Re: [weirds] switching to RFC 5988 link type registry

[Chris Wright <[e-mail]>]

Adding my support to this - I thinks its a great idea

Thanks

Chris
On 11/10/12 12:20 AM, "Andy Newton" <[e-mail]> wrote:

>All,
>
>We currently define a structure for linking to registry related things
>using URIs. It looks like this:
>
>"uris" :
>  [
>    {
>      "type" : "source",
>      "uri" : "http://whois-rws.net/contact/xxxx"
>    },
>    {
>      "type" : "held",
>      "uri" : "http://example.net/location/xxxx"
>    }
>  ]
>
>
>And we define the types in Appendix A.3
>(http://tools.ietf.org/html/draft-ietf-weirds-json-response-00#appendix-A.
>3
>).
>
>Does anybody see a downside to switching to a more Atom like approach and
>re-using link types from the RFC 5988 IANA registry? In other words,
>instead of doing our own thing move to a pre-established method.
>
>Given the entries in the 5988 registry and what we have, here's the
>re-arrangement I see:
>
>'self' - use 'self' from 5988 registry.
>'parent' - use 'up' from 5988 registry.
>'source' - use 'alternate' from 5988 with an appropriate media type in the
>'type' parameter.
>'web' - use 'alternate' from 5988 with text/html or text/xhtml or whatever
>media type
>'rdap' - we may have to register something here, but that may also be an
>indication that this type is not needed.
>'held' and 'geo' - let the GEOPRIV wonks register these as they are
>outside our bailiwick.
>
>Comments and suggestions?
>
>-andy
>
>_______________________________________________
>weirds mailing list
>[e-mail]
>https://www.ietf.org/mailman/listinfo/weirds