[tei-council] @cRef survey

[James Cummings]

Could you make sure that the Library SIG people circulate it? Not 
sure if they use it at all, but one of the comments on my P4 
Survey results were that many Library people probably don't see 
it because they don't read TEI-L.  (I don't think this will 
change your results in the slightest however...)

-James


On 07/02/12 13:47, Martin Holmes wrote:
> Hi all,
>
> I've had only nine responses to my @cRef survey so far, which suggests
> it's not in very wide use. People using it tend to be using it slightly
> incorrectly -- sometimes they're using canonical references with spaces
> in them, or they're using #id-in-different-document or
> #id-which-dunt-exist, and in no case has anyone created a<cRefPattern>.
>
> So if we deprecated it, we'd annoy a small number of people, but there's
> no fix we could make to it that would bring them all into the fold of
> correct usage anyway. Usage is as much of a mess as the definitions are.
>
> To deal with this issue, as well as that of @key, I think we need to
> take three steps:
>
> 1. Develop a proper strategy for the use of private URI schemes (with
> something similar to<cRefPattern>  where people can properly document
> how their private schemes are to be dereferenced or understood). This
> should divert a large proportion of the current usage of @cRef (and
> @key) over to @ref and @target.
>
> 2. Decide whether we want to keep @key and @cRef, or deprecate one or both.
>
> 3. If we're keeping them, decide what their datatypes should be. In the
> case of @cRef, create a proper class for it.
>
> 4. If we're not keeping them, do we need replacement attribute(s) for
> their use-cases?
>
> Cheers,
> Martin


-- 
Dr James Cummings, InfoDev,
Computing Services, University of Oxford