Open akuckartz opened 10 years ago
Andreas, please also post this question to the mailing if you want it to be discussed. Only a few people follow the issue tracker whereas everyone follows the mailing list. Thanks!
Hi Andreas,
On Sunday, February 09, 2014 10:02 PM, Andreas Kuckartz wrote:
The Linked Data Platform specifies Container and Page ontologies. How can Hydra ontologies mapped to them (and vice versa)?
A LDP container also defines the interaction model. In contrast, a Hydra collection is just that, a collection. In Hydra, the interaction model is described separately by operations. That has the advantage that you can use the same mechanism to describe interaction models of other resources, to restrict it, to expand it etc. So, in a sense, LDP containers are specializations of Hydra collections. Does that answer your question?
Cheers, Markus
I think there is more we can say about the relation between ldp:Container and hydra:Collection. An LDP container as a membership subject and a membership predicate which members must match for being members. I think a hydra:Collection can be modelled as an ldp:Container whose membership subject is the container or the collection (thus the IRI thereof) and the membership predicate is something like xxx:isMemberOf or any property which is common for all resources in the collection.
:+1:
LDP Paging now a separate draft http://www.w3.org/TR/ldp-paging/
I just committed over Hydra CG & LDP NEXT CG mailing lists to help with further exploration of this topic: https://lists.w3.org/Archives/Public/public-hydra/2015Oct/0081.html
I also created a wiki page, please feel invited to contribute or propose another place to document this topic: https://github.com/w3c-social/Social-APIs-Brainstorming/wiki/LDP,-Hydra,-LDF
Seems like LDP Paging was published as a stable and complete note in 2015. Compatibility with it would be nice, but we should internalise the examples into the Hydra spec and convert them to JSON, since the Turtle and Link
header syntax will be alien to a lot of Hydra consumers.
Link header syntax will be alien to a lot of Hydra consumers.
Yes, we want the controls inside of the message!
Indeed! 😃
There's also the W3C Activity Streams Collection interface (which is used by the W3C Web Annotation Data Model Collection interface), that we should look into.
If we want to look at AS2.0 Collection & OrderedCollection, we should pay extra attention to paging. Each CollectionPage and OrderedCollectionPage relates to the collection with https://www.w3.org/TR/activitystreams-vocabulary/#dfn-partof
"@context": "https://www.w3.org/ns/activitystreams",
"summary": "Page 1 of Sally's notes",
"type": "CollectionPage",
"id": "http://example.org/collection?page=1",
"partOf": "http://example.org/collection",
"items": [
{
"type": "Note",
"name": "Pizza Toppings to Try"
},
{
"type": "Note",
"name": "Thought about California"
}
]
}
So each page (not the collection itself) has items
relationship with each element/member. Which seems to me like step backwards from the views based approach documented on https://www.w3.org/community/hydra/wiki/Collection_Design#Pagination
I had conversation about AS2.0 paging with some of the Annotation WG members in https://github.com/w3c/web-annotation/issues/50#issuecomment-151049075
The Linked Data Platform specifies Container and Page ontologies. How can Hydra ontologies mapped to them (and vice versa)?
https://dvcs.w3.org/hg/ldpwg/raw-file/default/ldp.html