Skip to content

Live testing of remote categories

ActivityPub Test Kategorie
63 10 2.1k
  • 0 Stimmen
    3 Beiträge
    1 Aufrufe
    julian@community.nodebb.orgJ
    douginamug@mastodon.xyz this entire series right here https://seb.jambor.dev/posts/understanding-activitypub/
  • What drew you to ActivityPub?

    ActivityPub activitypub dotsocial blogs
    5
    0 Stimmen
    5 Beiträge
    1 Aufrufe
    julian@community.nodebb.orgJ
    > Getting a critical mass of people to create yet another account was always a major obstacle. I see and have experienced this effect time and time again, and we're getting closer and closer to the point where the protocol implementations can abstract away the messy bits. Gaining critical mass among devs is the first step!
  • 0 Stimmen
    14 Beiträge
    185 Aufrufe
    julian@community.nodebb.orgJ
    renchap@oisaur.com edent@mastodon.social as it should, as the followers collection can be gamed and should not be trusted.
  • Backfilling Conversations: Two Major Approaches

    ActivityPub activitypub fep 7888 f228 171b
    26
    0 Stimmen
    26 Beiträge
    318 Aufrufe
    julian@community.nodebb.orgJ
    > One weakness I have noticed in NodeBB's current federation is that posts which are in reply to a topic (e.g. a Lemmy comment) show up as individual threads until (or if) the root post of that topic shows up in the local NodeBB. No, Lemmy does not implement either strategy, they rely on 1b12 only. If NodeBB is receiving parts of a topic that don't resolve up to the root-level post that might be something we can fix. I'll try to take a look at it.
  • 0 Stimmen
    17 Beiträge
    532 Aufrufe
    julian@community.nodebb.orgJ
    Hey rimu@piefed.social thanks for responding (and sorry for the late reply!) I am not married to the Announce([Article|Note|Page]) approach, so I am definitely open to Create([Article|Note|Page]) with a back-reference. I think I went the former direction because there is a known fallback mechanism — the Announce is treated as a share/boost/repost as normal. However, sending the Create also is fine I think. However, do we need a backreference? In my limited research, it seems that Piefed, et al. picks the first Group actor and associates the post with that community. If I sent over a Create(Article) with two Group actors addressed, could Piefed associate the post with the first, and initiate a cross-post with the remaining Group actors? Secondly, is how to handle sync. 1b12 relies on communities having reciprocal followers in order for two-way synchronization to be established. On my end since I know it is cross-posted I will now send 1b12 activities to cross-posted communities, but can Piefed, et al. send 1b12 activities back as well, in the absence of followers? cc andrew_s@piefed.social nutomic@lemmy.ml melroy@kbin.melroy.org bentigorlich@gehirneimer.de
  • 0 Stimmen
    8 Beiträge
    285 Aufrufe
    julian@community.nodebb.orgJ
    @evan@cosocial.ca hmm. I agree in the sense that any combination of recipients can be addressed, but the specific term "follower only" (to the exclusion of the public pseudo-user) isn't AP specific... could be wrong on that one. Either way I do think it's a good courtesy to assume equal or narrower visibility when replying to any post. The specific issue you outlined in OP seems to be a Mastodon bug for sure.
  • 0 Stimmen
    1 Beiträge
    34 Aufrufe
    Niemand hat geantwortet
  • Peertube....

    ActivityPub Test Kategorie peertube activitypub
    12
    1
    0 Stimmen
    12 Beiträge
    235 Aufrufe
    julian@community.nodebb.orgJ
    @mario@hub.somaton.com thanks, I tested that payload and it successfully edited the post, so at least from the payload point of view there is no issue. Perhaps there is a problem with the http signature on update? Is that handled differently than a create?