HN Explorer

Ask HN: Could something like HN work on the fediverse?

6 points by ParadiseEclipse ↗ HN
Recently reading up on things like Mastodon as a replacement for Twitter, and PeerTube as a replacement for Youtube. What about something as a replacement for things like reddit/HN? Is that feasible?

8 comments

[ 3.1 ms ] story [ 24.0 ms ] thread
SubGenius
Not the fediverse, but I'm working on moving Gurlic's backend to matrix and opening up the code. Which means people should be able to host their own matrix server and run a Gurlic instance. Still a heavy WIP though.

I don't see why it wouldn't work. Matrix is a solid protocol to work with. The only thing I want to be careful about is ensuring that the UI has none of the weirdness usually found in decentralized apps.

Jtsummers
Usenet (via NNTP) was pretty widespread at one point, and is akin to a federated/distributed Reddit. So I don't see why it couldn't technically work now. But you have to find a way to drive up adoption to hit a sufficient level of users and hosts.
simplecto
I've had this conversation a few times in the week or so. Usenet still exists, but to what extent I have no idea. It might be worth to try out Giganews or any number of other providers.
Jtsummers
A few newsgroups are still reasonably active. I still frequent comp.lang.lisp and comp.lang.ada. Though they now have tons of spam in them so the interesting parts are somewhere in the garbage. I think most groups have moved to other discussion boards/systems as their primary forum now, and newer groups don't try to create a newsgroup. For instance, I doubt there's a comp.lang.elixir or comp.lang.rust.

As a discussion system, it's pretty much defunct at this point.

kazinator
You don't see very much spam arriving at the Eternal September or AIOE news servers.

It's pretty much fine as a discussion forum.

Things have been lively in comp.lang.c recently with a little coding exercise going on and such.

comp.unix.programmer and comp.unix.shell are active also.

comp.theory has been quite active due to a crackpot who insists that he has solved the halting problem, or at least refuted common halting proofs using a nested x86 simulator that detects recursion, haha.