Our app was banned because the button says “Report User” and not just “Report” | Hacker News Reader

Our app was banned because the button says “Report User” and not just “Report”

twitter.com

355 pointsigitur216 comments

For those not from South Africa, this app is primarily for people to be able to get notifications or look up when their area will be affected by a "planned power outage", or as Eskom (our state owned power utility) refers to it, "load shedding". "Load shedding" is enacted whenever they have a power production capacity issue where different areas take turns without power so as to not let the power grid collapse.

This app aggregates information from various data sources and provides push notifications too. It is without a doubt the best source of this information and the app being unavailable significantly affects the day to day lives of a significant portion of the South African population who use it to plan around these power outages.

While the app also has a "chat" feature, it's really tangental to the primary purpose of the app and I expect that most users of the app, like myself, don't use that feature at all and only care about knowing when they will be without electricity.

In case you're wondering why they hell we have load shedding, it's because Eskom is grossly incompetent. Their incompetence is hugely exacerbated by nepitism and corruption where government has historically appointed people to Eskom management positions solely as "favours", rather than on qualification for the job.

And here's a link to the app:

https://play.google.com/store/apps/details?id=com.ashwhale.s...

OMG. This is as bad or even worse than Texas or California...

It is far worse, but South Africa has an inept and corrupt government, I wonder what the excuse is for Texas?

I suspect the truth is, in the narrow domain of electric power distribution, Texas too has an inept government that’s beholden to certain vested interests in the garb of “free markets”.

This is a state that has fetishized “light touch regulation” so much that it, for most of the state, has its own grid to avoid dealing with Federal authorities. And repeatedly failed to take action to get power companies to winterize their equipment even though cold snaps in Texas are not unheard of — they’ve had disruption in 1989 and 2011, in addition to other near misses.

Let’s not even get into the wisdom of a market that allows Griddy to offer wholesale prices without caps to retail customers.

So hang on, why is it that South Africa is “corrupt”, but Texas is “beholden to certain vested interests”?

Did any one person say both of those things?

That's just more words for corrupt

The distinction-without-a-difference between these two phrases is exactly the point of the comment you're responding to.

...and presumably also the point of the comment it was responding to.

> I suspect the truth is, in the narrow domain of electric power distribution, Texas too has an inept government...

This is an absurd comparison. South Africa owns and runs its power utility. The result is that they struggle to provide electricity on an ongoing basis. According to Wikipedia, small business owners in South Africa said that load shedding was the number one challenge that they faced in Q1 of 2019.[0]

Texas does not own or run its utilities. They are privately run, and are given wide latitude by the state. The result is that the worst of the utilities are tragically only able to provide 99.95% uptime, with some power consumers experiencing outages for a few days per decade.

[0] https://en.wikipedia.org/wiki/South_African_energy_crisis

Yes, Texas in the realm of power distribution is probably a paradise by comparison with South Africa, but compared to its peers (other US states), Texas has a very non-resilient, or fragile, power distribution system that we know will "crash" when faced with a cold snap. Given that they have known this since at least 1989, I think it's not unfair to use the word "inept".

The giant power bills are just icing on the cake of bad policy-making.

Why would you think it is any different? ERCOT was recommended to protect its power generation capabilities from severe winter weather for over 10 years, yet the Republican leadership did nothing about it. That's either corruption or sheer ineptitude.

There is no comparison.

This sounds like a nightmare!

Eskom seems to have been around for a long time, but this problem seems to have kicked in a little over a decade ago.

Did something recently happen to Eskom to make them incompetent?

Corruption.

It's two-sided.

Apartheid South Africa's government focused on providing electricity mostly for the white minority, which it did well, but the majority of South African, especially in the rural regions, lived with no or unreliable power supply. Post 1994, with removal of sanctions, the economy opened up. There was a lot of international investment in South Africa. The economy grew strongly and the demand for electricity significantly increased.

The supply side was woefully mismanaged and didn't nearly keep up with the pace of the growth in demand. Corruption is definitely a key reason, as others here mention, with the effect being that power plants are not adequatly maintained and new planned power plants going far over budget and missing deadlines [1].

Besides corruption, there are some policitical constraints. Eskom provides power to municipalities who often don't pay. Soweto, with 1.3 million residents owe Eskom in excess of $1 billion [2]. The South African Supreme Court of Appeal has ruled that Eskom is not allowed to cut power to non-paying municipalities [3]. In contrast to a free-market system, Eskom price hikes are subject to regulatory approval [4].

One could debate the relative significance of each of these factors, but it's by no means due to only a single one of these.

[1] https://www.news24.com/fin24/Budget/how-medupi-and-kusile-ar...

[2] https://mg.co.za/article/2019-10-18-00-why-we-dont-pay-for-p...

[3] https://www.iol.co.za/pretoria-news/news/eskom-cannot-cut-el...

[4] https://www.moneyweb.co.za/news/companies-and-deals/electric...

I think you are missing the most important part. Even after Apartheid, the existing utilities provided for the entire population. However, there has not been any new generation capacity since 1994. Instead, the maximum generation capacity has shrunk year on year as boilers and other key parts of the power plants fail and aren't replaced.

I'm not OP - that they explicitly mentioned this:

>> The supply side was woefully mismanaged and didn't nearly keep up with the pace of the growth in demand.

> However, there has not been any new generation capacity since 1994.

That's not true. Majuba Power Station was commissioned between 1006 and 2001, although planning started in the Apartheid era. Gourikwa Power Station and Newcastle Cogeneration Plant were commissioned in 2007.

There are a few wind farms built after 2001: Hopefield, Coega, Darling, Klipheuwel, Dassiesklip Wind Energy Facility and Jeffreys Bay.

