Re: First draft of ActivityPub HTML Discovery report

On Sat, Nov 16, 2024, 06:51 Melvin Carvalho <melvincarvalho@gmail.com>
wrote:



so 16. 11. 2024 v 4:30 odesílatel Evan Prodromou <evan@prodromou.name>
napsal:

Hello, all. There's a first draft of the ActivityPub HTML Discovery report
here:

https://swicg.github.io/activitypub-html-discovery/


This looks wrong to me:

{
  "@context": "https://www.w3.org/ns/activitystreams",
  "id": "https://ap.example/some/path/person-1.jsonld",
  "type": "Person",
  "name": "Person One",
  "url": "https://html.example/profile/person-1.html"
}

It seems the id is tying the person’s identifier to the file name
(person-1.jsonld), which feels like a conflation of concepts. That said, I
assume this is an intentional design choice—perhaps a feature, not a bug?,
that AP will just have to live with


my understanding is that some examples are meant to demonstrate cases where
content negotiation is *not* used, or where the jsonld/html documents are
served by different servers, possibly as static files with no rewrite rules.

in other words, the assumption for this example is that ap.example and
html.example are two servers with very basic configurations that don't
include rewrite or try_files directives.