What I’m saying is that in order to make these claims you’d need to be able to show that it’s more than just a normal rate of error.
If you just take a few reports (or indeed even a few hundred) and ignore the size of the sample, then the conclusion is likely to be flawed.
Do you actually know what the rate of ‘capricious’ rejections is? I have seen a few 10s of cases reported, and many of them aren’t actually capricious, although a few clearly are.
My guess is that the chances of a capricious rejection of an App Store submission is less than 1 in 1,000,000.
I’d say that is neither unjust, nor capricious, and simply a reflection of scale.
I agree that when they make a mistake in the form of applying the rules inequitably, get called out, and fix it, that's a mistake.
But I claim that when they make a mistake in the form of applying the rules inequitably, get called out, and double down on applying the rules inequitably, that is not a mistake, that is capricious.
There may only be a small number of times they have been capricious, but they are not "making a mistake." They are applying the rules capriciously.
And while nobody is getting killed, the principle of justice is the same here as in our society. Injustice anywhere is a threat to justice everywhere.
One person imprisoned for a crime they didn't commit and the police knew they didn't commit taints the entire justice system. It removes confidence for everyone.
It's the same principle here. Mistakes, fine. But applying the rules inequitably and doubling down on that even once taints the system for everyone.
TL;DR I agree that the system usually works. But for a mistake to be a mistake, Apple has to be prepared to fix the mistake, not double down on continuing to have different rules for different developers.
In the Hey case it does seem like Apple was simply applying it’s rules, and DHH didn’t quite understand them.
I was far from simply applying the rules. The only documented rule they were applying was the one, which states that you must use Apple's in-app purchase if you offer sign-up within the app. There are many high profile apps (Netflix, Kindle, Fastmail etc) that require paid memberships but which don't offer sign-up within the app (not to mention several apps that Basecamp already offer in the app store), and Apple has apparently been fine with those for a long time. For that reason, Basecamp had every reason to think that they were in compliance with both the letter of the guideline as well as Apple's interpretation of it.
The distinction between what Apple calls reader apps and whatever they deem an email app to be, as well as the consumer/business distinction that an Apple spokesperson also offered as an explanation to one publication, are not mentioned in the app store review guidelines, and Apple has not talked about them before, so how on earth can DHH be blamed for not understanding that these distinctions mattered?
There is literally no evidence at all that developers are spinning the wheel of fortune or that Apple is capricious or unjust.
Given that number of developers, hundreds of mistakes every year would be an amazing track record for Apple.