The issue seems to be that we have two camps with different opinions of what constitues the PSHB project:
The "Push Feeds" (PF) camp with it's main argument: *Feeds should retain the expected format or we need separate parsers for everything* The "Push Something" (PS) camp with it's main argument: *Push and fat pings are really useful, why should we limit them to only feeds? I want to push ABC* Both sides have valid points but, as Charles pointed out, are defending different things. The real argument here is whether PSHB as a project is all about the whole life cycle, or just about how to push fat pings. It started with a focus on feeds, but now a lot of people have realized that the tech which pushes the feeds could be applied elsewhere. Perhaps what is needed is to split off, or start a new project that is solely dedicated to the "push something" idea, and then PSHB can remain the "push feeds" projects but defers all the push tech itself to the other project. (or start a new feeds project, I don't think it maters which) Similar to how Webfinger, OStatus and so on are really just a definition of how to use other tech (including PSHB) to achieve a certain end. I personally feel that while using the push tech for other payloads is a logical, and inevitable evolution, keeping PSHB as a "push feeds" project only is important, so that consumers can see a PubSubHubbub link and confidently consume it without having to know if it is really a feed or a custom payload. The push tech itself can then move to another project where it would be free to grow without the limitations of feeds. (and of course PSHB just leverages that + Atom to be the PSHB spec) Or in an easy to understand analogy: It's time to stop defining the "car" as a *requirement* for the "combustion engine". Both can, and probably should be developed independently and then brought together for a better result.
