It normally publishes about midday (UK time), but if the crawler fails for that edition, it picks up the next one at roughly 6 p.m., and then it gets into the habit of publishing then every day until I remember to set it back to midday.

Anyway, it looks like it’s just missed today’s 6 p.m. edition, but it might get the next one due around about now.

I read that tweet as something that wasn’t really about Fallout: New Vegas, and more as something using it as a vehicle for a joke (about adult women being nostalgic for the games they played as teenage boys).

I got the post I wrote there: https://pythag.net/post/4644

The actor thing might have just been a bug, because that post did have an actor in Create. It also doesn’t help that it seems to be a bit quiet there, unfortunately.

I’ve asked a question here: https://community.nodebb.org/topic/18547/any-plans-to-add-actor-to-create-activities-from-nodebb-users

I’ll see how it goes.

I’ve been playing around with it on my dev instance.

It’s possible to bring their forums in as communities, but they don’t supply an outbox, so it doesn’t come with any posts.

I added the ability to manually retrieve posts from there, including the ability to bring in some replies too (it’s currently set at 10). It’s quite nice that it’s convenient to also get replies (to replicate that with Lemmy, we’d need to query their API).

I subscribed to a community, but have had to hack around getting the first post that happened to show up to appear, because they don’t include an ‘actor’ in the Create object that’s embedded in the Announce. This is very unusual, and only seems to apply if the author was from nodebb, not Mastodon or Lemmy or whatever. I’ve signed up for an account there, to see if it also happens if a Create is sent directly (I suspect it does). As a new user though, I’m having to wait for my comment to be approved by a moderator (whereas with AP I can just yeet anything in there without delay).

I also spent some time banging my head against the desk, because it turns out posts from there are very similar to Poll Votes from Mastodon.

So, yeah - that’s a couple of things to chat about. It would be better if they were a bit more like other established platforms, but there’s already some inter-op: e.g. here and here

EDIT: my comment from there was approved, but nothing federated out. Looks like they have more work to do on this.

That community is deleted as far as anyone else can see (try your link with Private Browsing, or when logged out). It will remain visible to you, to allow you to change your mind and restore it, for 7 days I think.

Therefore, you should be able to start up a new community with the name you want, and just forget about the old one.

If you fetch a community that your instance hasn’t previously heard of, you can typically query the community’s ‘outbox’ collection to get recent posts. So in Lemmy, you get 50 old posts, and then - once someone has subscribed - new posts start coming in.

Different platforms have different formats for their outboxes - Lemmy uses Announce/Create/Page, a.gup.pe and PeerTube use Announce, with a URL that leads to a Note or Video, wordpress uses Create/Article. Because Lemmy already understands its own outbox format, it’s able to get old posts from other Lemmy instances. It doesn’t get old stuff for a.gup.pe, PeerTube, or wordpress though.

So you might be wondering what outbox format nodebb uses - to which the answer is none. The outbox leads nowhere useful (they’re in good company with MBIN on this). Anyway - this is why fetching a nodebb community won’t come with any of its existing posts (but - as mentioned - new stuff will come in for subscribers)

Season 3 of a TV show comes with a significant wage increase for everyone involved, so 3 seasons (at least) is something that the sellers of a show always want, but the buyers are trying to avoid.

On Netflix, it’s become a pattern of all shows only getting 1 or 2 seasons, unless they’re mega-hits, or dirt-cheap to produce in the first place.

How well a show wraps up after 2 seasons often depends on how much the writers want to do the streamer’s job for them. Tokyo Vice was a (rare) example of a good, self-contained, 2-season show.

naked mole rat as their fursona

Would probably lead to your application getting rejected. I’ve just looked, and they’re also known as ‘sand puppies’, so it’d be better go with that.

A comment here distinguishes between the ‘plain text’ that’s allowed by the spec, and MarkDown as a markup language (it’s confusingly named, I guess, but that’s what Wikipedia categorises it as too)

Lemmy mostly federates with Lemmy, but everything else out there (PeerTube, PixelFed, etc) has been developed to work well with Mastodon (sometimes to the Fediverse’s detriment), so your Mastodon account should be all you need.

Another commenter (who’s contributed code to Lemmy) pointed to a link that provides the specification for that field: “A simple, human-readable, plain-text name for the object. HTML markup MUST NOT be included.”

