And let's be clear, this isn't an argument over competing methods of payment, it's an argument about whether Apple should be entitled to a percentage of revenue for iOS app sales, regardless how they are sold, regardless where they are sold.
We should be mindful to keep this question separate from whether there should be more than one place to acquire apps, or whether it matters if the financial transaction occurs prior to downloading or after the app is installed.
If it's the latter, then Epic disagrees with you.
Epic's own business model says they're entitled to a cut of your revenue if you use their tools and libraries. They don't care what payment method you use. They don't care if you sell copies or sell in-game hats. If you make revenue, they're entitled to a slice of it.
(Yes, Epic does waive their fee for low revenue games. That's very nice of them, though in reality it's obviously a clever strategic move to lure game developers over to their ecosystem, in the hope that more breakout indie successes happen to be built with Unreal Engine. But that doesn't change the underlying principle: they would be entitled to it if they had asked for it.)
Though I disagree with arguments about whether Apple are "extremely well compensated" already. Arguments of principle should not take into account any company's particular monetisation strategy. For example, it's often stated that game consoles are sold at razor thin margins, or even a slight loss, and this justifies why they can charge high license fees for games. I think that's a perfectly valid strategy, but the fact that the game consoles aren't a profit centre (whereas iPhones definitely are for Apple) shouldn't affect whether it's decided that a 30% manufacturer's margin on digital software sales is appropriate or not.
But I would also argue that when the software in question is a game, a game console can indeed be a relevant analogy. Games are games—whether you are buying them to play on a smartphone or a GameBoy or an Xbox. The "business models" and "stakes for society" are the same regardless.
If I were in charge of Apple, I would lower their store percentage to a very low 10% for non-game apps. 30% might have been a reasonable amount in the early days, but Apple's market success means they don't need it. (I would also completely abandon their pointless stance over reader content and allow Spotify, Netflix etc to bill customers directly in-app.)
But for games, I would have Apple match the prevailing percentage set by Sony, Microsoft and Nintendo—which I believe is 30% across the board.
I suppose you could bolt a rudimentary spreadsheet feature onto Call Of Duty but that doesn't stop it from being any less of a video game. This isn't interesting and it certainly isn't clever.
A more interesting question would be if Microsoft Excel iOS had a little Sudoku mini-game buried in a submenu.
They would do this by transparently showing the markup to the user, and allowing the user to choose if they want to pay 30% more or not.
That seems like a pretty reasonable solution to all of this. Give users the choice to buy on the app, and have them pay the full fee if that is actually what the user wants.
I guess my point is that App Store metrics (total revenue, subscription revenue, average revenue per user, etc) continues to grow and maintain their lead over Google Play so the chilling effect can't be too bad.
I suggest that in terms of metrics, Apple's App Store vs Google's Play Store isn't really the right comparison to see the chilling effect I'm talking about, because both mobile ecosystems have broadly similar dynamics. A more telling comparison might be something like iOS app developers vs. web developers offering SAAS. How many businesses that run successful SAAS websites charging subscription fees for access also offer an equivalent mobile app where customers can subscribe, accepting Apple's 30% cut and other restrictions?