the git data model is a bunch of hash trees and hashed blobs. like wouldn't it be cool to idk represent a metadata graph that indicates hashed blobs just by extending the hooks for the data model because it was designed that way, to be understood by people who want a command and to mess around with it in unanticipated ways.

Show thread

also if anyone knows of things like that^ I'd love to hear it. trying to work on the tooling side of and start showing its use with a very concrete case of literal data formatting before moving on to more fun things like using it for interface design

Show thread

obvious rant without trivial solution: most of the time when i want to sent a file to someone on the same network i just make a torrent, flip on the tracker in my client, and send someone a magnet link. if it was THAT EASY to use then maybe more people would. !/

triplet links should actually be quadruplet links
author : object : subject : predicate
and like they are implicitly by the context of which URI they're under or otherwise in the context of their declaration but I'm like having that as an explicit part of the spec with authorship signing would make it about a billion times more useful !/

found it! I won't bother christine with an @ but for my future reference JONNY the post re: social challenged with people is HERE!

Show thread

I have asked and bookmarked it like a dozen times now, but does anyone remember that blog post where one of the authors of talks about the difficulty of the people and the people being friends??? !/

I am galaxy braining right now and drafting an interface to dissolve the document as such into recursive capability-based cells that can be continuous with activitypub (eg. by indicating that a document cell is in reply to post) and also be a generalized multi-interface representation of the underlying data. !/

finishing this section on and I think I need an adult. i mean i think i need a linguist. !/

the development of as a system to reconcile the fundamental contradiction between as a language-like system of vernacular expression and the desire for consistent automaticity is an extremely fascinating and subtle story of navigating design tension. I am still early in reading the history but it has taken me years to even appreciate why it's important. !/

I am once again, for maybe the 20th time, polling the room for any cultural histories of or , in the vein of Lindsay Poirier's "a turn for the scruffy," or more primary documents like blogs or mailing lists or anything you've got!!!! I have some of the broad outlines, but my overriding question is always "what the heck happened" THANK YOU !/

can anyone help me substantiate the claim that acquisition in 2010 was part of dominance over ? or am I way off? !/

just learned today that I'll be working on a project with maybe some resources come winter that I think fedi ppl might like. something something ... mebs we can put a team together ;) !/

any of my masto friends know of anyone who has used tools for management eg. for linking data to a deployable analysis pipeline/DAG? /! !/

where do I look for the transitional links between the / world and the development of ? People, writing, events, decisions, etc. I have been given cryptic hints that AP is where a lot of semweb energy ended up, but haven't been able to find much concrete information there. ty ❤️❤️

ok any or people: I've heard of the cultural rift that was "neat" v. "scruffy," but who can tell me about the focus between ppl focused on indexing and linking so ppl could use a on their own, vs. making automated inference/reasoning systems? one seems to be more profit-oriented/prioritized in industry, no?

A Fediverse instance for people interested in cooperative and collective projects.