[DDI-users] DDI v2-1: elements TXT and CONCEPT

[Mark Diggory]

DDI Users,

I've corrected the txt and concept tags in the cvs tree and the  
schema should be tested by someone to verify the changes.

http://ddi-alliance.cvs.sourceforge.net/*checkout*/ddi-alliance/ddi/ 
w3c/Version2-1.xsd

The differences can be verified here:

http://ddi-alliance.cvs.sourceforge.net/ddi-alliance/ddi/w3c/ 
Version2-1.xsd?r1=1.9&r2=1.10&diff_format=l

When evaluated and determined to be correct, this should replace the  
copy at:

http://www.icpsr.umich.edu/DDI/Version2-1.xsd

Sincerely,
Mark Diggory



On Jul 21, 2007, at 10:46 AM, Mark Diggory wrote:

> This appears to be a change that Matt Richardson and Sandra Ionescu
> has put into the schema (see revision 1.4 in the list.)
>
> http://ddi-alliance.cvs.sourceforge.net/ddi-alliance/ddi/w3c/
> Version2-1.xsd?view=log
>
> Sanda and Matt, can you comment on this change in November 2005 which
> appears to deviate from the DTD?  I have a email archive that from
> that time period I've reviewed and not been able to determine if
> there was discussion concerning it?  Was there consideration if this
> is an error in the DTD technical Implementation of the Specification
> as well? Or was this even an appropriate change to have made in the
> w3c Schema implementation?
>
> -Mark
>
>
> On Jul 12, 2007, at 2:49 PM, Wendy Thomas wrote:
>
>> For DDI 2.1 the cannonical version is the DTD. I will ask the guys
>> handling this on the SRG to look into the problem and get back to the
>> list. (they may be doing it already :)
>>
>> Wendy
>>
>>
>> On Thu, 12 Jul 2007, Fredy Kuhn wrote:
>>
>>> Hi all,
>>>
>>> I have a problem with the usage of sub element pairs 'txt' -
>>> 'concept' in DDI 2-1:
>>>
>>> The web pages 'Tree Structure' (*) and 'Whole Library' (**)
>>> display in DDI section 2 several of these pairs; example:
>>> <dataKind> 2.2.3.10.
>>>
>>> In schema 2-1, and according to Sourceforge history (***), these
>>> sub-elements were removed from study description elements, see
>>> revision 1.4:
>>>
>>> --- QUOTE ---
>>>
>>> The <txt> - <concept> pair illegally appears under <nation>,
>>> <geogCover>, <geogUnit>, <anlyUnit>, <universe>, <dataKind>,
>>> <timeMeth>, <sampProc>, <collMode>, <resInstru> in Study Desc.,
>>> also under <universe> 4.1.5 ... ...
>>>
>>> !!!----- <txt> and <concept> are VALID FIELDS under <varGrp>
>>> (4.1.2 and 4.1.3) and nCubeGrp (4.2.2 and 4.2.3) --------- !!!
>>>
>>> --- QUOTE END ---
>>>
>>> Thus, the TXT and CONCEPT tags are valid in DTD2.1, but not in
>>> schema2-1. Should we use them or not?
>>>
>>> Further more, by removing the sub-element <concept> from e.g.
>>> <datakind> or <sampProc>, how to markup the usage of a controlled
>>> vocabulary?
>>>
>>>
>>> (*) Tree structure:
>>> http://www.ddialliance.org/DDI/dtd/version2-1-tree.html
>>>
>>> (**) Whole library:
>>> http://www.ddialliance.org/DDI/dtd/version2-1-all.html
>>>
>>> (***) Sourceforge DTD v2.1 (see revision 1.4):
>>> http://ddi-alliance.cvs.sourceforge.net/ddi-alliance/ddi/w3c/
>>> Version2-1.xsd?view=log
>>>
>>>
>>> BTW: Looking at DDI-3, I'm asking me how elements for which we use
>>> a controlled vocabulary (CV), e.g. 'data collection mode', should
>>> be marked up, i.e. how to specify the CV. Could one of the
>>> 'experts' give me a hint? I'm not yet too familiar with DDI-3.
>>>
>>> Best regards
>>> freddy

~~~~~~~~~~~~~
Mark R. Diggory - DSpace Systems Manager
MIT Libraries, Systems and Technology Services
Massachusetts Institute of Technology