« March 2007 | Main | May 2007 »

April 30, 2007

ProLearn Project virtual seminars on metadata harmonization

On a note related to the work of the DCMI/IEEE LTSC Task Force which I mentioned in the previous post, the EC-funded ProLearn project is currently running a series of "virtual seminars", facilitated by Erik Duval and Mikael Nilsson, on the topic of "Learning Objects, Metadata and Interoperability (LOMI)" and more particularly on the harmonization of metadata models, with a focus on the IEEE Learning Object Metadata standard and Dublin Core.

The project plans to draft a document outlining some of the issues in this area, and the seminars themselves are quite wide-ranging discussions.

The sessions are held on Flashmeeting, and are open for anyone interested to join. The next session is scheduled for this Thursday 3 May 2007 at 13.00 BST/14.00 CET. Recordings and chatlogs from the previous sessions are also available.

JISC CETIS Metadata SIG, Manchester

A bit slow off the mark again here, but I spent the Monday before last at a meeting of the JISC CETIS Metadata & Digital Repositories Special Interest Group in Manchester. The focus of the meeting was Dublin Core, and I gave a presentation on the recently circulated version of the DCMI Abstract Model (slides).

I was slightly apprehensive as to how the presentation would go down, in part because I was first up (on a Monday morning...) and I was plunging straight into a topic which almost by definition is somewhat, well, "abstract", and in part because it was my first attempt to present this version of the DCAM. As it turned out, I think it was OK. I concluded with a couple of slides where I attempted to summarise how I think we ("the DC community") need to reframe how we present "what Dublin Core is (and isn't)". I suggested that the focus is moving away from the notion of DC as a (more or less fixed) "set of terms", and particularly away from that of DC as "15 elements", and towards an emphasis on the idea that the DCAM provides a framework for the construction of sets of resource descriptions of arbitrary complexity which can reference whatever metadata terms are required to meet the requirements of the particular application; those sets of descriptions can be represented/encoded using a range of digital formats; and the approach is compatible with the approaches underpinning the Semantic Web. I think that view is broadly consistent with the perspectives given by Tom Baker in his opening presentations to the last two Dublin Core conferences. 

I found it very helpful that directly after my presentation, Julie Allinson from UKOLN described (slides) the work on the ePrints DC Application Profile for describing scholarly works. I think Julie succeeded in making rather more concrete some of the issues that I had addressed in a rather general way, by illustrating very clearly how they were applied to a specific problem (and it meant that I could keep saying, "That point might be a bit clearer after you've heard from Julie"!) Julie's presentation was well-received, I think, and a couple of people came to chat to us later about the possibilities of applying this profile (or a similar profile based on a FRBR-derived model) to other types of documents for which similar requirements had been identified.

I also gave a (very!) brief report (slides) on the work of the DCMI/IEEE LTSC Task Force, which is seeking to facilitate interoperability between systems working with the IEEE Learning Object Metadata standard and systems working with Dublin Core by developing a mapping of the IEEE LOM standard to a set of terms that can be used in DC metadata. Essentially the initiative is developing a DC application profile which supports the representation of the same information as is expressed in a LOM instance, and providing a mapping from the LOM standard to that profile. There hasn't been much activity from the Task Force recently, in large part because the work had reached a point where some decisions were dependent on decisions about the domains and ranges of DCMI properties (see also Andy's comments) and on updates to the DCAM itself. But I think we are close to a point where the answers to many of those questions are rather clearer and we can start to pick up the LOM-DC work in the Task Force.

After lunch, Rosemary Russell from UKOLN introduced the work of the UK DCMI Affiliate. The DCMI Affiliate programme creates national or regional agencies which have a mediating role between DCMI and DC implementers in the countries or regions they represent. Affiliates act rather like two-way channels, disseminating up-to-date information from DCMI to the "local constituencies" (and maybe supplementing or contextualising that information - for example, through the provision of translations of key documents) and feeding "local" responses and requirements back from those constituencies to DCMI (Affiliates have representation on the DCMI Board of Trustees). Although the UK DCMI Affiliate is funded jointly by JISC and MLA, with UKOLN acting as "managing agent", the intention is that it should represent the broader communities of DC implementers in the UK. I think Rosemary was hoping for some feedback on what sort of activity might be useful, but during the meeting we struggled to come up with very concrete suggestions! A couple of possibilities which occurred to me later (the former perhaps more concrete and realisable, the latter perhaps more tentative!):

