 app. So spent $25 to create account , than spent 3-4 hours publishing app , waited 5 days to get released. Today it was showing " Closed testing " in app status. I was not aware of it. So searched on internet & read about Google's NEW REQUIREMENTS.

This month on 16- Nov i created account on it & that time i was not aware of this new policy!!!!!! From where the hell i am going to bring 20 testers now !!!!!

BRB, starting up AndroidAppDevTesting.io - faking test users to PlayStore-as-a-Service.

For just $29.95/month we will load your new app onto a fleet of 20 test Android devices and run your Appium or Selendroid test suites twice a day for 14 days.

you joke but this will happen.

I mean, it's kinda just BrowserStack or DeviceFarm and an Ansible script, right?

(I'd spend an evening or two setting up a landing page and duct taping that Ansible script onto my Stripe account, but I've got a personal policy of never ever trying to sell tools to developers...)

>I've got a personal policy of never ever trying to sell tools to developers...

It feels like there is a story there. Care to share?

Developers are some of the worst customers you'll ever have the displeasure of experiencing. We are incredibly cheap and price sensitive, have literally zero ethics around reasonable usage (if they didn't want me to do it they should have stopped me), we're loud and obnoxious with strongly held opinions, think the world revolves around us and our pet issues, and will work tirelessly to make sure you make no money. We're the Karen's of the digital age.

This is a self-callout, I'm guilty of all of these.

I've been selling dev tools and libraries for decades, and all of that exists (it exists with non-dev customers as well), but none of it is so awful that it makes addressing that market undesirable.

Cute but surely those devices, accounts and IPs will all be banned, along with the developer who used the service? I wouldn't play around with my account like that personally.

I'll be worth it to scammers. As so often with misguided policies, only the honest are punished.

I feel like this actually would be a weirdly therapeutic little side gig.

Why not resell testers from Amazon Turk with a hefty markup?

An unpopular opinion: app stores as a default means of app distribution need to die, yesterday. The intentions were good, but in the end, it's horrible for everyone — both the users and the developers. Especially for those outside of the US who don't share the American obsession with relentlessly protecting everyone from the very idea of sexuality, whether they want that protection or not. And the reviews are just never helpful to neither users nor developers. And apps begging for those reviews always feel pathetic.

Not likely. Both mobile platforms have app stores, windows has winget/chocolatey/..., macs are trying/have brew and linux has package managers since the 90s. It seems very much that this approach is winning over "side-loading" and for good reason.

Your complaint is somewhat orthogonal to this. It would be perfectly possible to just have multiple repositories/sources backing the same app store with different values of curation/censorship. Tick a box once to enable the sources and it is completely seamless. Of course, the store owners have little incentive to do that, since they earn a cut if you buy from their repo, but that is why you have to force "gatekeepers" (in the EU terminology). If they make it non-seamless, fine them and fine them again until they do.

I think you and GP are using different definitions of "App Store". GP seems to be taking issue with how walled gardens enable application censorship and erode user control. You counter that standardized application installation/management is good UX and compatible with models that allow anyone to spin up software repositories, which is true, but doesn't counter anything GP says.

GP is equating app stores with walled gardens. I am saying that this need not be the case and that you should keep the benefits while getting rid of the points he is criticizing.

> but doesn't counter anything GP says.

It says that razing the earth and getting rid of the whole concept is overkill and his criticism should not be of app stores as such but of "gatekeepers" abusing their position.

Package managers are not app stores. They are very different things. A package manager usually allows you to freely add non-default repositories — that alone makes it different enough.

They do have their own shortcomings, like how library packages work, but that's ultimately out of scope of this particular discussion. I didn't even have package managers in mind when I was writing that comment.

App stores are front-ends for managing/buying apps/packages, e.g.

winget install Netflix

will install netflix from the windows store.

Click on settings to enable new sources, e.g. join the beta channel. They are the same thing just graphical.

Not the same thing. You can't add a custom repository in App Store or Google Play. They only ever install apps from Apple's and Google's servers. Getting your app published on those servers is obviously subject to their owners' whims. Meanwhile, you can just run

    brew tap grishka/grishka

to add my personal repository to your homebrew configuration. Try doing that with mobile app stores. One might say that, ignoring some technical details, the functionality of package managers is a superset of the functionality of app stores.

> Not the same thing. You can't add a custom repository in App Store or Google Play.

