 |
 |
A few responses to Pierre-Julien's email below, but first an additional comment from him on the ISO MLR work: Work on MLR is done the ISO way with the help of National Bodies contributions at plenary sessions at the drafting level, then by balloting later on. This is actually the case as both MLR Part 1 Framework and Part 2 Core Elements will go through ballot resolution meetings, trying to address all comments to the satisfaction of everyone (not always an easy task!) at our September meeting in Toronto. Then, new drafts will be prepared and balloted again. We are not addressing DC interoperability at the moment but are planning to provide guidance in our future work. This is why there is no specific homepage on this project. However, I wrote a short summary of the project here: http://ntic.org/page.php3?t=patrimoine_educ_en&id_article=269 Sarah Currier wrote: > > I notice a similarity between the fundamental work on DC abstraction > and the > extensibility based approach we have taken for MLR to accommodate new > stakeholders' needs. It is necessary to explicitly and unambiguously > state > principles, construction rules and data structure. In my view, such a > work > would not result in constraining possible applications but leave them > opened > to accommodate applicaiton based on community needs (and at the same time > avoiding the debate about what a learning object really is!). Yes, this is a good point: I fear we may have been less than welcoming to comments regarding this issue: so I will emphasise again that the kind of thing Andy has noted is *very* important, and should definitely be part of what the DC-Ed AP does: I just think some of us were worried that people would back off from discussing the bits they were directly interested in, namely the usefulness of individual properties and the like, if the discussion got too far into the modelling angle. Both angles need work. > > Following are some comments about proposed educational elements: > ---- > 1) in introducing LOM Educational.LearningResourceType, you may want to > consider the ambiguity in its vocabulary which may address both the > nature > of the learning resource and its pedagogical application. In the MLR, we > proposed a split between two elements: > > Resource Type : Collection, Dataset, Interactive resource, Moving image, > Physical object, Service, Software, Sound, Still image, Text, Website > > Pedagogical Type : Learning measurement, Problem solving Activity, Tool, > Display, Description, Explanation We have already started to address this by having two DC properties in the DC-Ed AP: Type and InstructionalMethod (the latter is essentially pedagogical type). For each of these we will be recommending vocabularies. Type is a general DC property which we will be recommending education-specific vocabularies for. I've already noted in the draft AP that the LOM vocabulary is ambiguous and has elements that may apply to either property. I think we hope to recommend other vocabularies. In my experience the LOM Learning.Resource.Type vocabulary has long been one of the most problematic parts of the LOM for implementers and users! Anyway, it looks like your recommendation will map neatly to the DC-Ed AP. > > 2) having very limited FRBR related experience, I found it very > subjective > to document item such as InteractivityLevel, SemanticDensity and > Educational. Difficulty and seriously wonder about their relevance and > interoperability. Yes, these are the *other* LOM elements that are problematic for LOM implementers and users and in my experience are rarely used. In fact I don't think I know of any application profiles of the LOM that do use them (within individual implementations I mean). If they ever are useful it will likely be within a particular repository or community which has a shared understanding of their meaning, but even then the only one I can *really* see teachers or other users of LOs finding useful is Difficulty- and even that is a stretch due to the specificity to particular user level. > > 3) Educational.TypicalAgeRange. This is probably one of the most > important > federated search criteria. From an interoperability perspective, at > the time > of data sharing, education level should be translated into numerical age > values that can be processed without ambiguities (MLR has two elements: > minimumAge and maximumAge). But, of course, this doesn't need to be DC > embedded. I'm not sure I agree with you here! We had a long and fairly contentious discussion about this a while back on the CETIS Metadata & Digital Repositories SIG discussion list- I'll try to find it. I tend to come down on the side of "age is not useful for representing educational level": you can have gifted students at age 10 studying 1st year university-level materials, you can have adults learning to read for the first time, in fact, age has very little to do with what level a university resource is pitched at: 1st year engineering could have students of any age from 15 to 75 studying there. Adults in their thirties can return to college or secondary school to get basic qualifications, the examples go on and on. Trying to represent educational levels as age ranges is problematic on a number of levels. Where age range comes into it is in situations such as adults learning to read needing different reading materials than 5-year-olds, or materials on sensitive topics such as sex education pitched at different (childhood) age ranges. I think we need to accept that educational level cannot be truly mapped in an entirely technological process across international contexts, because of the differences in education systems, unless there is specific work and agreement between implementers. We've had a go in the UK just at mapping across the 4 nations within the UK (each has its own educational system) and this has been moderately successful, but really highlighted how difficult it is. There are some kinds of metadata that need human intervention at some point. > > 4) In our editorial work, Norm Friesen and myself have carefully reviewed > actual LOM definitions for both elements and vocabulary values. You > may want > to give a look at those definitions for DC candidates. The resulting > documents are publicly available. > > 36N1524 Text of ISO/IEC CD2 19788-1, ITLET - Metadata for Learning > Resources > - Part 1. Framework > > 36N1312 Working Draft (WD2) for ISO/IEC 19788-2 - Metadata for Learning > Resources - Part 2. Data Elements > > http://isotc.iso.org/livelink/livelink?func=ll&objId=1056984&objAction=brows > > e&sort=name > and > > WG4_N0185_Proposed_other_parts_for_ISO_IEC_19788 (PDF Package) > > http://isotc.iso.org/livelink/livelink?func=ll&objId=5791404&objAction=brows > > e&sort=name I will certainly look through those- and hope to raise any discussion with you arising from them. I'm pleased to have heard from you in plenty of time for the DC meetings later this month! > Hope this can be useful, Absolutely, thanks again, Sarah -- Sarah Currier Co-Moderator, Dublin Core Education Community Product Manager, Intrallect Ltd. http://www.intrallect.com 2nd Floor, Regent House Blackness Road Linlithgow EH49 7HU United Kingdom Tel: +44 870 234 3933 Mob: +44 (0)7980855801 E-mail: [log in to unmask] --