And solar: Kalkbult, Mulilo Renewable Energy Solar PV Prieska, De Aar Solar PV, Mulilo Renewable Energy Solar PV De Aar, SA Mainstream Renewable Power Droogfontein, Letsatsi Solar Park, Lesedi PV Project, Kathu Solar Energy Facility 1, Sishen Solar Facility and the Jasper Solar Energy Project.

But these were not enough to keep up with the demand.

Majuba Power Station was commissioned between 1996* and 2001

> Did something recently happen to Eskom to make them incompetent?

Demand for electricity consistently went up over the past 20+ years, with no matching increase in generation capacity.

Maintenance of current power plants hasn't been great either, and the supply/demand situation is precarious such that if any one of the power generation plant is offline for maintenance, there are likely to be compensatory blackouts somewhere.

Load shedding is the correct technical term for it. It's not just a South African thing either, industrial consumers in Europe have the same stipulation in their power contracts: in case of emergency, the power company (network operator) may choose to suspend power delivery to protect the grid.

However, in the EU this 1) isn't a regular occurrence, and 2) the grid has a tiered system for load shedding. I can't find a reference for the tier classifications right now, but if I remember correctly: tier 1 consumers (the ones who will be disconnected first) are heavy industries with huge power draw, tier 2 are other commercial uses, tier 3 is residential, and tier 4 is critical infrastructure (emergency services).

I don't think I've ever experienced a load shedding event that affected residential areas. There have been local blackouts but I don't recall any grid-wide emergency events.

Texas had tier 3 level of load shedding recently.

Based on the incompetent decisions that ERCOT has made over the past few "severe weather events", I would say ERCOT and ESKOM are eerily similar.

All my reading said that the decisions ERCOT made during the weather event were all very competent. They kept the grid from collapsing, which was their job.

The fact that extensive load shedding and brownouts etc needed to happen was not a result of decisions made during the weather event, but lack of infrastructure investment over decades (that may or may not have been under the control of ERCOT, rather than the legislature and state government regulators, but ERCOT certainly was on board with/lobbying for the conditions of deregulation that led to them).

During the weather event... ERCOT, the grid operator, did their job very competently. Making more capacity come online or reducing consumer demand was not really something that could be done during the weather event. They kept the grid from collapsing -- and it nearly did collapse, it was a close call. That was their job during the event.

> All my reading said that the decisions ERCOT made during the weather event were all very competent. They kept the grid from collapsing, which was their job.

I respectfully disagree. Demand might have been twice what supply was, but the way the load shedding was done wasn't smart at all. The rolling blackouts were done on a roughly 18-48 hour basis, i.e. if you lost power you probably lost it for at least 18 hours and I know plenty of people who lost it for much longer, up to 48 hours.

If demand was double what the supply was then it should have been possible to give people power for say 2 hours every 4 hours. Or maybe 6 hours every 12 hours. If demand outstripped supply by a factor of 4 then 3 hours every 12 hours.

But that's not what happened. A lot of houses have gas furnaces but they need electricity to run the fans and the ignition system and such. People died in their houses either from the cold or from running their cars in their garages trying to stay warm. A lot of pipes were burst due to not enough heat in the buildings.

I can think of at least a dozen ways in which the load shedding could have been better managed. I wouldn't agree that it was very competent.

> If demand was double what the supply was then it should have been possible to give people power for say 2 hours every 4 hours

The problem is if that electricity is being used for heat, the draw of the 2 hour on periods will increase to make up for the time off. Perhaps even resulting in more total usage, because when power is on, people will set the thermostat to 80 instead of 65, anticipating the power going off and it getting colder. When really you need everyone's expectations of comfort to change to 40 degrees (half the delta T).

The only way to win and supply some power to all houses is to keep the duty cycle so low that the amount of resistive heating equipment becomes the limiting factor (1 hour every 6 hours?). I don't know if they could have managed this, but it's certainly more complicated to figure out than simply lowering the period of the rolling blackouts.

Only so much of the load is heating, and heaters can only draw so much current.

Even 1 hour every few would stop food from rotting and let people do important tasks.

> Even 1 hour every few would stop food from rotting

I just put my food outside, in the car. I never lost power, but I didn't want to waste any electricity on cooling food when it was needed so badly.

ERCOT failed to adjust pricing appropriately after the initial emergency, which allowed energy producers to milk the situation for billions of dollars more than is fair or reasonable. It's not like their input costs went up by billions of dollars that required compensation.

> All my reading said

Comparing that to "All my expierincing" first hand. Not even close to the same thing. They would not have been in the stressed situtation in the first place had they enforced the recommended corrections for cold weather protection in the first place. The fact that they had to turn off power generation because pipes were freezing at the generating stations is an absolute joke, and is the prime cause of the stress.

The local weather forecasts remind people to move plants indoors, cover exposed plumbing, etc. Maybe they should also start including a friendly reminder to the local power plants to also cover exposed plumbing, but why would they listen to that when they've ignored government sanctioned reports from investigations into previous failures?

> ERCOT, the grid operator, did their job very competently "Turn off the power so the demand goes away" Check "Issue rolling black outs so that the demand is lowered and manageable, and give people a fighting chance" Nope. These were absolutely not rolling black outs, but static black outs. The method they chose to "save the grid" was just as incompetent as not maintaining their systems in the first place.

Thank you for more info!

Do you know of any good journalistic coverage of this, since apparently I've been reading bad stuff?

It seems important info for those who don't have first-hand professional experience to get.