Exactly the point. The problem is that you are not allowed to do that (not a problem of can't). As pointed out by people in other comments, alternative app stores on android totally do have that functionality. So the problem is not app stores as such but google and apple being gatekeepers. Give me the option to add fdroid repos in google play.

> Try doing that with mobile app stores

F-Droid?

Sigh. Mainstream mobile app stores.

You are conflating tools (apt, brew, pip) with stores (Play Store, Appstore, F-Droid).

Android already has the ability to allow a tool to install from different sources (this is how F-droid functions, however only from their own single source).

> however only from their own single source

F-Droid is actually both a tool as well as a repository (which is the one that's included by default in F-Droid the tool). You can add additional repositories to F-Droid the tool.

I honestly did not know that was possible. Thanks!

Sorry, are you saying that winget, chocolatey, and brew are equivalent to the mobile App Store? Because for the average user, that is not remotely true. The App Store is how you install programs on mobile by default. On desktop, however, the default installation process is to download an installer/executable from a website. Only very technically literate people (basically just software developers) will be installing things via winget, chocolatey, or brew

Since Google is blocked in China, the default process for installing Android applications there is also to download an .apk from the internet and install it manually.

It works without a hiccup on my phone running LineageOS. I haven't tried on something running stock Android.

But it leads me to suspect that there isn't much standing in the way of distributing your Android apps this way outside of China, either.

Sanctioned Russian banks distribute their apps as self-updating apks. It's much, MUCH worse with iOS — Tinkoff tried publishing an innocuous-looking app that turns into their usual bank app after you open a "secret" deeplink for example.

Two things are problematic, though: 1) app stores got normalized as a way to distribute apps to the point of people being afraid to sideload, and 2) the fact that you can't distribute an iOS app without Apple's permission is utterly bonkers.

> Android applications there is also to download an .apk from the internet and install it manually.

This is how I do it in the US whenever possible, too.

> Sorry, are you saying that

No.

> It seems very much that this approach is winning over "side-loading"

But it's a shame. I hate app stores, personally. Not just on smartphones, but everywhere else they've been popping up.

Web managed to survive and thrive without App Stores, so I agree with you. What we need is resilient implementation for applications, so it's safe to install anything, just like it's safe to browse anything. May be some curation is good for applications with very deep integration with OS, like file browsers or remote control. But for typical application which could work as a website, there's no reason to have any curation.

Modern mobile OSes (and macOS) are already resilient. An app can't access anything sensitive unless you give it a permission to do so.

> Web managed to survive and thrive without App Stores

Did it? It reached critical mass and started getting shittier, more complex and more closed.

Open is great whilst the majority of users are technical or at least competent in critical thinking. Once you pass that tipping point, protectionism is a necessity.

This sounds good but its really impractical in practice. Even well designed systems suffer from security holes. Telling dumb people they can install anything they like will always be a harmful pretense.

> Especially for those outside of the US who don't share the American obsession with relentlessly protecting everyone from the very idea of sexuality, whether they want that protection or not.

The first step would be to not have mobile telecom device OSes run by American advertising and media companies.

Unfortunately mobile Linux is limited to ultra-small vendors liek Purism and Pinephone.

90% of apps could just be websites, so developers also bear a lot of responsibility for jumping onto that bandwagon for little benefit

While I agree on the sexuality part of this and I do wish that wasn’t a problem on the App Store.

I disagree with this.

There are a lot of situations where an app is unnecessary. I feel like we have too many situations where it could just be a website.

But if I am going to download an app I prefer the App Store. Especially if it’s an app that I need to spend money on/with.

This sounds like another nail in the coffin of indie devs.

I have 2-3 apps. I’ve started with Android with eclipse.

Back then I was really enjoying how I could make a small utility and get it on devices quickly.

After my 3rd nexus just died I’ve switched to Apple walled garden and.. got back to developing for Apple eco-system some indie apps.

Since then, The amount of resources I’ve spent on just maintaining an existing app on Google play was so time consuming.

Let alone API and depreciation, the play store required me after all those years sooo much documents and procedures just to keep my account.

I’m not sure how Google got the play store experience to be so worse than Apple.

Exactly this, so much energy is needed just to keep things working as they were. It's like Google actively cultivates and accelerates the software rot. Personally I'm tired of it and I'm getting off the train with my app.