So in this case, it’s more that the JSON looks a bit ambiguous: ‘mediaType’ is only referring to the format of the text in a post’s body, but - unlike me - you’d also need to be aware of the spec to know that it doesn’t apply to the title.

Oh, wow. Thanks.

For clarity, I wasn’t intending to say that PieFed treats that field as HTML (it treats it as text), I just meant that if you were looking at that JSON, and being a bit lazy like me and not looking at specs, then it wouldn’t be unreasonable to assume that the ‘mediaType’ field also refers to ‘name’ (rather than a ‘content’ field which this post doesn’t happen to have).

Anyway, this seems to be even more reason why MD shouldn’t be put in titles, and front-ends shouldn’t be encouraging the practise by rendering it.

Yes - it’s easy to do from a command line. For this post, it would be:

curl --header 'accept: application/activity+json' --location https://lemmy.world/post/24241974 | jq .

{  
  "@context": [  
    "https://join-lemmy.org/context.json",  
    "https://www.w3.org/ns/activitystreams"  
  ],  
  "type": "Page",  
  "id": "https://lemmy.world/post/24241974",  
  "attributedTo": "https://lemmy.world/u/amon",  
  "to": [  
    "https://lemmy.world/c/fediverse",  
    "https://www.w3.org/ns/activitystreams#Public"  
  ],  
  "name": "By the way, you can have `Markdown` in Lemmy post titles",  
  "cc": [],  
  "mediaType": "text/html",  
  "attachment": [],  
  "commentsEnabled": true,  
  "sensitive": false,  
  "published": "2025-01-13T20:48:50.824942Z",  
  "language": {  
    "identifier": "en",  
    "name": "English"  
  },  
  "audience": "https://lemmy.world/c/fediverse"  
}   

You can, but maybe you shouldn’t. Given that this post is in the fediverse community, I don’t feel too bad about mentioning that Lemmy is part of a federated network with PieFed and MBIN (I try not to bollock on too much about the platform I happen to be using).

In the ActivityPub JSON for this post, there is no indication that this field contains MarkDown. If anything, it says the opposite, it says it contains HTML. It’s therefore not unreasonable for other platforms to render it as such.

Given this, and the poor support for mobile clients indicated in the comments, and the fact that it’s only a subset of MarkDown tags, but include ones that aren’t part of CommonMark standard, I’d argue that it’s not necessarily a good idea.

You can’t. resolve_object in the API needs a full URL - something you could also put in a browser to uniquely identify an object - a user, community, post, or comment, but not a modlog entry.

You’ll have to get the mods / admin at blahaj to flip the ban again.

Other instances have recorded that particular unban, so it’s possible that - at the time it was sent, lemmy.nz rejected it because the community didn’t have any local subscribers. So, considering the activity, it would be a lemmy bug. Maybe not though, maybe lemmy.nz was just down at the time. Anyway, it’s showing 1 local subscriber now, so assuming that’s a different account, another attempt should be acknowledged.

OP won’t see your image unfortunately, because of a difference in how embeds are handled (Mastodon’s use of image attachments to Lemmy’s use of MarkDown has been fixed in one direction, but not the other).

For anyone wondering what the image was, it’s here: https://lemmy.dbzer0.com/pictrs/image/5854765e-4707-4b21-b93f-00f3d781ad71.webp

Also, since I’m not replying directly to OP, I think I also have to do this: @mho@social.heise.de

The top ones I’m aware of are Veronica Explains and The Linux Experiment.

I also like ctrl-alt-rees

I see this has already been answered, but I’ll post what I was typing out anyway.

It’s not enough to just provide your auth token if you want to fetch the details for yourself (even though the JWT decodes to identify you). You have to use /user like you would for fetching anyone else, e.g.:

curl --header 'accept: application/json' --header 'authorization: Bearer xyzyzyz' --location https://lemmy.world/api/v3/user?username=okelote360 | jq .

Yeah, as mentioned, Lemmy seems to have made a mess of bringing that community in. It looks like they got the community details - and know that posting is restricted to the mod, but they didn’t fetch who the mod is. This meant that posts in the outbox were rejected, and all subsequent activity (that would normally cause them to resolve the missing posts) is being rejected too.

Screenshot (these activities are all Likes and Comments):

Full-size image link: https://postimg.cc/3481xLdq