Let me say this back to see if I understand it:
VGR is proposing that someone could maybe hack up a site that would ingest blygs and sort the posts into topics, into \"subreddits\"? Or, rather than it figuring that out, maybe people tag posts in some way that would indicate that they should be considered on a portal like that?
Then, I presume, people would do upvotes and downvotes to get those posts more or less attention?
This seems like a really fun way to create a decentralized version of reddit. It also gives a new form of discovery that I particularly would like. Just opening up someone's feed and trying to decide if I like their vibe is a little hard.
Look, the one subreddit I can't quit opening is r/typwriters. If folks start posting photos of their old typewriters and routing them through Blygger reddit, then I'm going to look. You have at least one guaranteed audience member.
That said, I'm not 100% sure I'm understanding the idea of Blygger reddit. But if that's it, it seems like a good idea.
The Thicket could use a portal (or, even better, several).
Stubbing: Comments sections, multi-author blygs, forums
Comments sections, multi-author blygs, forums
\nAnyone who wants to do multi-author blygs, comments sections, slack-like things, or forums — you should be fully unblocked now. Have your agent read the relevant technical notes.
\nStudio will do multi-author eventually, but not the rest.
\nThere’s a definite opportunity now for someone to build a Blygger reddit evolution. Blyddit. Or a Blyscord or Blyack. You just build it in the obvious way, and flatten/discard the user and content hierarchies before putting it out on the wire. There’s probably a way to reconstruct both on the receiving end by piggybacking/tunneling sufficient topological information as user data, but I doubt anyone wants their reader clients to feature a dozen idiosyncratic topologies.
\nYou could however do a kind of network of blyddits (or blyscords or blyacks) as a subgraph of the thicket that all speak the same topology. Your challenge isn’t building it, but defining and getting people to adopt a shared spec for it. If you want to do this, as a sort of L2 spec riding the basic L1 blygger as a transport layer, talk to me. I can probably provide some input.
\nThe danger is trying to get one client to speak too many topologies (or L2 specs), producing a Babel UX, so I recommend against that if you’re building a client. I’m trying to make the Studio UX a sort of topologically unconstrained thicket-walker on the UX side, that only uses stubs, quotes, transcludes, pins, and forks to present the view. If the content also carries other supervening topology, this will be ignored.
\nIf you want to build higher abstraction layers, think of the thicket as a compile target. Clients that don’t speak your specific non-basic topological language will have a more elemental, unfiltered experience. The way foreigners in a country where they don’t speak the local language find the higher abstraction stuff illegible but might be more attuned to the raw reality that natives have gotten blind to.
\n
—Brady Dale
November 7, 2026