I'll also stop using the Play Store because this change clearly signals what's welcome there. They don't want to host quirky-but-useful indie apps with personality. They would rather have the polished addiction forming platforms for corporations. Worst thing, they dress it up as a favor to their users.

F-Droid looks good, though.

20 people is just a start. I expect it to ramp up and effectively lock out indie developers. This alone would've prevented stuff like Flappy Bird from ever being published.

I've had issues with my camera manufacturer's app for syncing photos on Android for years, so I finally built a super lightweight replacement which is faster and more reliable.[0] I used it for a year or so before finally going through the hassle of publishing it to the Play Store.

The app has a very limited audience. In the first month, I've had approximately 28 purchases. If I had to find 20 people to use the app for a couple weeks before I could publish, it wouldn't have been published.

[0] https://play.google.com/store/apps/details?id=me.johnmaguire...

The other question that has come up, is what happens when one of the 20 testers has their account banned? Is the guilt-by-association thing going to apply; would the developer account also get shut down?

Yeah, this is really arbitrary, when I first got started in mobile apps I had one or two users to start with, I didn't know anyone who would help, it was really specialised as it was, catering to a niche market, I just had an idea and I could publish it (to apple in my case).

It didn't do well, but I still sell copies every month, and getting an app on a store considerably improved job prospects after a somewhat drastic change in career direction.

This sucks.

Realistically there is not much reason for flappy bird to be on the store. It could have been a website.

What if I want to program in Java, not JavaScript? What if I don't want to learn how to work with PWA and instead just want have an easily offline-accessible app? What if I wanted to add in-app purchases?

There's plenty of reasons for apps like Flappy Bird to be on the store.

The universe really doesn't cater to the howls of "but I just want to!"

Fair enough.

What if I need to program in Java, not JavaScript (because it is a requirement of the project)? What if I need it to not be PWA (because some of the libraries are native-only)? What if I need to add in-app purchases (because it is our business model)?

Does changing "want to" "need to" make a difference? It shouldn't.

No not really. The universe doesn't care much for stupid business decisions either - the business tends to just fail.

Your project requirement, native-only library or business model are choices that can change, not 'needs'.

What is a need in your model of the universe?

Making payroll.

That goes for virtually every corporate app that this policy change benefits.

"Wanting to get paid for your work" is a reason.

Some programmers like the idea of making a profit. Apps facilitate this.

This would have prevented me from publishing my first app, which ended up getting me my first job and a great career.

Same, got my first Android dev job from my side projects. Don't think current state of app stores would let something like that happen easily.

Wow. Could you tell us more about your journey?

I bought a t-mobile g1 the first android phone. At the time my school's wifi needed you to log in every time you joined the network. I wrote an app that logged you in automatically. Solved a need I had, I then put in in the app store for $0.99 and sold a few hundred.

I went to a small company career fair for the free lunch. When they saw I had an app in the app store, I had immediate interest and got an internship.

Wow? It sounds like he created a portfolio piece. lol.

Well there goes any hope I had of publishing to the Play Store. I don't know 20 people with an Android phone. What's the option here? Beg for random people to install your beta app?

There will be discords for fellow devs to install each other's apps and possibly companies to facilitate it for a fee (e.g., $3 per install + 50c per day of "testing"). Although, if testers have to be in the same market as where the app is to be sold, this could get a bit more expensive.

Developing-world startup idea here. +90% of Android market share here in Argentina, and probably most South American countries (if not all).

> Beg for random people to install your beta app?

I wouldn't call it "begging", but essentially, yes, this is what I've always done.

It's definitely weird rule if I understand it correctly. If the app developer selects 20 people, what benefit does it offer.

That seems absurdly onerous for indie devs. I hope Apple doesn’t do anything as extreme.

I cannot imagine any sane person coming up with this as a solution to any problem other than "it is too easy as an indie developer to make an app, solve a problem people have, and make it big in the world". Seriously whose ass did this idea get pulled from, and how can this be justified?

Yeah this is just going to spawn a cottage industry on Fiverr with jobs for "20 people testing your Android app", it honestly reminds me of my county's bureaucrats needlessly adding hoops to a process (which is so they can collect bribes).

Even for a small company of up to 10 people, I doubt many have 20 testing user lying around ready and waiting. Either pay/recruit or everyone get their extended families

My guess is that 2 weeks and the usage pattern of 20 users gives Google enough data points to determine whether the app is just AI generated bot spam. Because I'm betting there's a lot of that at the moment.

