[pjc50]

Apple's opacity leads to this kind of thing. The "rebuttal" doesn't have a citation either. How do we know that it itself is not also fake news? Apple should have a nice clear page we can all link to that explains how this works. But nobody seems to have found one.

Edit: no, really, this is important: we're in the middle of the lesson of "social media can amplify things that aren't true and you should check them". Great. So, how do we verify things? What should we be doing? The counter claim link at the top of this thread is also just a random unsourced tweet!

[tonyedgecombe]

It's in the terms:

"In such cases,Apple will have the right to retain its commission on the sale of that Licensed Application, notwithstanding the refund of the price to the End-User."

[ShroudedNight]

I have never engaged in any kind of professional legal training; I am most definitely not a lawyer.

That being said, my understanding of 'notwithstanding' in this context would imply that Apple reserves the right to retain its commission even when they refund the price to the End-User.

https://weagree.com/weblog/notwithstanding-is-a-tricky-one/

[voxic11]

What is in the terms? That part says they can keep the commission if they want to. But the question is do they actually keep it.

[jerf]

If I were writing these terms, I'd want to include that clause so that if someone attempts to attack Apple financially by getting a lot of people together to buy apps and then refund them, Apple would have some discretion in terms of keeping enough money to offset the attack. It would be a tricky situation, but at the same time, by including that clause they also make the situation less likely to come up in the first place by making that attack vector less appealing.

There's also some other attack vectors I can think of that this renders less likely, like "let's get my big YouTube fanbase to all buy this one app today, then refund it near the end of the refund window, thus pushing this app to the top of the store at no cost to my YouTube fanbase but at cost to Apple!" This would give Apple the cover to keep enough commission to cover the attack. Given the known shenanigans played on the app store, this seems less like some bizarre far-out possibility and more like something that would be a routine thing done by sketchier app developers if the terms didn't make it a bad proposition like this.

But in a normal day-to-day transaction, the logic works out in favor of refunding it even if they do nominally have the contractual right to refuse to do so.

[qwertygnu]

How would either of those situations (which are pretty much the same) cost Apple _financially_? If 10,000 people buy an app for 1$ each, the creator gets 7000$ and Apple gets 3000$. When they all get refunds. The creator refunds the 7000$ and Apple refunds the 3000$...

[Liquid_Fire]

Apple have to pay some amount (let's arbitrarily say 1%) for processing the payments, so in this example they would lose $100.

[xxs]

I doubt Apple rate is this high.

[LanceH]

That may not have a single answer.

[hardlianotion]

In other words: the rebuttal is not clearcut.

[KingOfCoders]

Shouldnt this be the top comment?

[jerf]

That is a good question, and what I'm about to say is not a full answer. But I do give some of credence to this tweet because it is the original claimant publicly admitting they are wrong, which is a costly act that puts skin in the game. Now if it is found that it is true that Apple does keep the %30, the claimant is going to take an even harder hit.

There is also the matter of prior probabilities; if you had asked me beforehand "How certain are you that if Apple refunds a charge, the refund is complete and they also refund their cut?" I'd have given you an answer in the 90%+ range, on the grounds that if they did something that nasty we'd have heard by now. Furthermore, if they did change that policy, I'd expect to hear about it precisely as a change in policy, rather than the sudden discovery that it's been like that all the time.

I am emphatically not holding these up as total answers to the question. I'm not sure there is such a thing. But these are reasons to consider this tweet likelier to be true than the original claim.

There is also the constant possibility "What if somebody knows about these signals and fakes them?" In which case the question is, who would have motivation to do such a thing? In this case I can't see how this Tweeter has any particular motivation to fake this matter, as this retraction mostly doesn't benefit them any. (I mean, I do respect them for it, but we still have the original error to consider. I don't have a concrete threshold, but you're still burning a bit of rep. To see it clearly, consider the strategy of "impress jerf by making lots of public mistakes and then publicly apologize for it"... I do respect the public retraction and maybe the first time it's even a net gain, but it's not a scalable strategy.)

Apple would clearly have a motivation to claim they don't keep the refund even if they in fact do, as clearly keeping that 30% would be bad PR, so this signal would weigh against them. Weighing for them, though, would be the sheer mass of people who could contradict them if they claimed not to keep it but in fact did; any app author of any significant size has direct experience with this and that's a large pool of people, which also includes some vocal people in it.

[Nition]

This is really common on Reddit, where if an article says one thing, and the top comment says it's wrong, everyone believes the comment. But sometimes the article was actually right and the comment is wrong, or they're both wrong.

[jasonlotito]

While I agree with you, I can't help but reference Hitchen's Razer: "What can be asserted without evidence can also be dismissed without evidence."

[baxtr]

So opacity justifies fake news?

[pjc50]

"Justifies"? "Fake news"?

No, but not having a clear policy makes it very easy for inaccurate gossip about your policy to circulate.

[damnyou]

Opacity means that the blame for any "fake news" rests solely on Apple for not being clear enough.