Hey man! sorry for the late reply, I know its been days but things got busy and this fell down the list longer than it should have.
First, thank you for the feedback back then I did take it seriously, even with the delay in acting on it. A few things came directly out of it:
The comparative chips idea “more like this, but faster / calmer / shorter / harder” I built it. It’s live on game pages now on similar games. You were right that it sidesteps the NLP problem entirely and turns it into a UI problem I can actually ship. There’s room to expand it further, but the core is in and working.
I also shipped automatic importing from Steam. It’s not OAuth, though since Steam doesn’t offer a per-developer OAuth flow the way itch does with profile:games. What I built instead is simpler: a dev pastes their Steam App ID or store URL, and the site pulls the store page data (title, description, screenshots, genres, etc.)
itch.io is next on my list, and the OAuth approach you laid out is genuinely the right one. I hadn’t looked at profile:games that closely and the argument for it (itch does the verification, no crawler, no claim-flow ambiguity) is convincing. I do want to be careful with it though, exactly for the reason you flagged that the payload apparently carries earnings and purchase counts, and I’d need to strip that at ingest and be explicit about it on the consent screen before I’d feel comfortable shipping it. Same with the published flag/ I’d need to filter on it so a dev doesn’t get an unpublished game exposed. So it’ll take a bit more thought than the Steam version did, but it’s on the roadmap.
On adding Bombercup manually as admin/ I’ve thought about this and I don’t think I can do it, and not because of effort. It would mean building a whole separate “admin creates and owns a listing” path, which directly contradicts the model the site is built on that the developer/user is the one who uploads and edits their own page, always. Adding an exception for that undermines the exact thing I’m trying to guarantee, and from an engineering standpoint it’s not something I want to bolt on as a one-off.
That said. I think I actually already got a submission notification for the indie dev form and approved it, so if that’s the one I’m thinking of, you’re one step away from having it live with your own text and screenshots rather than anything I’d write off the itch page. If you do want to finish it, another dev told me the whole flow took him about 4 minutes start to finish, so it’s not a heavy ask. But genuinely no pressure, if you’d rather wait for the itch.io OAuth import to land and do it that way instead, that’s completely fine too, it just might be a while before that’s ready.
Thanks again for the thought you put into all of this man it’s rare to get feedback this well-reasoned, and it’s already shaped what shipped. Cheers!!