Still seems like absurdly high bar for indie devs to clear.

Maybe that is the exact problem this is meant to solve. Maybe all these countless apps cost too much to host for google?

This lockdown of what used to be a relatively accessible app distribution system is sad. At least we have F-Droid :)

https://f-droid.org

Isn't this just going to create lots of dodgy services that offer a "test 20 apps and 20 people will test yours", that already exist for giving app 5 star reviews?

So glad I saw this on November 10 before the deadline. I quickly set up a Play Console account, verified my ID and set up a draft of my app to ensure I'd be in the clear.

Been working on an indie game for the last couple of years - would have been a huge pain trying to find 20 people with Android devices to meet this arbitrary threshold.

Ladies and gentleman, I present you the company, that prides itself on hiring only the best and the brightest!

To bypass that, make an org account and give them:

As organization’s name, type, size, detailed address, phone number, and website.

Wait what? So if you have a company, you don't need to test?

Edit: wow, that's how I'm reading it too. If you have a DUNS number, this will presumably not apply to you. So this isn't about quality, it's an anti-spam measure. Crazy.

I really don't understand why all the actually brilliant people at Google, Apple, etc, are never allowed to make user-facing decisions.

I wish this applied to existing apps I have to use at work which seem to be upgraded without any testing. Such gems as get an Android version upgrade and can't reinstall app; lose all location information on the app that provides maps of work to do; alerts stop working when graphics on the app are "improved". Sigh.

I don't think I even have 20 friends... nevermind 20 I could get to install my apps. guess I'm sticking to the cloud :(

What a terrible decision, how does that do anything other than fuck over small developers? The ones who want to publish "unsafe" apps will just cheat the system

How about they actually deal with the low quality, spammy apps that are already published?

I'll give a good example, VPN apps, specifically the ones published by "Country VPN LLC", they have 145 apps that are just titled copy-paste "(insert country) VPN", nor do they actually provide the VPN location they claim. And there's more just like that publisher, such as "Proxy Master & Turbo VPN" and "AltApps" both of which have 14 apps that are titled for different countries

https://play.google.com/store/apps/developer?id=Country+VPN+...

https://play.google.com/store/apps/developer?id=Proxy+Master...

https://play.google.com/store/apps/developer?id=AltApps

Has anybody considered this may not be a bug but a feature?

They might intend to get indie apps off the store. Maybe the maintenance is too much for them, and this is a lower backlash way to do it than outright banning indie apps. This way the indie devs leave on their own by becoming disgruntled.

You're paranoid, get help!

Only for new accounts

> "Without providing an exact timeframe, Google says that new developers with individual accounts (as opposed to new Organization accounts) will be required to test apps with 20 people or more for 2 weeks or longer before publishing to production"

I don't think that what's friends are for (to test your app). Meaning, you'll need to go to a paying service, making it harder to justify afterwards to make your app free without ads or in-app purchase.

I don't even have 20 friends. But if you/your team have 15 acquaintances who are willing, you can set up five Google accounts to test the app. After all, all you need for a Google account is a phone number. Which, at least in my country, can be acquired for free without much hassle.

This won't stop any app store spammers, if they can spam the app store it's no problem for them to get 20 fake devices and bypass this. Again it's the genuine indie developers getting punished.

I don't even know 20 people with Androids.

I dont even know 20 people who would install something for me.

I don't even know 20 people

I don’t even expect 20 users.

What if testing for launch is higher volume than post launch downloads?

Then Google doesn't care about your app and you.

In the apps I write, that very well may be the case.

From the official press release,

> ... developers with newly created personal Play Console accounts will soon be required to test their apps with at least 20 people for a minimum of two weeks before applying for access to production.

Does this mean that a developer has to recruit 20 users through a special workflow, or the users will simply see that they are 'testers' for a new app?

You add their email addresses to a list of tester accounts, then they have to install the app via the Play Store. So yes, developers need to recruit users (well, Google accounts).

I was about to say…isn’t this really easy to game? Is google checking IDs at the door?

Well that ends my app...

You guys know you can distribute your app outside Play Store right? Not saying that's a long term business plan for most, but if you are at the point you can't even get 20 people to install, world of mouth can clear at least this hurdle.

As for requirement itself, non-trivial apps that didn't get even this level of testing are likely to be quite broken. Will you really test on different form factors and Android versions / anticipate all user behavior? So I can see the case for protecting basic store reputation, in that anything you download is unlikely to crash right away or drain your battery etc.

