Game overlays and capture were working fine, and the controller API was designed to support any number of controllers (Steam's support is great but their interfaces are subpar, in our opinion). We were also able to pull from a well known database of controller configurations and device IDs, which really made this a non-issue.

Linux compatibility was fine as far as the client went (all of our code was cross-platform and not webkit frames or the like). The client even ran on Android and iOS.

If you're referring to Steam's Proton, we really didn't want to touch that area for a while. But we had much better systems for searching for new titles, including those that worked well on the system and also matched all of the criteria (tags and whatnot).

Our social system was also designed to support "cross-talk" between different marketplaces (Steam, GoG Universe and Epic) but we never got as far as building out any client functionality - just the initial blackbox proof of concepts.

The store aspect was indeed just a smaller part of it, though it was complicated in its own right.

The project was a great idea and we were executing well on it. Lots of cool new tech was developed for it. But nobody we talked to wanted it - including publishers, users, investors, or even friends. It didn't matter how compatible we made it, the fact that we didn't push you to re-buy games, etc.

We wanted to make an non-shitty experience for gaming and the market simply said "no".