Re: Link: grandfathered format-specific registered relations

On Sat, Aug 22, 2009 at 4:06 AM, Roy T. Fielding <fielding@gbiv.com> wrote:

> On Aug 21, 2009, at 6:46 PM, Mark Nottingham wrote:
>
>  So what does 'edit-media' mean when it's used with a non-Atom relation?
>>
>
> You mean "when it's used within a non-Atom format document, right?
> It means the same thing it does within an Atom format document.
>

RFC 5023 defines edit-media as "an IRI that can be used to modify a[n
associated] Media Resource" - there is nothing Atom-specific about that and
as a generic relation it seems like something that could be useful for OCCI
too (or more importantly something I'd have to redefine for OCCI if Atom
were able to claim a monopoly on this relation - e.g. occi-edit-media?).
Similarly, I'm not the only one thinking about what JSONpub might look
like<http://twitter.com/pkeane/status/1941800604>,
in which case we would need to create another [set of] relation(s) for it.

Whether or not that meaning is useful is presumably decided by the
> person who stuck it there.
>

Agreed - I don't see much risk in taking context into account (e.g. the
format of the document you're parsing) when deciphering relations.

We could even pre-define some relations of our own with a view to focusing
future development efforts - for example, 'start' and 'stop' relations are
intensely interesting for our WG but could be useful for the web of things
too (imagine turning on and off machines - even simple ones like light
bulbs) - and the semantics are obvious. I'd much rather see sensible generic
terms adopted than things like 'light-on' and 'light-off'.

Sam