  • The development of a set of modular "tutorial"-style resources for DC implementers, based firmly on the DCAM and presented in a consistent "style" (consistent use of terminology, but also reuse of graphics etc), such that the components could be used in various combinations to meet the requirements of different audiences. So, something like:

    • (?) short introduction to the entity-relational model
    • (?) short introduction to Web Architecture (URIs, resources and representations etc)
    • short introduction to the DCAM (notions of making statements about resources etc)
    • DCAM in detail
    • DCAM and RDF
    • the vocabulary model and how to create/describe vocabularies of metadata terms
    • the DC application profile model and how to create/describe DC application profiles
    • encoding DC metadata using X/HTML (and processing X/HTML-encoded DC metadata?)
    • encoding DC metadata using the DC-XML format(s) (and processing DC-XML-encoded metadata?)
    • the DCMI metadata vocabularies
    • Case study 1: "Simple DC" as a DC application profile
    • Case study 2: the ePrints DC application profile
    • Case study 3: some other DCAP? a service provider (metadata consumer) view?

    While DCMI does provide some materials via its own Web site, these materials have been developed by several different people and/or over a long period of time, and they vary at least in the terminology and presentational style they use if not also in some of the concepts described - though I should note that DCMI  is aware of these issues and has itself recently tendered for some work in this area. I think the first two of those items on my list have probably been under-emphasised in the past, but it seems to me that at least some grasp of them is helpful to (arguably, necessary for) an understanding of the others - certainly for the developers of DC application profiles.

  • The development of a regular programme of small f2f meetings within the UK, rather along the lines of the JISC CETIS SIGs. I appreciate the resource implications here are rather more substantial - and Rosemary may not thank me for suggesting it! And I'm also conscious that it's important not to draw energy away from the work of the DCMI Communities and Task Groups. I guess this suggestion is partly motivated by my admiration for the way CETIS have succeeded in building up active and supportive communities around areas of activity related to e-learning and e-learning-related standards/specifications, and the way the SIG meetings (or the ones I've attended at least) seem to work so well. They provide fairly low-key, informal forums in which people come together and exchange ideas and they really do seem to foster a spirit of collaboration, and generally avoid the problem of "meetings for meetings' sake". Is there scope to do something similar around the DC implementer community in the UK? Or is that community too disparate/fragmented? Maybe the fact that Dublin Core is longer established than most of the e-learning specs means that the context is different and it's harder to build up those sort of "interest groups" at this point in its evolution? I'm really not sure.

The final section of the day focussed on areas of DC-related activity closer to the CETIS community's e-learning-related interests, with a presentation by Sarah Currier of Intrallect on the work of the DCMI Education Community, of which Sarah is a co-moderator, and by Rob Tice of the Vocabulary Management Group on the Vocabulary Bank and Vocabulary Studio tools that have been developed for Becta to support the authoring and management of the controlled curriculum vocabularies used in UK education. Those presentations led into some discussions of the role of "metadata registries" of one form or another in the context of the Web, which has prompted me to think about resuscitating a draft post I was working on a while ago... So more on that topic coming up shortly, I hope.

Anyway, thanks to Phil Barker and Neil Fegen for organising an enjoyable day.

April 25, 2007

DC Collections Profile Reviewed

Andy noted a few posts back that the DCMI Usage Board met recently in Barcelona. One of the items on the UB agenda was to review an updated version of the Dublin Core Collections Application Profile prepared by the Collection Description Application Profile Task Group, and I'm pleased to note that - as Ann Apps reports here - the UB approved the profile as "conforming" to their current criteria for a DC application profile (DCAP): essentially, it's compatible with the DCMI Abstract Model (DCAM); it's internally consistent; and the documentation conforms to current guidelines.

Thanks to Sarah Shreeves of the University of Illinois Library at Urbana-Champaign and Muriel Foulonneau of the Centre national de la recherche scientifique for their work in co-ordinating the Task Group, steering the group towards conclusions, editing the (quite substantial) set of documents, and getting things in order for presentation to the UB - all on quite a tight timescale. Back in my previous life, I contributed to work done in the area of "collection-level description" at UKOLN, building particularly on the work done by Andy and by Michael Heaney of the University of Oxford on the RSLP Collection Description Model and Schema. As part of that work, for a while I chaired the DCMI Collection Description Working Group (as the group working on the profile was called back then), so I've continued to take an active interest in the development of the profile and it's good to hear the news that a stable version is now available for use.

