No blame or judgement here, just making observations.
@dd
> Depends on how one defines success, I suppose.
Indeed, this is in the eye of the beholder. But if you consider his constant frustration with the majority of the fediverse using ActivityPub instead of Zot (or The Protocol Formerly Known as Zot?), then it's reasonable to think his metric for success is adoption. Which the decisions I quoted and the habits I noted all work to undermine.
@tetrislife
> moving beyond prototying languages
Scuttlebutt started getting taken more seriously when there was an implementation that wasn't in JavaScript. Matrix started to get taken more seriously when there was progress on server implantations that weren't in Python. Perhaps this is just language snobbery, with no technical justification. But it's observable that it increased interest in and adoption of the protocol in both cases. As @dd said, it depends on your metric for success.
@tetrislife
> Maybe a UI for such a multi-faceted software artifact is difficult?
Indeed. If it wasn't, we wouldn't be having this discussion. But there are platforms that have as many features as Hubzilla or more, but their UI is easier to navigate, and self-teaching. UX is not just about UI either. It's also about thorough user-facing documentation, that's easy to find and easy to understand. Doing this well is important in making a feature-rich multi-tool more tractable for new users.
@strypey the part about not moving beyond prototying languages, I wish it was a slam-dunk, but
a) it may only be true nowadays, with Elixir around
b) it asks for working code to be thrown away, which may be insane to do before a re-implementation is around
It can be expected from the community, maybe not from Mike.
@dd