I read articles from every where. Local stations had coverage of the damage and lack of "rolling" black outs, national outlets had coverage as well, and I'm pretty sure I remember seeing international coverage from BBC, Daily Mail, etc. I can't remember where I was seeing the scathing reports on ERCOT's ineptitude, but they were plenty.

https://www.texastribune.org/2021/02/17/texas-power-grid-fai... https://www.dailymail.co.uk/news/article-9270343/Texas-perfe...

just a snippet of a quick search from soooo many articles

American here--I've experienced load shedding once. The problem was a lineman started a wildfire that took out the primary source of power for the town. Until the line was repaired there was nowhere near enough power available, they directed what they had to things like the hospital etc and everyone else was in the dark.

There is also the mess we currently have in California. High winds blew stuff into high power lines and started a bad wildfire. In the lawsuits that followed the utility was told to shut down the risky lines under high wind conditions--which of course causes massive outages. Of course the utility is "at fault" because the ecology types blocked efforts to keep the vegetation away from the lines.

The utility is at fault because they didn't spend enough money on maintenance. As a for-profit company in an established industry, PG&E won't ever be able to "pop" like a SaaS company does, so profits and salaries for executives and shareholders have to come from somewhere. Spending less on maintenance won't bite you - until it does - but until then, thats money that can be put elsewhere like CEO pockets.

PG&E is at fault because they refused to properly maintain the power lines, and lines that had been hanging for years from the same supports where the supports failed are the cause of the fires.

See here: https://www.nbcbayarea.com/news/local/long-term-wear-found-o...

https://www.kqed.org/news/11792217/1987-report-suggested-pge...

I can't find it right now, but there was an absolutely fantastic RCA report that included all the information on the wearing down of the hooks and how this could have all been prevented.

The better technical term for what South Africa has is rolling black-outs that increase in frequency (and ultimately duration) depending on the amount of load that needs to be shed.

South Africa also has actual load shedding, where industrial customers are asked then forced to reduce usage, but the black-outs are entirely different. They are region-by-region complete loss of power for all but the hospitals (but often take out hospitals and emergency services too).

Compounding this issue, the grid is not designed to be switched on and off frequently. While there is a set schedule for the rolling blackouts, transformers and other equipment will blow under the sudden loads, meaning your regional blackout may last a lot longer than planned. You may also get blackouts sooner or completely out of schedule for region, for which no explanation is given.

> industrial consumers in Europe have the same stipulation in their power contracts: in case of emergency, the power company (network operator) may choose to suspend power delivery to protect the grid.

You can't really compare.

Industries can chose in their contract to be part of the load shedding system. It's done with their agreement and the system is purely based on the grid frequency. The disconnection is done on site in an orderly fashion. In a way, it's probably more correct to view it as industries electing to temporary disconnect rather than the provider stopping supplying them.

> I don't recall any grid-wide emergency events.

The European electrical grid is extremely resilient. It was accidently split in two a couple months ago and that was barely noticeable for the customers of either half.

This app is also known for its humorous take the quite serious subject matter. The WHAT’S NEW section in the Google Play Store is known for sarcastic titbits with the fictional Jeff working hard to keep us up to date with the imminent darkness. The app's name itself, Eskom se Push, is wordplay on an extremely vulgar South African insult and feeds into the popular dislike of Eskom.

Unfortunately, an under-moderated chat feature is considered a risk these days, like an open email relay or an open proxy. I personally think Google erred too far on the side of caution here, but developers should be aware that chat features aren't as cheap to add as they used to be; they change the risk model of the app.

This. Another example: https://www.theverge.com/2021/2/22/22295496/malaysiakini-gui...

> Malaysian news site fined $124,000 for five reader comments

> Malaysiakini’s co-founder avoided jail time

IMHO a lot of sites and apps should just remove chat/comment functionality and only allow emoji reactions.

The company I work for has forbidden the use of Google and their services such as GCP etc. due to how they treat their Play Store developers and other customer, in particular that there never seems to be any human being that you can talk to and find out what you need to do to fix the situation. We do not want the same to occur to our servers or if there is an overflow from Play Store ban to GCP etc. The business risk is too high when you rely on Google's services.

Some examples:

Terraria banned - https://twitter.com/Demilogic/status/1358661840402845696

New project banned - https://medium.com/@amton15127/why-you-should-not-use-fireba...

Google bans company - https://www.reddit.com/r/tifu/comments/8kvias/tifu_by_gettin...

Google bans mail - https://talkingpointsmemo.com/edblog/a-serf-on-googles-farm

Ban app for communicating changes during covid - https://news.ycombinator.com/item?id=23221447

Adwords ban - https://news.ycombinator.com/item?id=23224791

Serverpunch bad support - https://news.ycombinator.com/item?id=17431609

Delete app - https://news.ycombinator.com/item?id=20826618

Google bans game with pandemic - https://news.ycombinator.com/item?id=23229073

Google bans dev with no recourse - https://news.ycombinator.com/item?id=15197357

Another good example of an awful implementation can be seen in Google Workspace. Sometimes the crappy ML algorithms will generate a false positive and suspend an account for a suspicious login. Fine. I get it. It’s hard to do at scale and nothing’s perfect.

The thing I’d like to have Google explain to me is why they think it’s a good idea to bounce incoming mail for a user that’s been auto-suspended by an algorithm. In what scenario would I want that, especially when the account is locked at 3:00 AM local time?

That’s something that actually happened to me this weekend. For anyone at Google, NO ONE wants their incoming email bounced because of your crappy ML algorithms.

Same thing happened to me yesterday, I had to switch my MX records over to fastmail as that was easier than trying to figure out how to resolve the issue with Google directly.

