I think we already have the answer in the field of computer science: ACM. I.e., a non-profit professional organization that runs conferences with peer-reviewed

I think we already have the answer in the field of computer science: ACM. I.e., a non-profit professional organization that runs conferences with peer-reviewed content (and some journals and magazines). A cheap membership (<$100) gives you access to all the content, which could even be free (money comes from conferences and donations). Fundamentally what is needed is a process that produces peer-reviewed, immutable, discrete pieces of globally published work. Self-published blogs don't really achieve that well.