What do you base this assumption on exactly? Back in the day I released my first app without even a physical Android device to countless happy users. In the enterprises I have professionally worked we didn't have 20 dedicated testers.

Exactly. As a startup asking the first 20 users some other way of installing the app which is not in play store doesn't give any confidence.

You can also set up your own repo for fdroid. With Android 12 and above, neostore can auto update apps. It actually works now.

And I believe Android 14 will bring even more updates here but I don't remember exactly what it is ..

Stop building apps. Just build WPAs.

Yeah, my response to this is that I'm glad I quit Android app development a while ago to focus on PWAs. Just hope for better PWA support from Apple/Safari one day, even though it is explicitly against their self-interest

Safari on iPhone even supports notifications now! Check out revolt chat. It looks and feels like a native app.

Apple made some really good progress in the latest MacOS and iOS releases. I have been using the "Add to Dock" feature in Safari for things like YT Music and Twitter/X. It's pretty neat and works well.

The native experience benefits the user and the device more because it's using the APIs as they were intended, WPA's will forever be the inferior choice.

20 is too many. How many of us have 20 friends we can corral into testing an Android app that isn't even on the store yet?

If we had 20 friends we would be in sales.

IT is getting more and more industrialized and I expect those moves will happen again and again.

Fuck Google. Every time I think they can't get worse. Looks like my app will be fdroid only.

In addition to developer verification, which will display your real name in Play Store, it seems like Google wants to kick all hobbyists off the platform. Anyone care to guess if malware count will drop sharply?

Hmm, this will probably create indie devs going around and begging people to test their apps. Finding 20 people is not easy if you are just one person having idea, skills to build the idea but not much else.

What happens to apps that are IoT/Bluetooth-only requiring hardware? Is demo mode enough?

Will there be detailed documentation other than a blogpost?

Horrible decision

It's a good thing Google accounts are hard to generate in bulk and Android simulators don't exist.

They have the tech to block 99% of those easily. The last 1% who can get around the block will have a nice source of revenue.

Google is basically acknowledging they don’t want and or care about portfolio/small utility/first time project apps on their store. This obviously isn’t a great look, especially when both companies nominally support educational and vocational efforts around app development. IMO, onus is on these companies to resolve the inherent contradiction here.

That said, it might be the right move. Despite some FUD in the comments that this will hinder indie devs or prevent the next Flappy Bird, such claims seem a bit exaggerated. Also, overcoming obstacles is part of the challenge of app development so if somehow this relatively low bar is the line too far…good riddance. Anecdotally, I’ve interviewed quite a few mobile developers with clear resume projects who then claim they’ve “published multiple apps to the [Play/App] store”; almost universally these candidates acknowledge without much pressing that said apps are unmaintained bitrot or even delisted due to some other unmet requirement anyhow.

I suppose the best argument is the same mentality that leads to spam on the store could be used to get around the 20 tester limit. Sure, but I’d expect that Google expects at least meaningful improvement even with malevolent actors, as that’s the new normal with such mitigations in any online marketplace these days.

Very annoying for indie devs

Android is great for reach, otherwise much more trouble than its worth

I cannot imagine using the play store or the apple store ever again. What hell. This is the true value of Android vs. iOS though: Google can do whatever corporate nonsense Google does, and I can live in my own parallel universe that works for me. Hopefully, Android stays open until other open alternatives emerge. But I think we have several more years before Google could feasibly pull back on AOSP even if they wanted to.

This is the last straw for me.

I know I cannot just move my business to some non-existent competition in Soviet America, but they don't control everything on the web just yet. Mail, storage, even ads can be had elsewhere. And this is now giving me the motivation to finally reduce my dependence on google.

If they can fuck over my app business just like that, they can do the same with ads, mail, storage or SAAS and it's a risk I cannot take.

If Google wants Android apps to have better quality, they need to take a step back and look at their mess of a platform.