I had been forwarding all of my email from my domain to my personal Gmail (the irony) so haven't lost anything except a Sundays worth of incoming emails which bounced.

In my case I was still using a free account for my custom domain, grandfathered in. I guess I get what I paid for...

Was it your admin account that got locked? That's a big fear of mine. If so, did you have any of the account recovery options set up? Ex: 2FA, phone number, recovery email, etc.?

IMHO a lot of account "security" is intentionally made to be overly aggressive so tech companies can lock your account and verify your ID / get your phone number. If you're using anything for business the only option is to diligently set up all the recovery options that are available, even though that means giving them a ton of personal information.

Yeah, locked my primary account and the admin account. There was no option to use any recovery options that I could see.

And today I get an email saying my accounts are now enabled?

The computers are definitely in control.

Glad I was forced to move away at this point, locking accounts is one thing, but bouncing emails not cool.

It's rife in the industry. The drama this week is that my wife can't log in to Facebook because their code generator just doesn't work. The number in the app shows up, she types it, but she can't log in because apparently it's just wrong.

Due to their marvelous design, you need a personal account to run an advertising account for your place of work. Luckily she hasn't been logged out on her work laptop yet or she couldn't do a large part of her job.

Business support? Yeah good luck, try googling and you'll get pages of dead links or suggestions to click something when you're already logged in.

If you haven’t already, check the time / date / timezone on the device that is generating the codes. Facebook uses TOTP for this, which is time-based. If the time on your device is wrong, the codes will be wrong. I would expect this problem to pop up more frequently this time of year due to errors relating to daylight saving time.

DST has nothing to do with TOTP. The TOTP spec specifies that unix time has to be used as the time source, which does not have DST.

That doesn't matter, DST-related events can still break TOTP.

Example: Device is set to the wrong timezone. It’s nearby, but the difference between the two timezones is that one has DST and one doesn’t. DST comes along, either the time changes by an hour when it isn’t supposed to, or it doesn’t change when it is supposed to. The most visible thing that looks broken to the user is not the timezone, but the time. So they adjust the time. The user-visible clock is now “correct” for the next six months as far as the user sees, but the system clock – including UNIX timestamps – is incorrect by an hour. This results in broken TOTP.

I find this awfully thin. About every device that does TOTP by default takes its time from the network provider, and lets the user set the correct timezone, and makes it even really easy to do so. Since most DST settings for timezones in the world are quite predictable (yes I am aware that there are few that have last-minute changes, but most don't, especially not in developed nations), most people never have to adjust their timezone because of DST.

Though I do suspect that Google Authenticator even has some logic to reduce these problems, since you can "sync" its time in the settings.

“Lets the user set the correct timezone” is where this falls down. The user doesn’t always get this right. Once that happens, the time looks correct for a while, then looks wrong all of a sudden. So they go into their settings and manually change the time. At that point taking time from the network provider no longer happens, because they overrode that.

The predictability of DST doesn’t help, and “most people never have to adjust their timezone” isn’t relevant either. Perhaps you didn’t fully grasp what was happening in the example? All of the machinery works correctly, it’s just the first time they set up the computer the timezone was wrong, or they moved. Everything after that point can work correctly, but if that thing is wrong, UNIX time will be wrong even if the time looks right to the user.

I’ve literally had to fix this problem for people. “Correct” time, incorrect timezone. It’s not a theoretical example.

> check the time / date / timezone on the device that is generating the codes.

Sounds like that'd be on the FB infrastructure side though?

The clocks on both devices (i.e. fb's server and your phone) need to be fairly closely synchronised. TOTP effectively encrypts a timestamp using a shared secret then decrypts it on the other side and checks you're within some bound. If either clock is out it won't work.

TOTP generates the codes on both the client and server sides. The user enters the client generated code and the server validates it.

So if the time on the client device is wrong (perhaps it was manually set and not updated for daylight savings), then the server will disagree with what the code should be.

It's the FB app on a new iPhone, set to Sydney where we live. There are no DST options I assume "it just works" covers all that.

There is an option to resync inside the FB app, which refreshes and... doesn't change a thing.

Your advice is appareciated though and I'll keep it in mind for the future.

Back to my original comment though. Want support? Maybe you will go to https://www.facebook.com/business/help and read the part that says "Find answers or contact support" and click the button.

This brings up an overlay with the usual SaaS buffet of simple answers for simple questions and not too much else. Click on one, don't find your answer, but wait! There is a message under it!

"Need more help? Contact support."

Yet no link. No button. Just dead text.

This is how multi-billion dollar companies operate in the 21st century.

This is a prime example of why Google and Apple's mobile app distribution duopoly needs to be disrupted. They're poor stewards of this space, and they've kept an iron grip on the mobile app market for over a decade, now.

Makes sense. The best thing you can do with Google is to stay as far away as possible.

Smartphone app development is a poor investment. Just build a good website.

That brings up a question I've had -- so many web sites seem to work better as an app. Is that because app development is easier, or is there some things that you can do with an app that you can't do through web standards? Specifically what types of items can't be [easily] done as a web app but are easy as a smartphone app?

Push notifications. One of the primary reasons why companies haven't switched to PWA for simple apps.

Then what is that popup I get all the time visiting sites that says "Allow site xyz to send notifications"? Is that different than push notifications?

Doesn't work on Safari / iOS.

App push notifications can be silent, but web based notifications can't. That means you can push updated content without disturbing the user in an app, but not for a webpage.

I would rather just get an email at this point.

Mostly its just about user data and money. It's AOL all over again. Only this time, they just might win.

Apple refuses to implement standards that make web applications usable on mobile devices, even though Chrome and Firefox have supported them for ages.

