@cwebber Is the AP outbox OrderedCollection supposed to list objects created by the actor directly, or list the "create" activities wrapping those objects?

@cwebber How do you think I could express the "unlisted" option in AP? I admit it's not a strong choice in the first place.. public but not listed in public timelines.. And yet it was always a choice for youtube videos, for example, so I think there is a use case. Would it be to-followers and cc-public?


@cwebber Oh also unclear how I could mark media as nsfw

@gargron We don't have a specific way to mark media as NSFW, but for now a good way might be to add a special tag to it with a common URI. Or we can add a new property?

I'm assuming you'll also want to deal with Content Warnings and that isn't in there yet; I've been thinking maybe we could make a special ContentWarning type for tags in AP; what do you think?

These might be worth filing as issues?

@cwebber the content warning is in essence a "summary" vs the "content", so it's already encodeable

@cwebber @gargron

For encoding CWs, I would recommending using a content and extendedContent property, with a human-readable link in the content that says "Show More"—that way it degrades gracefully on other instances.

@nightpool @gargron I don't think we have any extendedContent type thing in AS2 presently?

@cwebber @gargron That's what makes it degrade gracefully—you don't want users whose instances don't support content warnings to be exposed to the cut content. So absent something with a specific "click through to see this" semantic, it seems like you'd run into issues?

@nightpool @gargron I definitely get the desire to degrade gracefully here, though "summary" at least maps to the idea of "pre-content info", and making "content" now mean *either* full content or pre-content-info seems kinda like too much overloading to me...

@cwebber @gargron To me, "summary" means "you can use this *instead* of the content"—that is, in different parts of the application you would show one or the other (depending on how much space you have, for example).

I'm not sure it maps very cleanly onto "pre-content info"

@nightpool @cwebber example: summary "spoilers about American Gods 1x08" content: "blah blah blah"

@cwebber @nightpool i feel like that's ok and semantic and it's how mastodon does it with ostatus too

@gargron @cwebber yeah, you're probably right. I could see it going either way.

@gargron @nightpool It may be more than Mastodon wishes to do, but it would also be possible to stick a type on the summary field, like:

{"summary": {"id": "https://.../#cw", "type": "ContentWarning", "content": "spoilers about American Gods 1x08""}}

but maybe that's getting over-done. It would clearly delineate between summaries-as-blog-summaries/paper and summaries-as-in-content-warnings

@cwebber @gargron Question that came up yesterday in the chat—how do you resolve a fragment identifier? JSON-LD says "Fragment identifiers used with application/ld+json are treated as in RDF syntaxes, as per RDF 1.1 Concepts and Abstract Syntax [RDF11-CONCEPTS]"

but nobody knows what that means.

@nightpool @gargron I think the best thing to do is to fetcthe object at the uri, then search through the returned as2 object for the sub-object with the fragment. Similar to how you'd "jump" to the right section in html based on the fragment.

What do you think?

@cwebber @gargron You mean, the sub-object who's ID is equal to the full IRI you're trying to resolve?

@cwebber I don't *think* so? I just wanted to make sure I was understanding the spec correctly. It's a pretty ambiguous part of the RDF spec, so maybe a quick note of clarification would be useful.

Sign in to participate in the conversation

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!