UI Support? Completely broken. Apple has simple UI frameworks. You either use UIKit or SwiftUI. Very clear which one is being used in which file and the code won't compile if you mess up. No worry about color changes, design mismatch, etc. Everything looks the same and matches. They are compatible and interoperable. Begin with UIKit and you can slowly transition to SwiftUI if you want to. Android? Support library, material 2, material 3, androidx, jetpack compose, AppCompat, etc. I confused myself even trying to look this list up. wtf Google? Code compiles, you run. Either crashes right off the bat which is the lucky case (you are using component from material 3 but view theme is not a descendant of a material 3 theme), or if you're unlucky and it runs, half things are one design and color, the rest don't match? Why? Oh you forgot this one view uses a theme that is descendant of a different library. Off you go hunting down the style that component uses in 30 xml files (which btw, for some reason, are broken down based on dpi and api version too??) Good luck with that. 2 hours later you narrow it down, fix it, sigh of relief, then bam, it hits you back just like a window blind. The color matches but the button now looks different. Hmmm

Build system? Somehow this is even more broken than the ui support. Open project that used to build fine 1 month ago, hit build, it fails. Hmm what's wrong? Oh there is a gradle version mismatch between what the IDE has and what the project has downloaded in the main folder (wtf x2! Why is a binary for the build system living in the project source in the first place???) Shouldn't the android development tools included in the IDE already contain the last and best version of the build system that is backwards compatible? Imagine if Xcode asked you to download and store xcodebuild in the project source. It's ok. Update to new gradle version, hit build, it fails again. what now?? Oh the version for `com.android.tools.build:gradle` doesn't match! Weird, shouldn't android studio just be able to download and use the appropriate version of the android gradle plugin based on the gradle version you have specified (uhum I mean the one Android Studio itself should have included and you should never have to specify manually or download or store in your project source???) Oh and how could I forget our wonderful friend, the multidex! See, our build tools aren't even smart enough to be able to tell when your project needs to multi dex. It just errors out and you have to manually go ahead and enable it and include the library. To their credit, newer versions seem to have fixed this but this shouldn't have ever been even left to the developer.

Dependencies? This one isn't even debatable. It's yet another fragmented mess. Xcode dependency management is very simple. You either use their provided SwiftPM or you can use Cocoapods. Or both. They work well together. With android? Oh, Google does not want to commit a small storage/bandwidth from their massive capability to delivering good dependency hosting to developers. You pull the code from repo, dependencies can't be fetched! Wtf? Teammate swears it builds fine on their side (huh doesn't it always??). You check what's wrong! Oh Bintray/JCenter decided to shut down forever! Bye bye. They left and left you the generous parting gift of 10 unresolvable dependencies! You got hunting down the Github rabbit hole, issues galore. Everyone's freaking out. Project maintainer has left and hasn't pushed code to other dependency repos. You think: Good, the code is on Github, I should be able to just use the GitHub url directly in the dependencies instead, right? NO! Wrong!!! Things can't be that simple with android! You realize gradle dependency management doesn't allow you to just pull that code and use it unlike SwiftPM and Cocoapods. You are now at the mercy of yet another third party website (jitpack.io) to act as middle man between Github and gradle. This website shuts down any second and you are back to square one! Why god?? Why can't I just pull the god forsaken code and build it locally myself!? Did I mention that besides jitpack.io there's also maven central? What happens if that shuts down? No one knows!

I'm not even going to talk about the mess that is/was app signing and release, billing sdk updates, target sdk version deprecation, etc.

Fix your broken ecosystem first before blaming the developers! Why do the android apps for even bigger players like Tinder, Bumble, and Hinge have significantly lower rating than their iOS counterpart? They surely do test properly and can afford proper development. This just goes to show there is something really wrong with the android ecosystem and Google needs to look inwards to figure out how to get it out of the state its in right now.

I have all of these problems, and more (since I inherited a react native app). I DREAD having up do framework updates.

Maintaining this app for iOS is HALF the work that Android is, we have a lot less Android users, and in general it seems like iOS users are just a better audience.

It really feels like there's no point to doing an Android app, just give those users a website, PWA if they're lucky. And with this change? Pfffff.

And I agree - it does all descend from Google's horrible tooling and ecosystem.

The funny thing is Android Studio itself is now much more usable than XCode but as you say it is everything you are doing in it that is maddening.

What kills me about Android is the extent to which the official best practices have always been out of touch with reality and somehow get more so every time they add something new. The indoctrination in the Android dev community is so strong most people cannot separate the few core good bits of the platform from the layers of utter nonsense on top because they only know the nonsense.

20 seems pretty low. I'd be very, very nervous about releasing any software that only had 20 beta testers. I am disappointed to hear that there is such a problem with apps being inadequately tested.

But how is Google going to enforce this?

Reply on news.ycombinator.com