Hi Amber,
There are three issues here:
1. What attributes are you storing in your database?
2. What attributes do your users see?
3. What attributes appear in your data when you are interoperating with
another provider who uses the LOM?
Each of these questions can have a different answer so long as you can
map your stored attributes to the corresponding LOM attributes when
interoperating.
I suggest that what goes into your database (or whatever your metadata
is stored in) are the LOM pairs. Your software should then map these to
a more
user friendly and descriptive set of pairs when the information is
displayed on your website. If your software does not allow for this then
you will have to decided from one of three choices:
1. Your users have to live with funny looking and unfriendly descriptors
2. You put *your* descriptors into the records in the database but map
them back to the LOM ones for interoperation
3. You get more flexible software.
Bear in mind that in theory LOM compliance (or for that matter
compliance with any other standard) is a mutable thing. It is only
necessary when somebody wants to see your records as LOM XML. This means
that *IN THEORY* you can have whatever data structure you like and put
whatever descriptors you want in your records *PROVIDED THAT* you can
map back to LOM conformant attributes and structures when somebody wants
LOM records from you. The enormously tricky bit is ensuring you can map
directly. To do this you really need semantic equivalence of terms and
equivalence in the granularity and richness of descriptions.
Since in this instance there is a one to one mapping between your user
friendly terms and the LOM terms you should be OK.
Hope all that made sense.
Cheers.
Amber Thomas wrote:
> Fellow Metadata Users,
>
> I manage the Ferl website, and we are using IEEE LOM compliant metadata
>
> _http://ferl.becta.org.uk/_
>
> There’s a CETIS case study on our use of metadata at:
>
> _http://metadata.cetis.ac.uk/usage_survey/cs_ferl.pdf_
>
>
> I need your help on our use of relations
>
>
>
> *Relations Pairs*
>
> We use the relations pairs (references/referenced by replaces/is
> replaced by etc) to relate content objects within the site. For example,
>
> we link a case study, lesson plan and weblink together
> _http://ferl.becta.org.uk/display.cfm?resID=2251_
>
> or we link an event item to agendas and booking forms
> _http://ferl.becta.org.uk/display.cfm?resID=5421_
>
> We present them to the user in the left hand column of the site. (We
> are re-launching in mid November, and then the relations will be
> displayed in a better way)
>
> Relations are a useful tool, but they can be confusing to users. We
> need to use plain English to display the relation type. We also need
> to make sure we use the pairings consistently
>
> We’ve already changed “has part” to “contains”. We’re considering
> making more changes. There’s a strong case for changing the reference
> pair to references/references because in the way we use it, it’s a
> sibling relationship rather than parent/child (if you see what I mean!)
>
> When our xml implementation is complete, each metadata record will be
> translated back into the standard terminology and made available as an
> xml rtf.
>
>
> */However/* in changing the term to make more sense in the context of
> our website, we risk deviating from the standard uses of the terms.
> That’s why I’m emailing this list.
>
>
> Can you see any dangers in us changing to the proposed terms below [in
> brackets]?
>
> Can you suggest any better terms?
>
>
> Grateful if any metadata people out there could spend 5 minutes
> feeding back to me/the list by* 3**^rd ** November.*
>
>
> Hope this makes sense. Thanks in advance
>
> Amber
>
>
>
>
>
> Terms in [brackets] are our proposed replacement terms for use on our site
>
>
> *Pai**r*
>
> Is version of Has version
>
> See whether this is used more than replacement pair. Remove one from
> the menu
>
> Is replaced by Replaces
>
> See whether this is used more than version pair. Remove one from the menu
>
> Is required by Requires
>
> Standard function: provide filetype support?
>
> For Ferl the cms file type field produces file type support for each
> resource, eg zip, pdf. So we don’t need this pair to indicate
> plug-ins. Ensure that XML records translate this info into relations?
> Take requires pair off the menu
>
> Is part of [belongs to] Has part [contains]
>
> This is our most useful pair for managing packages of resources. We
> already replace “has part” with “contains”. Lets also replace “is part
> of” with “belongs to”
>
> Is referenced by [references] References
>
> We don’t use this for parent-child relations, so can we display both
> as “references”? Ensure that XML records still translate as “is
> referenced by”
>
> Is format of [alternative format] Has format [alternative format]
>
> Standard function: signposting templates?
>
> We can use this for resources we host in more than one format: eg we
> might have it in word and in pdf or in html and in word. Use
> “alternative format”
>
> Is based on Is basis for [is used by]
>
> We need this pair for instances of repurposed and re-submitted
> content. “Is basis for” is not an intuitive label. Suggest “is used by”
>
>
>
> Amber Thomas
>
> Ferl Content Editor
>
> Becta
>
> Millburn Hill Road, Science Park, Coventry, CV4 7JJ
>
> tel: 024 76 847167
>
> email: [log in to unmask]
>
> website: http://ferl.becta.org.uk/
>
> Ferl Five: the fifth Annual Conference for e-learning policy and
> practice in Post-16 Education
>
> 24^th and 25^th November. London
>
> Can you afford to miss it?
>
> *http://ferl.becta.org.uk/conference2003*
>
> |
>
> **********************************************************************
> This email and any files transmitted with it are confidential and
> intended solely for the use of the individual or entity to whom they
> are addressed. If you have received this email in error please notify
> the system manager.
> This footnote also confirms that this email message has been swept by
> MIMEsweeper for the presence of computer viruses.
> www.mimesweeper.com
> **********************************************************************
> |
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
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/>
|
