Hi everyone,

The issues that Lorna raises about how to implement a faceted vocabulary
in practice are of course very important but are secondary to decisions
about the composition of the vocabulary, and the two issues need to be
considered separately.

How the final vocabulary is used by cataloguers will in many cases
differ from how it is presented to searchers and is entirely dependent
in both cases upon the context of that use.

There are a number of precedents for using such a vocabulary for both
cataloguing and searching within the RDN going way back to the ROADS
eLib project and I'm sure scores of other examples could be cited
contributors to this list :)

It is technically relatively easy to present a useable facted vocabulary
for cataloguing (where the hierarchy is presented) and flatten that out
into a single list of terms for presentation to a searcher.

I like Aida's model as it seems the best fit for producing rich
descriptions. I personally believe that dumbing down the vocabulary to
make things simple for the end user would be a mistake if it meant that
the descriptions resulting were poorer and the vocabulary is a kludge.

I'm sure we can achieve a faceted vocabulary that is *both* richly
descriptive and useable for cataloguing and searching.

Cheers.

Lorna Campbell wrote:

>
>Date:    Fri, 5 Sep 2003 12:05:01 +0100
>From:    "Lorna M. Campbell" <[log in to unmask]>
>Subject: Re: learningResourceType vocabulary discussion
>
>Hi Aida,
>
>Thank you very much for your mail. You certainly haven't missed the
>point and your comments add significantly to the debate.  The technique
>that you have outlined below has already been considered by various
>groups who have tried to address this particularly thorny issue.  I
>absolutely agree that this is the obvious ideal solution although a
>couple of issues remain.  Clearly we would need to create the two facets
>of the vocabulary and decide on their terms, this in itself could
>constitute a large, but in my opinion, very valid research project.  We
>also need to consider how we would present such a  faceted vocabulary to
>metadata creators and searchers. Would the vocab be presented as two
>separate lists with at least one classification from each list being
>mandatory?  Or would it be better to present users with a single list?
> Any further comments or suggestions on the practical creation and
>implementation of a faceted vocabulary of this kind would be greatly
>appreciated.
>
>Thanks again
>Lorna
>
>Aida Slavic wrote:
>
>
>
>>Lorna,
>>I have only one suggestion for a vocabulary building technique
>>that may help.
>>
>>Analysing the suggested vocabulary I can see at least two main criteria
>>for the organization of concepts
>>      A) type by the function the resource
>>      has in education
>>
>>      B) type by form of presentation
>>
>>These are mutually exclusive categories of concepts and the educational
>>resources
>>are often a combination of those. Dublin Core Type is a good example of
>>a 'pure' vocabulary, it has only one criterion. In the case of educational
>>material I think at least two are needed. This means two facets of
>>mutually exclusive concepts otherwise the resulting vocabulary is
>>cross-classification ('apples and bicycles')
>>I think the following would facilitate discussion (please note
>>that this is only an example of vocabulary structure and not
>>suggestion of real vocabulary and I may have made mistakes)
>>
>>A FUNCTION
>>
>>A1 for practice / training
>>              exercise
>>                      written
>>                      practical
>>              experiments (?)
>>A2 for evaluation
>>              A21 test
>>                      A211 type of test 1
>>                      A212 type of test 2 etc.
>>              A22 questionnaire
>>              A23 written exercise(?)
>>              A24 interview
>>              A25 mixed
>>
>>A3 instructional material
>>              A31 lecture notes
>>              A32 slides
>>              A33 examples
>>              A34 mixed
>>
>>A4 combination of exercise/evaluation/lecture
>>
>>
>>B FORM OF PRESENTATION
>>      B1 simulation (?)
>>      B2 graphical presentation
>>                      B21 images
>>                              B211 still images
>>                              B212 moving images
>>                      B22 graphs
>>                              diagrams
>>                              charts
>>                      B23 drawings
>>                      B24 sketches etc.
>>      B3 audio recording
>>      B4 text
>>              B51 questionnaire
>>              B52 narrative etc.
>>      B5 objects
>>      B6 events
>>      B7 mixed forms
>>
>>This kind of presentation may help to see what kind of vocabulary may
>>be needed to work together ...for instance
>>one can have
>>A31 + B4 Lecture in the form of Video recording (moving images)
>>I think that provision needs to be made for expressing a combination.
>>
>>Hierarchical organization helps to expand or collapse specificity.
>>It enables easier mapping and vocabulary expansion - while a discussion is
>>going on.
>>It well shows gaps and 'holes'
>>If this is done properly the hierarchy can easily be collapsed and
>>the number of concepts reduced. The last stage is to decide what
>>terms/symbols/codes will be used to present concepts.
>>
>>I accept that you may have already gone through the phase above and that
>>the suggested vocabulary is the result but it would facilitate discussion
>>if it would be grouped by criterion.
>>
>>I hope I haven't missed the point here.
>>
>>Aida
>>
>>
>>

--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Paul Hollands <[log in to unmask]>
LTSN-01 Information and Web Support Officer
University of Newcastle, 16/17 Framlington Place
Newcastle upon Tyne, NE2 4AB
0191 222 5888
<http://www.ltsn-01.ac.uk/>