On Mon, 8 Jul 2002, Diane I. Hillmann wrote:
> At 09:47 PM 7/7/2002 +0100, Andy Powell wrote:
>
> >I was chair of the DC Format WG which defined the various qualifiers for
> >dc:format. Although the wording isn't perhaps quite as clear as it might
> >have been, my recollection is that dcterms:medium was *only* intended for
> >use with physical resources (e.g. a valid value might be 'canvas' for a
> >painting).
> >
> >A MIME type should be given using dc:format and the IMT encoding scheme
> >(i.e. *not* using a refinement of dc:format). I presume this means using
> >the rather verbose RDF encoding
> >
> ><dc:format>
> > <dcterms:IMT>
> > <rdf:value>text/html</rdf:value>
> > </dcterms:IMT>
> ></dc:format>
> >
> >which you seem to reject for date elsewhere in your proposals?
>
> Andy:
>
> I was at some of the meetings you chaired, and though I do recall that the
> need for a place for physical medium was the impetus behind that qualifier,
> I'm not sure that there was consensus on the restriction you recall. In any
> case, even if that had been the intention, I'm not sure the term "medium"
> can be restricted so easily. While the qualifier definition certainly
> emphasizes the "physical" medium, the presentation of these terms at
> http://dublincore.org/documents/dcmes-qualifiers/ (particularly the chart)
> rather implies that format can be either extent or medium, and that medium
> can have derive a vocabulary from IMT.
This was raised (by Rebecca) during the DC-Libraries discussions. The
message I sent at the time follows. In short, the DCQ recommendation is
*wrong* to indicate that IMT can be used with dcterms:medium. We need to
correct it. I recommend that the table at the top of the DCQ document is
corrected and a new version issued (the table is non-normative so this
isn't too drastic IMHO). I apologise for having allowed this to slip by
unnoticed at the time :-(
--- cut ---
> From [log in to unmask] Mon Jul 8 16:21:21 2002
> Date: Mon, 25 Feb 2002 22:48:57 +0000 (GMT)
> From: Andy Powell <[log in to unmask]>
> To: [log in to unmask]
> Subject: Format/medium confusion (was DC-Q Email Ballot #2 (fwd))
>
> All,
> I've checked back thru my email logs (there's an awful lot of them!)...
>
> The final format-related vote is shown below.
>
> We voted on a proposal that clearly indicated that 'medium' is only
> applicable to physical resources and that 'IMT' is only applicable to
> virtual resources.
>
> Apologies for mis-representing this in my previous email to Rebecca. As
> chair of the format group, my memory should have been better!
>
> So, the definitions in the DCQ document are correct, but the table at the
> top of the document is incorrect. 'IMT' should *not* be shown as a valid
> qualifier for 'medium' - it is only a valid qualifier for 'format'.
>
> I'm not sure what we do now - recommend a revision of the DCQ document?
>
> Andy
> --
> Distributed Systems, UKOLN, University of Bath, Bath, BA2 7AY, UK
> http://www.ukoln.ac.uk/ukoln/staff/a.powell +44 1225 383933
> Resource Discovery Network http://www.rdn.ac.uk/
>
>
> ---------- Forwarded message ----------
> Date: Thu, 11 May 2000 07:36:00 -0400
> From: "Weibel,Stu" <[log in to unmask]>
> Reply-To: [log in to unmask]
> To: "[log in to unmask]" <[log in to unmask]>
> Subject: DC-Q Email Ballot #2
>
>
> 2nd DC Qualifier Editorial Email Ballot (Format only)
>
> The following is a proposed modification to the DC Qualifier Ballot. The DC
> Qualifier specification will be modified as indicated by the results of
> this ballot.
>
> Please review' at your earliest opportunity. Discussion or
> lobbying for a particular result should take place on DC-Usage list.
>
> Deadline for balloting: May 24
>
> Two thirds majority required to change the result of the original ballot
>
> --------------------------------------------------------
>
> 1. Format qualifiers.