It's because the web was developed as a means for delivering documents and app development has been bolted on as an afterthought.

Unfortunately, Apple refuses to implement basic features into Safari that Chrome and Firefox have supported for a while, and they refuse to allow developers to use browser engines other than the embedded Safari on iOS. This forces developers to release iOS apps if they want to target iOS users.

What we need is some anti-trust litigation with teeth.

What features?

Notifications is the big one. Only way to push a notification to a user on iOS is to have an app (even through Safari on desktop supports notifications).

Web notifications aren’t power efficient the way APNS notifications are.

That’s why it’s easy to build support for it on Desktop safari but not iOS.

I except they’ll support Web notifications on iOS once the technology is mature enough to meet the power budget.

web notifications work exactly the same way APNS do. If OS vendors are reasonable and expose the APIs needed to use them efficently that is.

This is not correct. Web push notifications rely on running a persistent service worker in a JavaScript runtime, which is not anything at all like how they are received by native iOS Apps.

Additionally with iOS apps, Apple can and does work directly with developers whose apps are overusing the notification system and draining power, which they cannot do with arbitrary web apps.

> What we need is some anti-trust litigation with teeth.

Indeed. We need big tech to divest some of their eye watering profits into actual user support. With so much depending on tech, kicking out users at a whimsical ML bit flip can be super damaging.

I really don't understand why google doesn't offer some sort of human support. It's not terribly hard to implement something that is revenue neutral (or even positive), and the customers with the most money to spend are really going to want it.

Because they’re a “tech” company through and through and to them it’s a step backwards.

Some Google employee usually chimes in on these to provide a link to some forum where if you just follow these specific steps for one specific problem, you might be able to fix it. But it ignores everyone else who doesn't fit into that narrow range. Google has worse customer service than Comcast.

This is the other side of the Play-store problem. The other side is the tax.

Play store employee can ban your app = destroy your business at any time. The reason can be 'new policy', 'misunderstanding' or something more problematic, such as influence from your competitor to the employee.

A 'power to destroy business', should not be a click away from some random employee.

Except it's not, they give you time to comply with the policy.

How much time was given in this instance?

I don't think they read the tweets.

> Play store employee can ban your app = destroy your business at any time

No one forces / forced businesses to rely on apps for their business. Websites do exist and work well, so there are alternatives.

> A 'power to destroy business', should not be a click away from some random employee.

It exists everywhere, it is called monopoly and this is only an example thereof.

Businesses could also, for instance, be destroyed by ARM Holdings refusing to license the ARM architecture to them, or Valve deciding to pull a game from Steam.

Especially with technology, there are a great deal of monopolies that exist.

ARM can't unilaterally revoke a contract (and even if they tried, it'd end up in court; your production lines wouldn't just suddenly stop working one day), and there's plenty of ways to distribute games to users on PCs other than Steam.

> ARM can't unilaterally revoke a contract (and even if they tried, it'd end up in court; your production lines wouldn't just suddenly stop working one day)

And the only reason Google can do that in this case is because the terms were specified as such, and ARM Holdings could do that too.

> and there's plenty of ways to distribute games to users on PCs other than Steam.

And there is also a plenty of ways to distribute Android software other than the Google Play Store, but the effect in both cases is that one's business will likely die being denied access by either.

> And there is also a plenty of ways to distribute Android software other than the Google Play Store, but the effect in both cases is that one's business will likely die being denied access by either.

What's the rationale behind wanting multiple appstores on iOS, if multiple appstores on Android has no positive effect for developers anyway?

On the plus side there are multiple app stores for the android and the ability to simply install your app directly to the device. So even if Google play store bans you outright you aren't dead in the water.

Now that being said. If you are dependent on google for revenues...

And they can even do it inadvertently. I've run into multiple examples recently due to the Android file system becoming ever more locked down.

Yes, the vast majority of apps have no business writing to any location other than their own storage, and in general even reading other areas should be subject to severe restrictions.

However, there are some apps that have a *legitimate* need to be able to wander freely through the file system. Specifically, apps whose purpose in life is dealing with files.

The latest run-in I've had with this: The Goodsync Android client, which now appears to be basically useless. It's a file synchronization tool, what good is it if it can't wander where the user wants it to? Now I have to plug my phone into the computer to do the same task (the file system lockdown doesn't apply to access from the PC) that I used to be able to do simply by having the phone in the room.

Daily reminder that your business/project/hobby should not depend on anything controlled by Google.

Edit: Despite all their shortcomings, I wouldn't mind seeing android phone makers join force and create an independent app store.

That's unrealistic if your target is android phones.

You can wax poetic about the idea of an independent app store all you want, it doesn't change the facts on the ground in the current moment.

Or you can just make a web app. Unless you are doing games or anything requiring lower-level hardware access, like a bluetooth scanner or something.

Why stop there? Web apps can trivially be taken down if your domain registrar, hosting provider, datacenter partner or payment solution decides to kick you out.

There many hosting provider, datacenter, payment gatway to choose from. But just 2 appstore to use (technically android can have as many as you want, but how many pp use 3rd store right now?)

compare to china where google is blocked, which as a result has dozens of competing appstores. each phone brand has their own, and then some.

i do believe that if google didn't have an appstore that it forces android licensees to use, then we would have a similar variety of appstores world wide.

The problem I have with third party app stores is you have to open up permissions to install any third-party app -- you can't just say "Trust apps from the following app stores" (at least that I've found, although it has been a while since the last time I looked into it).

You have alternatives for those things though, in other countries if necessary. Ok, ok, accepting payments is hard -- if you're getting booted from your processors because, say, you're in a controversial industry, you might have to get creative (look at the US cannabis industry for such creativity).

