14 messages in com.googlegroups.social-graph-apiRe: How to be a cooperative social ne...| From | Sent On | Attachments |
|---|---|---|
| Dale Newfield | 02 Feb 2008 09:47 | |
| Martin Atkins | 02 Feb 2008 09:58 | |
| Dale Newfield | 02 Feb 2008 10:02 | |
| Dale Newfield | 02 Feb 2008 10:08 | |
| Brad Fitzpatrick | 02 Feb 2008 11:53 | |
| Martin Atkins | 02 Feb 2008 12:15 | |
| Dale Newfield | 02 Feb 2008 12:42 | |
| Dan Brickley | 02 Feb 2008 14:29 | |
| Dan Brickley | 02 Feb 2008 14:38 | |
| Julian Bond | 03 Feb 2008 03:43 | |
| Martin Atkins | 03 Feb 2008 03:49 | |
| John Breslin | 04 Feb 2008 06:37 | |
| Brad Fitzpatrick | 04 Feb 2008 09:43 | |
| Dale Newfield | 10 Feb 2008 21:02 |
| Subject: | Re: How to be a cooperative social network? |
|---|---|
| From: | Martin Atkins (m....@degeneration.co.uk) |
| Date: | 02/03/2008 03:49:28 AM |
| List: | com.googlegroups.social-graph-api |
Dan Brickley wrote:
Actually I think we can tease apart the different aspects of XFN. The relationship types ie. the list of defined kinds of relationship that can exist between people (or agents) have utility outside of the specific HTML idioms promoted by microformats.org. In other words, there's XFN as a 'vocabulary' versus XFN as a 'notation'. I find value in the former mostly for non-public data, and in the latter as a quick authoring format for stuff I can import into RDF for SPARQL query and data merging. The applicability of the vocabulary shouldn't be limited by the scope of the notation; specifically we should be able to mix it with any other vocabulary for making claims, without needing the blessing of the notation-designers.
Certainly. It was the notation to which I was referring. This makes the assumption that a document will be self-describing; there's no allowance for storing this metadata in a separate document, and therefore you can't use it for metadata on things that are not documents.
RDF was of course design with this (storing the metadata in a decentralized fashion) in mind. However, having the metadata *in* the resource that they are describing is useful in that this data can be "trusted" to a certain extent. If I publish an edge from your mbox_sha1 to mine (as opposed to the other way around), it's probably better if that doesn't get returned from the Social Graph API!