While I use microblogging to frame the initial narrative, as explained in the article, this isn’t limited to Twitter-like products. Tangled is “GitHub on atprot

While I use microblogging to frame the initial narrative, as explained in the article, this isn’t limited to Twitter-like products. Tangled is “GitHub on atproto”, Leaflet is “Medium on atproto”, and so on. The problem with client-side P2P is you can’t do large-scale aggregation with consistency. Large-scale aggregation with consistency is what normal people expect from social apps. Re: the OpenAI thing you mentioned, that's actually a perfect example of something atproto excels at. Since the data already exists in the network, you can crawl/index it and run your own tooling that does something proactive on cron jobs etc. See https://github.com/graze-social/iftta https://github.com/graze-social/iftta for some initial work in that area.