You mean that thing which google controls 90% of the rendering engines for? [Yes, they have much less power, but its not exactly independence either]

Let's not go into hypotheticals about what might happen if google started... what, inserting spyware, basically, into their own open source browser engine?

Chrome and even Firefox have already begun political policing of their browsers (https://reclaimthenet.org/google-chrome-web-store-bans-disse...). The next step seems like it’ll be forced curation of the web or blocking of IP addresses, since they’re already knocking on that door.

They banned a sleazy-looking extension from their extension stores. That's a far cry from "forced curation of the web or blocking of IP addresses". And, if that happens? Someone will just fork the browser.

The hypothetical is that they remove support for part of a web standard and screw over your app.

Which they certainly have done in the past (albeit,usually for really good reason. Google is a bit in a damned if you do damned if you dont position)

Users, for the most part, don't want web apps and won't use them. It's an unrealistic stance.

Actually, users, for the most part, don't want to install apps.

Maybe HN users, but an increasingly overwhelming percentage of the general population prefers to download an app to do something instead of using a mobile website.

Various industry surveys, like the one from comscore in 2019 suggest otherwise. IIRC it found that most American phone owners install only 1-2 new apps per year. I think they would've gone to more new websites than that.

That study doesn't imply what you suggest. Yes, most Americans install 1-2 apps per year: Facebook, Instagram and similar, and that's all they ever use. That segment of the population doesn't use the mobile web either; they spend all of their time in 1-2 walled gardens and you're not going to reach them with an app or website.

Couldn't you create a web-first app, then have the "app" that appears in the app just install a launcher icon that launches a browser window pointing to your website? This method (PWA?) I believe also lets you launch the browser without the browser control bar at the top (so it appears as a local app).

The best part of this is even if your app gets banned from the app store, it is still accessible to anyone that wants to bookmark your URL, and I believe that users can also install the local PWA icon without having to go through an app store (I may be wrong about that, just starting to learn about PWAs and what is currently possible, and what is coming up).

Do you have a source for that?

You seem to know what you are talking about, but from my personal experience this does not see correct. To me, it seems that it used to be that way, but now people only download apps for very specific purposes.

Do you have pointers to data on this subject? I am very interested in this subject!

Webapps don't show up on the home screen, can't be targets of share pane actions, can't access contacts, can't send push notifications.

The current state of webapps on Android has made good progress on these issues:

Supported on all major Android browsers:

- Progressive Web Apps can be added to your home screen - Web Push lets web apps receive push notifications (Brave's implementation is broken, but they try to support the feature)

Experimental APIs supported by Chrome for Android:

- Web Share Target API, which lets homescreen'd web apps receive shares from other apps (the share pane makes no distinction between sharing to a web app or a native app) - Contacts APIs to read the user's contacts list

I think PWA's competitiveness will get a lot better once other browsers adopt the Web Share Target (or something like it), but I'm skeptical that they'll really take off as long as Apple continues hold off on implementing features that would make PWAs competitive with native apps.

I guess all business/project/hobby websites should shut down then? Your website won't get very far at all when most people rely on google's search results, browser, and email service.

The same is true for Facebook. My website and ad accounts were banned one day without a reason. They said it did not meet community standards. There has been no recourse since 2 years.

I once tried to advertise on facebook. I was editing my ad, and when I was trying upload an image or something, it went into some sort of infinite loop of trying to reload the page. Then, seconds later, it was banned, taking down my business page and my personal account as well. Still down after two years and going through their "appeals" process, which involves trying to log in with my old password, saying it was a mistake, and nothing ever happening.

There is no appeals process unfortunately. For the past 2 years I have been asking for an update regularly and the only response I get is an automated response saying that they have checked my ad account and there is nothing they can do about it.

My website is still blocked by Facebook's sharing debugger and my ad accounts with balance in them have been blocked without a reason given.

You could sue them in small claims court.

I'm completely in favor of everyone suing facebook for literally any reason, but to be clear, what would the legal standing/basis be for this small claims court?

The balance in their ad account belongs to them, not to Facebook.

Beyond that, the point is just to make them notice you, so anything might be good enough. Reporting to a regulator like the CFPB can also be effective if the case looks financey - I got my PayPal account back recently that way after they didn't like how I typed in a tracking number and banned me for life.

I got banned from Gumtree for a similar system. I was using a script blocker, uMatrix (but have since stopped because of the hassle), and the site wasn't coded to handle something with reCAPTCHAs not loading and the error messages were useless. Support said it's an automated system and for contacting them about the ban, they blocked my email as well.

You do have a recourse, same as with any other business: you can always take them to court.

Unless they force binding arbitration in their terms. This is quite common and precludes you from going to court.

Which is why such terms should be illegal.

Yup. Binding arbitration clauses should be taken out with a RKEW as nuke from orbit means you're way too close.

// RKEW - Relativistic kinetic energy weapon

They're not even getting arbitration now- a professional arbitrator would likely find in their favor, at least as far as monies owed.

Repeating this same line is just harmful noise at this point. Rather, we should - legally - demand more from tech utility platforms like Google, Facebook or Apple.

> Repeating this same line is just harmful noise

More like a strong disclaimer worth repeating for newcomers in the space.

These stories have come up several times over the years on HN and it is worth reiterating:

It is very risky if your product/company relies solely on another company's success and openness to you - especially if that company is not contractually obligated to continue providing you service.

> Daily reminder that your business/project/hobby should not depend on anything controlled by Google.

