I was thinking about what the definition of a core Feed Collecting Service Specification would look like and authentication is such a PITA. Of course what I’d specify wouldn’t include auth, but it’s still painful to think about.
What would a minimal feed collecting service need?
Add Feed Remove Feed Get Feeds Get Feed Add Category Remove Category Add Category to Feed
Now, I may be missing something that should be in a core specification but that seems kind of like a minimum to me. Even the Category functions may be too much for the core of it. Of course the service could still have their own proprietary way of managing feeds. They could choose to build that on top of this core set of features or next to it or add this spec on top of their existing API. You know what I’m saying. 😄
Dave Winer’s FeedLand got me thinking about this. There is also this great podcast episode where he explains what he wants to do. His tool could implement this specification if he wanted it to. I don’t think there’s an API to it but a very generic specification, implemented by multiple third parties, would open up feed readers to supporting multiple feed services without having to do special client side work to support each service. Just do it once and connect to any. Specs aren’t ever perfect but they 100% allow for interop between clients and other services.
I’d love to be part of a group working through something like this. Specifically I’d love to get it added to the same collective that includes ActivityPub and ActivityStream, perhaps what I’m after is WebSub? I haven’t read through it yet. I also need to read Social Web Protocols.
I think what I’m looking for is less social protocol and more simiple, agreed upon, API implementation. It is, of course, kind of selfish because I work on a feed reader and would love to be able to connect through a well specified API.
Oh, one more thing! I listened to Dave’s latest podcast episode on the way into the coffee shop this morning and he put the final touches on his vision for a rebooted weblogging system. He’s now covered writing, feed subscription, and finally, discourse. His idea is quite good because it would really make folks who reply to your post think before posting. Listen to the podcast and it will become obvious — I think — why.