@puppygirlhornypost2 @trwnh @cwebber that’s flatly illegal. id must be a string that encodes a URI.

@erincandescent@akko.erincandescent.net @trwnh@mastodon.social @cwebber@social.coop OH I know why i am confusing this i am using AS logic not AP

@puppygirlhornypost2 @trwnh @cwebber no its flat out illegal AS2 (and indeed flat out illegal JSON-LD), AP imposes no additional requirements here.

@erincandescent@akko.erincandescent.net @trwnh@mastodon.social @cwebber@social.coop im so confused to why it specifies Object | Link here then unless I am misinterpreting what that means which is always possible

@puppygirlhornypost2 @trwnh @cwebber Domain is the types this property applies to

Range is the type of the value(s) of this property

@puppygirlhornypost2 @erincandescent could, but it doesn't make for semantics. what does it mean to reply to a Link instead of replying to a Note? like what's the link context there. we know the link target is the as:href and we might have some link relation declared as the as:rel, but what does the arc of that graph edge look like? can you even draw such an edge? technically you are not relating A to B, you are relating A to an intermediate blank node whose href is B. how to reason around that?

@puppygirlhornypost2 @cwebber @trwnh Anyway the explainers for “id” and “type” are a bit wonky because they’re special magical RDF terms unlike anything else and so their domain is actually everything and their range is always a URI (or in the case of type: one or more URIs).

@erincandescent@akko.erincandescent.net @cwebber@social.coop @trwnh@mastodon.social well that certainly doesn't help with confusion lol

@erincandescent@akko.erincandescent.net @cwebber@social.coop @trwnh@mastodon.social Here we are I have found a much more valid example. https://www.w3.org/TR/activitystreams-vocabulary/#dfn-subject a subject can be either.

@puppygirlhornypost2 @erincandescent in rdf you have the concept of "type" which points to a Class. if everything is a triple of the form "alice knows bob", then we can start to describe each term individually:

- alice isA Person
- bob isA Person
- knows isA Property
- Person isA Class

as2.0 "id" is just a straight alias for jsonld `@\id` and that basically just indicates the subject for that node. you can reconstruct statements of the form "id property value"

as2.0 "type" is rdf:type p much.

@puppygirlhornypost2 @erincandescent @cwebber actually speaking of as:subject and the entire as:Relationship model

why reinvent rdf triples? reifying a Relationship is basically exactly the same as reifying a Statement. https://www.w3.org/wiki/RdfReification

as:subject == rdf:subject
as:relationship == rdf:predicate
as:object in this case is used equivalently to rdf:object despite not being semantically equivalent????

@trwnh @puppygirlhornypost2 @erincandescent oh god I don't want to talk about the linked data and activitypub and etc stuff more than I have to on here because people just end up being really nasty from either pro-RDF or anti-RDF camps

But hey, since ActivityPub is using json-ld, technically it's the largest deployed "linked data" or "semantic web" system ever

So there

@cwebber @puppygirlhornypost2 @erincandescent no nastiness intended! just a lot of confusion about the intentions and (RDF entirely aside) the data model itself, especially as defined in the vocabulary! i'll untag you going forward if this conversation proceeds, dw

@cwebber@social.coop @trwnh@mastodon.social @erincandescent@akko.erincandescent.net the only real opinion I have on rdf is that I am glad we use json-ld instead of the xml representation because based on the Wikipedia page’s example of rdf in xml it’s fucking disgusting and I say this as someone who generally doesn’t hate xml

@cwebber@social.coop @trwnh@mastodon.social @erincandescent@akko.erincandescent.net specifically I mean that representing AP within RDF/XML would be horrendous for both implementations and for people trying to learn about how these sort of things are represented.

<?xml version="1.0" encoding="utf-8"?>
<rdf:RDF xmlns:contact="http://www.w3.org/2000/10/swap/pim/contact#" xmlns:eric="http://www.w3.org/People/EM/contact#" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
  <rdf:Description rdf:about="http://www.w3.org/People/EM/contact#me">
    <contact:fullName>Eric Miller</contact:fullName>
  </rdf:Description>
  <rdf:Description rdf:about="http://www.w3.org/People/EM/contact#me">
    <contact:mailbox rdf:resource="mailto:e.miller123(at)example"/>
  </rdf:Description>
  <rdf:Description rdf:about="http://www.w3.org/People/EM/contact#me">
    <contact:personalTitle>Dr.</contact:personalTitle>
  </rdf:Description>
  <rdf:Description rdf:about="http://www.w3.org/People/EM/contact#me">
    <rdf:type rdf:resource="http://www.w3.org/2000/10/swap/pim/contact#Person"/>
  </rdf:Description>
</rdf:RDF>

@cwebber@social.coop @trwnh@mastodon.social @erincandescent@akko.erincandescent.net one of the things I like about the json-ld representation is that unless we get into more complicated relationships (such as coincidentally the Relationship object) for the most part you don’t have to consciously think about it. With the xml way… oh god.

@puppygirlhornypost2 @erincandescent putting rdf aside entirely, i actually hate a lot of how the "plain json" of as2.0 looks like. you could remove the jsonld context entirely and you would be left with just horrendous data modeling and representation. people complain about rdf or ld but none of what they complain about is due to rdf or ld. if anything, a lot of the problems are things that context would solve, or at least expansion would solve. like imo we should be using expanded form instead

@trwnh@mastodon.social @erincandescent@akko.erincandescent.net I mean we’re going to have to agree to disagree there lol

@puppygirlhornypost2 @erincandescent ?? the lack of expanding ids is directly responsible for there being infinitely many ways to express the same thing. that is Infinitely Bad. built-in ambiguity is not a good thing for machines. normalizing/canonizing things ahead-of-time solves this. requiring everyone else to implicitly share the exact same context as you is what causes the issue, especially as you deviate away from a commonly-agreed-upon base. otherwise you need a central registry