Note that this doesn’t mean don’t participate on platforms controlled by Google, it just means diversify your business so you have multiple income streams rather than just putting all your eggs in one App Store basket. Apps should really just be a client facing portal tied in to a whole ecosystem of solutions you provide.

There's already F-Droid and the Amazon App store.

f-droid doesn't have payments, Amazon is yet another faceless mega-corp that has had its fair share of sellers accusing them of various shady things.

Creating a Play Store alternative is hard for many reasons:

1) finance: the more countries you operate in, the messier it becomes. International taxes are one hard mess, KYC/AML regulations differ between countries, and to top it off there is the whole "international sanctions" issue especially regarding Iran (where it's fine and explicitly encouraged by the EU to do business with Iran, but any entity that has US exposure exposes themselves to liability in the US for violations).

2) vetting of apps against a constant onslaught of spam, malware, copyright violations: f-droid has it a bit easier since they require all apps be open source, but a commercial, widely used alternative will have to run static analysis, dynamic analysis (to catch runtime exploit attempts) and manual testing. All of this is expensive and requires expert knowledge of Android as well as IT security.

3) Implementation and hosting: an app store worth its name has a lot of binary assets to distribute to users (and again, you have to avoid getting into trouble with people abusing your service to spread illegal content, because there will be such cases rather sooner than later), the store itself has to be implemented, regularly adapted to account for changes in the Android core, you definitely want a focus on security to avoid some hacker distributing malware to all your users with a push...

4) Customer and developer support: it's a well-known meme that FB/Amazon/Twitter/Google are almost impossible to reach for ordinary people without raising a shitstorm on Hacker News or a well-funded lawyer team... but the key thing is, support is expensive to run.

The hardest part is getting any traction with it.

Developers have no incentive to go the extra mile to publish on your store, because there are no users and it is extra work.

Users have no incentive to install your store, because there is barely anything on it.

OEMs have no incentive to preinstall your store, because you don't have any content (which devalues their product) and they don't gain anything from it. If they are willing to roll their own store, they at least rake in all the profit from it.

> f-droid doesn't have payments

I'm just a normal F-Droid user, but couldn't you build In-App purchases or link accounts to your website?

So, register for a paid account on mygreatapp.com and use the login details in your app?

I think that's what Google and Apple specifically DON'T WANT for financial reasons, but how about F-Droid?

Unfortunately, that's hard.

Developers like in-app payment methods because all they have to do is integrate an SDK for payment, and then they get a monthly payment on their bank account and a bill for accounting, that's it.

If you want to handle payments yourself, you'll have to:

1) implement user management to deal with storing what stuff a user has purchased, with all the GDPR and customer support (forgotten passwords, hacked accounts, lost MFA creds) headache that comes from that

2) find a payment processor that operates in all your target countries (no, Stripe and Paypal alone won't cut it), and integrate these (and hope they don't run into the same issue with Paypal, who are known for deciding on a whim to withhold funds)

3) Issue individual bills to customers, account for stuff like cross-border VAT, insanities like county/city sales taxes, deal with refund laws

4) deal with fraud attempts, angry parents, ...

5) Re-implement recurring payment schemes if your business model wants these

In the end, app stores (and ad SDKs) are a matter of convenience. Big shops like Epic, Spotify, Netflix can get away with running lots of this infrastructure on their own, but 99% of small devs don't have the time, knowledge and legal requirements to deal with that.

Doesn't f-droid require all app submitted there to be open source? Kinda hard to implement in-app payment while keeping your app open.

Yeah, SEO has become hell hole in the last 6 months. Crippled businesses across the board from what I've seen and heard on the grape vine.

After 8 years of SEO expertise I honestly can say that there's nothing about SEO these days that makes sense, pages are not being indexed for weeks at a time, spam/crap at the top result. Throw in the endless (and ambiguous) requirements to adhere to the mighty G's requirements.

There are independent app stores.

The problem with the Parler situation is more that their average user is too incompetent and inexperienced with computers to understand how to download something that isn't literally laid out in front of them.

I refuse to use Google Play or create a google account so I use The Aurora Store which replaces the Google Play Store.

If I want to download some independent apps I head to F droid store or Aptoide.

There are even more independent and alternative app stores, but these are the ones I use.

Oh, and don't forget about github and app downloads directly from the web.

F droid allows you to install without Google play.

>> android phone makers join force and create an independent app store

And why wouldn’t this App Store have any takedown policies? Or have to follow the same kind of legal restrictions? In the discussed example, this wasn’t even automated; an employee manually reviewed and chose to ban their app.

This is why we need anti-trust action when it comes to mobile app distribution. Google and Apple have abused their duopoly in this market for over a decade, now.

> We don't allow apps whose primary purpose is featuring or hosting objectionable [User Generated Content]. [...] your app currently only includes feature to flag inappropriate users. [...] the app must provide a user-friendly, in-app system for reporting objectionable UCG.

It's user generated content. How is reporting a user functionally different from reporting content?

This makes me even more driven to switch to F-Droid for everything I possibly can. Hopefully we'll see a daily-drivable Linux smartphone before too long so I can ditch Android all together.

The difference is the same between malice and ignorance.

A user may just be wrong about something or not quite understand something. They're not trying to deliberately cause disruption, there's no need to fault the user. Remove the content, inform the user, move on.

It's only a problem is a user continues to submit reported material.

Agreed, that is an important distinction. As written, the button might discourage users from ever clicking it except in the most extreme cases or until after observing a pattern over time.

Which isn't to say that Google's actions are at all justified here. This kind of thing makes me extremely wary of continuing to do business with them.

Yup. I reported a message on a forum a couple of days ago. The user made a completely innocent mistake and there was nothing at all objectionable about the message--the problem is that he inadvertently doxxed himself.

