On Monday, June 9, 2003, at 05:56 PM, Andy Powell wrote:
> On Sun, 8 Jun 2003, Liddy Nevile wrote:
>
>> I originally thought that we should think about solving the
>> accessibility community's problem of pointing to alternative resources
>> by using a new dc:relation:isEquivalentTo qualifier - now I am more
>> interested in having a dc:relation:isAlternativeTo qualification.
>> Equivalent and alternative content are not exactly the same but
>> equivalent content is alternative, so that works.
>>
>> Please comment on this.
>
> Sounds reasonable, but we already have isFormatOF and hasFormat (the
> latter being defined as
>
> The described resource is the same intellectual content of the
> referenced resource, but presented in another format.
>
> ) so why not use them?
The reason is because this is not for the case where it is simply the
same content in another modality - but where there is alternative
content - eg for a visionless version, it might be better to use
something else rather than just describe an image.
>
>> Instead, as conformance is often important, why not go for a new dc
>> element dc:conformsTo and then it is easily identified, useful in many
>> contexts, and can have suitable meaningful qualifiers such as
>> dc:conformsTo:accessibility (resource's accessibility) and
>> dc:conformsTo:metadata (resource's metadata) and
>> dc:conformsTo:encoding
>> (stuff about the resource with values like XHTML) etc...
>
> We already have dcterms:conformsTo, defined as
>
> A reference to an established standard to which the resource
> conforms.
>
> Provided that you can define a URI that corresponds to appropriate
> accessibility standard/conformance level I don't see any problem with
> using this?
>
I was thinking we'd use the existing conformsTo (in fact i was an
initiator of that) but now that DC is being proposed for wide use in
the context of conformance, and there is conformance across a range of
areas, it seems neater to have it as a separate element - for
conforming to accessibility, privacy, interoperability specs, etc etc
- and also there seems to me to be a need for both the declaration and
the certification - so it would all be too messy if it fitted inside
relation - and anyway,m I believe that does not work well when you have
to send the user agent/person off to somewhere else to interpret the
declaration.
I am not being assertive here - just explaining my thinking and glad to
be shown a better way.
> I don't underdstand what you mean by dc:conformsTo:metadata ?
>
Here I have been thinking about some way that we could say that our
metadata conforms to some standard - ... but the problem is that it is
the metadata not the resource that is conforming, so it does not make a
lot of sense to just shove that info into the dc:relation element
either - (again, I am not sure, just trying to work it out).
> dc:conformsTo:encoding is just dc:format isn't it?
Liddy
|