Long post; relevant to people looking at extending the LOM and tools
(Reload?) for creating extended metadata records based on application
profiles, and touches on accessibility metadata.
Hello Jay, everyone
[Jay and I have been talking off-line, so not all of this reply arises
from issues in jay's query below]
As Jay says, there are two ways of providing metadata information not in
the LOM model; there are two ways of doing one of those.
1. Have a two seperate records, one for LOM data and one for the stuff
not in the LOM. I know that this is the approach taken for QTI metadata,
where usage data (information on the statistical analysis for how many
people get the question right) is stored as a seperate section of XML
outwith the LOM stuff. Similarly, in Intralibrary (correct me if I'm
wrong) the IPR information is captured in an ODRL XML instance is
included along side the LOM record.
This approach seems right to me when the information is not related to
anything else in the LOM and/or there is an existing well developed
schema for encoding it.
2. Extend the LOM by
2.a) using a classification or controlled vocabulary that expresses what
you want. This approach is taken in the UK LOM Core in adapting the
vocabulary used for educational context, and by the RLLOMAP in
developing vocabularies to classify resources by policy theme and
pedagogy. -- no need to change the schema definition to do this.
2.b) adding elements to existing categories in the LOM, as seen, for
example, with the Educational.LearningPrinciples element in the
Celebrate application profile.
I believe the best practice advise for creating new elements is to
describe them in a schema seperate to the LOM xsd and to reference them
in using name space prefixes in the LOM XML instance (that way you avoid
having to reference local copies of schemas)... but it's a while since I
looked at this, so please let me know if there's any better way.
[Perhaps folk who use accessibility metadata have experience here, do
you use put Access for Metadata in Lom instances as extended elements,
or use it as a stand-alone additional instance outside the <lom>..</lom>
tags]
I know that the extensions that Jay needs to make will have to be taken
on a case-by-case basis. For example, he needs to encode information on
the regional accent of the speaker, which could be done using the
language encoding in the form en-GB-cockney, but this may be
constraining information intended for humans by using an encoding
intended for machines, so perhaps a classification with a non-standard
purpose would be more appropriate. Jay also needs to encode information
about subtitles, so again perhaps there's something to be learned from
the accessibility metadata work. There are also some others (something
like instructions for teachers and learners) that might be similar to
new elements created for other projects -- Jay, if you share your
"problem elements" there may be someone on list who has dealt with
something similar.
Another concern for Jay is how to create metadata instances and content
packages once he has built on and bent the LOM to his own needs: in
other words a content packaging tool that can cope with application
profiles. I'm sure Reload can do this in theory, but it would require
some Java and XML know-how to do the necessary modifications. Is anyone
aware of anything better? I.e more easily configurable (like SHAME for
RDF metadata?) and more focussed on the end user (like WCKER?).
Hope this helps, Phil.
References / resources
QTI meta-data and usage data
http://www.imsglobal.org/question/qti_v2p0/imsqti_mdudv2p0.html
UK LOM Core http://www.cetis.ac.uk/profiles/uklomcore
RLLOMAP http://www.rdn.ac.uk/publications/rdn-ltsn/ap/
CELEBRATE
http://www.eun.org/ww/en/pub/celebrate_help/application_profile.htm
LOM XML Binding
http://ltsc.ieee.org/wg12/files/IEEE_1484_12_03_d8_submitted.pdf
Accessibility metadata, e.g. IMS AccessForAll Meta-data
http://www.imsglobal.org/accessibility/index.html
Reload http://www.reload.ac.uk/
SHAME http://kmr.nada.kth.se/shame/
WCKER http://wcker.conted.ox.ac.uk/cgi-bin/trac.cgi
eLanguages technical wrote:
>
>
> Hi, I am rather new to world this world of metadata and VLO’s, and am
> currently looking at producing a metadata profile for a project I have
> been assigned to.
>
> I have undertaken the initial research into the various standards and
> frameworks surrounding Metadata. I have some questions about creating a
> customised profile.
>
> Where ever possible I have mapped the current ‘meta-data’ collected by
> an existing repository (this was designed a few years ago, and was not
> as far as I am aware designed around a standard – rather to address a
> need) but I have data that I need to tag resources with that does not
> ‘fit’ an existing category.
>
> I am using the IEEE P1484.12/D6.4 LOM as a template.
>
>
>
> My guess is that you add the data in a custom element in the
> group/subgroup deemed most appropriate (for example, ‘Description for
> teacher – e.g. main task, skill strategy’ and ‘Description for learner –
> e.g. main learning outcome, skill’ would fit within the 5.x group of
> Education. Bear in mind this is a separate set of descriptions to the
> both the 1.4 General Description and the 5.10 Educational Description
> categories…)
>
>
>
> I guess another strategy would be to created two profiles, a totally LOM
> based meta-set and a totally ‘custom’ meta-set, and apply both sets of
> meta to the LO’s/resources.
>
>
>
> Can anyone suggest a _/correct/_ or better way of doing this?
>
>
>
> Thanks for your time
>
>
>
> Jay Gattuso
>
>
>
> eLanguages Technical
>
> University of Southampton
>
> New College
>
> The Avenue
>
> Southampton, SO17 1BG
>
> UK
>
> _____________________
>
> (W) + 44 (0)23 8059 7510
>
> (M) + 44 (0) 7733 322554
>
> http://www.elanguages.ac.uk
>
>
>
--
Phil Barker Learning Technology Adviser
ICBL, School of Mathematical and Computer Sciences
Mountbatten Building, Heriot-Watt University,
Edinburgh, EH14 4AS
Tel: 0131 451 3278 Fax: 0131 451 3327
Web: http://www.icbl.hw.ac.uk/~philb/
|