Elsewhere I am a moderator on a forum and I have seen exactly the same mistake, zapped it (the errant line, I left the rest of the message alone) on the spot and PMed the user--and been thanked for doing so.

Google doesn't understand customer service. They really don't. When speaking to customers, I ask them what is the biggest issue with using a google product.They will most likely say - reaching a human being.

PS: I work for a competitor of one of google's products.

Watch the amount of these posts rise steeply in the future. Mass moderation is a Don Quixote fight, and in addition to their own products, big platforms increasingly force these policies on everything available in their app stores.

There was a great thread on moderation just now: https://news.ycombinator.com/item?id=26458826

It is probably a reaction to the pressure platforms (IMO to some extent rightfully) receive from regulators. Kicking downwards... but platforms especially app stores will probably be forced to do this anyway in the middle-term future - be it social pressure or regulatory enforcements.

Our app recently got banned a full week before being reinstated because of a non-published outdated .apk in an alpha test track that was not used.

To be fair, we cleaned up our artefact management but still. Google is evil.

What's hostile is the timing and bluntness of Google's treatment of developers on their platform, as well as the difficulty of getting a decent human conversation with the party you're doing business with.

But if Google sets clear rules about how your bad people/behavior reporting button must look, I don't see Obvious fault in such policy.

When did Google ever set clear rules? It's all rubber paragraphs that are enforced at the whim of anonymous bureaucrats who will never be held responsible for whatever decisions they make.

> enforced at the whim of anonymous bureaucrats

Not all of whom are human.

You could argue that someone or a some group asked for a change order and/or approved a change request with a “bot” or a piece of code.

This is my new favorite ban reason. "Reviewer glanced over functionality and decided that a required feature is probably not implemented although it is implemented. Ban immediately on a Saturday."

The app has been restored https://twitter.com/EskomSePush/status/1371423310878183426?s...

After reading this, I think Signal is also violating this policy. I was not able to find a way to report messages or users. In comparison on WhatsApp I can report a user or group and the "most recent messages" will be forwarded to moderation.

I think it is kind of strange that even a private messenger needs to implement a report system for messages.

> In comparison on WhatsApp I can report a user or group and the "most recent messages" will be forwarded to moderation.

Really? So if a random person decide to report me, plain text uncrypted messages will be sent to WhatsApp?

Even if this does not happen regularly, this means that WhatsApp have means to read my messages, which is scary enough.

I dont think this confirms that WhatsApp can ready our messages. In theory it could be implemented as a simple message forward, just as any message can be forwarded in Whatsapp.

No. It means the person making the report has the ability to read the message. Of course they do, otherwise it would be completely useless. You can't stop the recipient from doing something you don't want with the message you send them.

It's sad and ironic that Google used the exact same banning procedure they incorrecftly surmised the app used.

It won't solve these issues with gatekeepers fundamentally and doesn't work for all cases, but embracing PWAs means anyone can click the 'install' button for your application and side step some of this nonsense while getting a fairly robust feature set.

What is a PWA?

Progressive web apps - i.e. web apps

This seems so unnecessarily hostile. Honestly if I build a side project it has to be viable as a web service, I can't imagine hinging my business on the arbitrary whims of the apple Google duopoly.

Kind of ironic that their own mod tool apparently uses bans as the first step!

I think people forget that you don't need an app on the app on an app store to use it. The app is out there already, just give it to people. The app store simply gets rid of the "bad boy message".

Stop complaining and find a way around it.

How long until we have to validate our code against a list of banned terms, like "master" or any new truly egregiously offensive terms of the day, before we are able to host our code or binaries on a third-party service?

This troubles me enough to start considering webviews again for my next project.

A more fundamental question is why it is Google’s business at all to police apps or the content users generate on them. If someone finds content “objectionable”, they should just move on, and not complain or try to shut down that content. What an utterly unnecessary fiasco. We need decentralized platforms and alternatives to the hegemony of Google, Apple, and other big tech companies.

Imagine what it would be like if the section 230 protections were eliminated.

For anyone unaware, the name of the app is a play on a swear word. "EskomSePush" is an intentional play on the phrase "Eskom Se Poes". Eskom being the power utility that provides electricity to South Africa, and has been woefully inadequate at doing so.

See here for a meaning of the word Poes in Afrikaans. https://en.wiktionary.org/wiki/poes#Afrikaans

Wonder if Google is aware of that and has been getting lots of "reports" by genuine users that are offended by this? And this weird "technicality" is just a cover for why they really got banned/taken down.

They have a one-hit wonder that they're trying to ride the wave on. So instead of sticking to the functionality they had (notifications/schedule/etc) which was perfect for users, they decided to add a "comments section" it seems for people to vent their frustrations. So no sympathy from me, despite Google being in the wrong here.

So you dislike the devs because they added an optional feature?

There's not much more to expand on the core functionality of notifications & schedule visibility.

I didn't say I dislike them. I'm just saying they don't get sympathy from me regarding this "User Generated Content" feature that is now being targeted by Google. They should have known this feature would be a place for people to vent in ugly ways rather than to discuss or offer advice. And South Africa right now is a hotbed of civil discontent, especially between different racial/cultural groups.

Herman Maritz, the author, has stated that they were overwhelmed with app users that complained to them when there were normal unplanned power outages or when Eskom deviated from the planned load shedding schedules. The chat feature was a way for user of close proximity to confirm a power outage with each other [1]. I wouldn't call it a 'comments section'.

[1] https://twitter.com/hermaritz/status/1371017176593928194

Reply on news.ycombinator.com