What is Nostr?
mikedilger /
npub1acg…p35c
2023-04-26 22:31:25

Update on Gossip Development and what I'm up to

I spent an hour this morning preparing for packaging of gossip 0.6 (which is not yet released). Generally I work through the dependencies that I have forked and update each one's upstream dependencies and run it's tests, and bubble up the chain until gossip itself has it's dependencies updated. That part is done, so now it is on to compiling on each system and then building the outputs into packages for each architecture.

I just bought a bottom-of-the-line Apple Mac Mini (M2 based though) that I don't even want for anything other than compiling, just so that I can compile and release for Apple silicon more easily, without getting hung up in all the cross-compile or virtualization issues (believe me there are many, and I tried a lot of options before deciding it was too complex and error-prone, and I would still need extra tools to create a '.pkg'). If you run gossip on Apple and want to use the packaged version when it is available, you really ought to throw me a zap as I went NZ$1067 out of pocket for you all (of course you are not obligated to do anything).

This upcoming version will have video thanks to bu5hm4nn (npub1hlq…z5wg) . If you want to try the 'unstable' branch right now, you can have video right now.

I'm sorry about zaps. I still haven't made time to finish the work, and the weighty decisions about when to fetch which data don't have obviously correct answers, and when I'm not clear whether I'm doing it the best way it makes me hesitate. There are only a few other things that I really want gossip to have soon, such as 'directory' relays to lookup relay list metadata on.

After that I think I will start work on gossip-personal-relay which will be a linux-only server using mmap (for events) and sled+abomination for indexes (no SQL). Because without this personal relay the gossip relay model (and NIP-65) isn't really complete. And I came here to solve the "fully distributed and censorship resistant social media microblogging problem" (not the zap people bitcoin problem, or the github-on-nostr problem or any of a number of other problems), and I can't rest until that problem is fully solved to my satisfaction.

And even though I intend all that, I've got a lot going on that has slowed down my nostr work quite a bit. But I'm still here and I think development will pick up again soon.
Author Public Key
npub1acg6thl5psv62405rljzkj8spesceyfz2c32udakc2ak0dmvfeyse9p35c