#ActivityBud - the exciting practice of intentionally misinterpreting the #ActivityPub and #ActivityStreams specs in order to derive ridiculous conclusions.

#ActivityBud - the exciting practice of intentionally misinterpreting the #ActivityPub and #ActivityStreams specs in order to derive ridiculous conclusions. By analogy with the movie Air Bud, where the team proved that there ain't no rule that a dog can't play basketball.

Another analogy might be the centaur tower in Dungeons and Dragons.

The definition of centaurs in D&D says that a centaur can carry a medium-sized creature on its back -- the equine part.

But a centaur is a medium-sized creature. Does that mean that a centaur can carry another centaur on its back? And could *that* centaur carry another centaur on its own back?

Taken to the extreme you get a centaur tower - an infinite stack of centaurs. They'd probably do even better than a dog at basketball!

These are fun thought experiments and great for conversations. I guess where it gets to be a problem is when you're so entranced with basketball-playing dogs and infinite centaur towers that we can't have a conversation about, y'know, the actual game. Or you fight tooth and nail to keep those loopholes from being closed.

As someone who was part of creating both specifications, I can assure everyone that the mistakes were unintentional. We really did try to make specifications that worked.

@evan the real question is how one properly represents a centaur tower under ActivityPub.

@evan maybe some examples would clarify what you mean, but i'm not sure how to define "the actual game" in a way that everyone would agree they're playing it. like, what counts as a "loophole" and what counts as a fundamental use case? what makes a conclusion "ridiculous", or an interpretation "intentionally misinterpret[ed]"? these questions all depend on the frame of reference of whoever's asking them. i generally assume good faith whenever these things are being discussed. should we not?

@evan one example that illustrates just how much contention there is about even supposedly basic things: are activities content? one might argue they are the primary reason the specs exist at all, to publish activities to activity streams. someone else might equally argue that they are simply vehicles for distributed CRUD of notes, which are the real content. the specs currently support both interpretations. should one be closed as a loophole? is one an intentional misinterpretation? which one?

@trwnh thank you so much!

@evan sorry, i have no idea how to interpret this response in context of what came before it. what am i being thanked for? i was expressing confusion at the original premise.

@trwnh I know you don't. I also know you love thinking and talking about dog basketball, and that you like open questions.

ActivityPub is a social API and federation protocol. One way to think about questions like the ones you ask is could it be applicable to an API or federation protocol?

In this case, yes. We already have a few activity types in the ActivityPub spec that aren't CRUD, and we have a whole Activity Vocabulary with even more. We can and should distribute them.

@trwnh we can also compare against existing social software.

@trwnh it's good to undercut assumptions. I appreciate that you have an incisive mind and that you don't accept axioms. It's a healthy practice.

@evan it wasn't intended as open question nor dog basketball. my point was we can either accommodate diverse viewpoints on what the "actual game" is and assume people are playing in good faith (and let them keep playing), or we close the door to those interpretations and uphold one of them as the correct interpretation (which might mean someone takes their ball and goes home). personally i'm team activity, but i'm not going to close the door on team note. i can't assume anyone else's position!

@evan but what specifically incited my comment was the bit about people's intentions. i again can't speak for anyone else's intentions, but whenever i get involved in any conversation, i'm doing it to work toward some kind of practical answer. i'd like to assume others are doing the same, and that people aren't generally participating in bad faith or simply to have conversations with no conclusion.

@trwnh absolutely, we can and should accommodate diverse viewpoints.

@trwnh so, when you introduced the question about activities being content, was that to help me get to an answer about the separate topic I had raised?

@evan i plead character limit and getting distracted by other things i'm doing rn, but i was using that as an example of a question that might *look* like an "intentional misinterpretation" but actually isn't. if you wanna publish activities then it's good faith to have discussions within the framework of publishing activities. but it could look like you're not playing "the actual game" by someone else's interpretation within a different framework.

@evan for me, "the actual game" is maximizing people's expressiveness and letting them describe their lives one activity at a time. the goalpost is high fidelity and low miscommunication. a different goalpost might aim to lower fidelity just to get something to show up. if any conclusion seems ridiculous, then it's either not as ridiculous as it seems within a different framework, or there's something else that could be done to make it internally consistent. it's driven by use cases.

@evan tldr i don't think it's healthy to look at these discussions through the lens that people are "intentionally misinterpreting the specs in order to derive ridiculous conclusions", merely "fun thought experiments" intended so that "we can't have a conversation about the actual game", "fight[ing] tooth and nail to keep those loopholes from being closed". that reading is tantamount to bad faith, whereas a good faith reading is that people interpreting the spec as best they can in their frame.

@evan@cosocial.ca

Did someone 'Well actually' the centaur tower already?
And if not, would you like me to?