>
> Discussions among some members of the Format WG resulting from the Qualifier
> ballot suggest that it is necessary to distinguish among three things:
>
> 1) genre - the representational class of the resource
> 2) medium - the physical carrier of the information
> 3) encoding - the way in which the data is encoded on the medium
>
> Typically description schemes (including DC) have conflated these three
> categories into two, with a resulting redundency or confusion (for example
> Text/HTML from IMT... a mix of genre and encoding format).
>
> By revising the definitions very slightly, we can distinguish fairly cleanly
> among these three categories within DC:
>
> | Genre | Physical | Encoding |
> | | Medium | |
> -----------------+--------+----------+------------+
> DC.Type | X | | |
> -----------------+--------+----------+------------+
> DC.Format | | | |
> refined by | | X | |
> 'Medium' | | | |
> -----------------+--------+----------+------------+
> DC.Format | | | |
> refined by | | | |
> encoding scheme | | | X |
> IMT | | | |
> -----------------+--------+----------+------------+
>
> The definitions for two of the existing Format qualifiers (Medium and IMT)
> have been revised to reflect this. This proposal clearly separates the
> physical material or carrier of the resource (Medium) from the way in which
> the resource is encoded on the material or carrier (Encoding). Thus, Medium
> is applicable only to physical resources and the IMT encoding scheme
> provides a well-accepted (if not perfect)enumerated list of encodings for
> Internet resources.
>
> This proposal does not solve every possible point of confusion, but it is
> perhaps the clearest resolution to date of this problem that plagues all
> extant description schemes.
>
> Note that acceptance of this proposal supercedes issue 3 on the previous
> email ballot... this is important, because 'Medium' as a representation of
> physical manifestation in this proposal is distinctly NOT media type. It
> also supercedes the wording of IMT (issue 7) on the previous email ballot,
> but this proposed changed
> is simply a correction to the definition of IMT raised by Priscilla Caplan
>
>
> Proposed Revisions:
>
> Element Refinement for 'Format': Medium
> Label: Medium
> Name: medium
> Definition: The material or physical carrier of the resource.
> Type: Element Refinement
>
> Encoding Scheme for 'Format': IMT
> Label: IMT
> Name: IMT
> Definition: The Internet media type of the resource.
> Type: Encoding Scheme
> See also:
> http://www.isi.edu/in-notes/iana/assignments/media-types/media-types
>
>
> ----- cut and submit vote to DC-USAGE ----
>
> [ ] I support the original definitions as modified in the previous email
> ballot
>
> [ ] I support the proposed revisions as described in this ballot
>
> [ ] I do not support either position for the reasons specified below:
>
>
> other comments:
>
>
> --- cut ---
>
--- cut ---
> In the project I'm working on now, we've been trying to set up some methods
> for "normalizing" large gobs of undifferentiated data
> meaningfully. Knowing that data in a format element that comes in with a
> qualifier "IMT" is not about extent is useful enough that we're inferring
> from it that the information is about the medium (broadly speaking).
It is not about the 'medium' as defined in DCQ. IMT indicates an Internet
Media Type. The 'medium' as defined by DCQ is
The material or physical carrier of the resource.
which is quite clearly (to me) only applicable to physical resources.
> If you think we're on the wrong track here, we may have to reconsider that
> approach. And for sure, this should be explained in "Using Dublin Core,"
> since we currently have nothing to say about the issue there.
It's not clear to me whether you are on the wrong track or not. It would
certainly be helpful to get the table in DCQ updated and a note about this
added to the 'Using DC' document.
Andy.
> Diane
>
>
> *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
> Diane I. Hillmann
> Project Manager & Metadata Specialist
> National Science Digital Library Project at Cornell
> Department of Computer Science Voice: 607/255-5691
> 417 Rhodes Hall Fax: 607/255-4428
> Ithaca, NY 14853 Email: [log in to unmask]
> *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
>
Andy
--
Distributed Systems, UKOLN, University of Bath, Bath, BA2 7AY, UK
http://www.ukoln.ac.uk/ukoln/staff/a.powell +44 1225 383933
Resource Discovery Network http://www.rdn.ac.uk/
|