I'm not sure that this does encourage native Linux game development. Why bother putting in the porting work for 0.1% of potential users when someone else might do the work for me via Wine/Proton.
I used to think the same way about native first and forcing devs to adopt Linux. The whole "No Tux, no bucks" thing (which has had pretty much zero impact since it's not enough bucks). But, over the years, and especially with Proton being so good with Steam, I've completely switched. If Linux gaming is to be a thing, then there needs to be an adoption first perspective.
So, yeah... maybe that means most devs will just say "my Linux support is just it runs on Proton probably, good luck!" but the thing is... there are games on Linux now. Lots of them. Lots of good ones. I was playing The Witness last night by just pushing play on Steam. No winecfg or winetools or separate DriveC Steam installation. No messing with drivers. I pressed play, the game loaded, I played it. I've repeated this loop in the last few months with most of the games in my library. Endless Legend is back in my rotation again. All of the dumb anime Japanese games where they don't even know Linux is a thing that exists suddenly work. It's glorious.
Wine/Proton may be the lazy way to develop for Linux and might not give people the coveted title of Linux exclusive gamer, but it's working really well if all you care about is playing games and not installing Windows.
Compatibility is a stepping stone to increasing the 0.8% (actual numbers) Linux population on Steam to higher numbers. If 2% or even 10% of users were Linux-based then developers would have second thoughts about choosing DirectX over Vulkan for example when it makes it more difficult for them to reach those customers.
Also the "no tux no bucks" philosophy many Linux users take in avoiding paying for non-native games.
