The Itch.io desktop app gathers information on how much the user played a game. My suggestion is that, after some time and after a given number of people have downloaded and played the game, information about average play time is made available either for developers alone or for developers and users.
As a developer having that kind of information would be very useful, it'd tell a lot about how people play the games we make and could even lead to improvements in the game itself. It'd be even better if we could get an "average play session" info as well.
As an user I think this could be an interesting tool. I've seen paid games on Itch whose store pages are full of comments indicating the game doesn't even run, this kind of extra information could be useful for consumers to make more informed decisions. I think it'd also give people another way to find diamonds in the rough, some very inconspicuous games with few ratings have a lot of content to show.
The con to this proposal is that... Well, let's compare Itch to Steam. Steam shows what games you own, what achievements you got, which game you are playing now, how much you've played for each game in your library, so on. That's a lot of information being shown and I know I'm not the only one who dislikes it.
So, keeping the suggestion of play time analytics in mind, does it cause any harm to the users' privacy?
It would if Itch did it like Steam does. But if all Itch displayed was average session time and play time across users, and if it only was displayed after a number of people played the game (this way you wouldn't even be able to track the play time of the first adopters), I don't think there'd be any harm. It'd be impossible to single out individual users and their habits. And it's not like any extra data would be collected, the desktop app already gathers the users' play time data.
Another counter-argument I can see against this proposal is that "only a fraction of players use the desktop app, any data you'd get from this would be incomplete."
...And, yes, that is objectively correct. The data would be incomplete. But even then it'd still be useful for developers and players, no one needs perfect Panopticon-esque information here. Being given a rough estimate of how much people play a game — even if it's gathered from the comparably smaller crowd using the desktop app — would be useful enough for developers and players.