I've mentioned a number of issues raised by draft-kunze-dc-00.txt - "Dublin
Core Metadata for Simple Resource Description". There follows a summary of
their current status.
Issue Status Note
----------------------------------------------------------------------------
1 "(" or "%" at start of element value Open
2 Characters which would break HTML Open
3 Role of the Appendix Open
4 Language codes Open See separate thread
5 Dates and times Open See separate thread
----------------------------------------------------------------------------
Stu wrote:
>Misha,
>
>the use of qualifiers and substructure in DC elements will not replace
>the unstructured use of the DC set. We have from the beginning
>expected at least two distinct modes of deployment:
>
> Unstructured free text in elements (RFC #1 is intended to provide
> sufficient technical detail to implement this approach).
>
> Scheme-qualified elements (possibly with subtypes) that afford
> more complex implementations (RFC #2 is expected to provide a guide
> to implement this approach).
This is good stuff, but it doesn't deal with the problems I raised.
>Your remarks about character set restrictions seem right on.
Do you mean issue 1 or issue 2 or both?
In each case, which approach would you prefer RFC #1 to take:
- To tell the user not to use certain characters (or maybe not to use them
in particular circumstances)?
- To tell the user how to use these characters?
Either approach would be reasonable. To do neither would be unreasonable.
Misha
|