The DC Collections AP specifies how to construct a DC metadata description set that provides one or more collection-level descriptions, i.e. descriptions where the resource described is the collection, rather than an individual item within that collection. In his paper describing the entity-relational model on which the simpler model underpinning the DC Collections AP is based, Mike Heaney used a geographical metaphor (which I admit I used to quote an awful lot when I was doing presentations on the topic!), which I think captures quite nicely the sort of "view" of resources that a collection-level description can provide:

The information landscape can be seen as a contour map in which there are mountains, hillocks, valleys, plains and plateaux. A large general collection of information – say a research library – can be seen as a plateau, raised above the surrounding plain. A specialized collection of particular importance is like a sharp peak. Upon a plateau there might be undulations representing strengths and weaknesses.

The scholar surveying this landscape is looking for the high points. A high point represents an area where the potential for gleaning desired information by visiting that spot (physically or by remote means) is greater than that of other areas. To continue the analogy, the scholar is concerned at the initial survey to identify areas rather than specific features – to identify rainforest rather than to retrieve an analysis of the canopy fauna of the Amazon basin. This model attempts to characterise that initial part of the process of information retrieval.

The work within DCMI on the collections DCAP has been something of a protracted process, I readily admit! But it's worth noting that the contributors have faced the challenges not only of grappling with some occasionally complex issues related to what we wanted to say about collections and their relationships to resources of other types, but also of adjusting to what has been something of an "evolving" understanding within DCMI of "what a DC application profile really is" - a topic about which I jotted some thoughts a few weeks ago - and about how a DCAP should be presented and documented. However, I'd like to think that the outcome is that the resulting profile is now based on firm foundations within the framework provided by the DCAM.

April 20, 2007

TeacherTube

I came across TeacherTube the other day - a teacher-oriented variant of YouTube.

An interesting idea, though one wonders about critical mass and so on.  It seems to me that there's a useful question to be asked about when it makes sense to build a vertical application like this (by which I mean, an application targeted at a particular audience) and when it makes more sense to use the tagging facilities in existing social tools to partition various sub-sets of the content.

For example, one presumably wouldn't ever invent Flickr4learners??  Rather, one would use a suitable set of tags to highlight images particularly intended to be used as learning resources (i.e. tags that indicate curriculum area, educational level or whatever).

April 18, 2007

Slideshare gets real

A while back I blogged about Slideshare being an example on 'fake' sharing...

I'm pleased to say that it has got 'real' and now offers the ability to download the PPT or PDF file for each presentation, as well as making the slides available thru the embedded display facility.  Nice.

Note that you have to manually enable this feature for any existing presentations in Slideshare - but doing so doesn't mean uploading the presentation again, just selecting a new tick box on the 'edit' page.

A tag for the Symposium

Following the convention of event organisers recommending a tag which can be applied to resources within the various systems that support tagging to flag resources related in some way to the event, I suggest that we use the tag

efsym2007

for resources related to the upcoming Eduserv Foundation Symposium on virtual worlds and learning.

I've tagged a few items on del.icio.us related either to the general theme, the event itself, or to the speakers, and I'll tag this post so that it's picked up by Technorati

del.icio.us del.icio.us icon Technorati Technorati icon Flickr Flickr icon

And picking up on my previous post, I've created a del.icio.us tag description for that tag, but it's only visible on my personal collection of entries for the tag.

Hmm. So, back on the topic of tag descriptions, I wonder whether it would be a good thing if del.icio.us surfaced links on the all-user/tag list (e.g. http://del.icio.us/tag/efsym2007) to individual user/tag lists (e.g. http://del.icio.us/PeteJ/efsym2007) for those cases where users provide tag descriptions. e.g. a column over on the right-hand side of the page headed something like "Descriptions of this tag are provided by...." (and similar "see also" links in the corresponding RSS feed too?). I can't make up my mind whether it would be the right thing to do or not: on the one hand, there's value in the approach that tag descriptions are user-specific, "local"; on the other, as I suggested some while ago, some element of "negotiation" of tag "meaning" is part of what puts the "social" in social tagging, and it seems to me facilitating access to other taggers' descriptions of the same tag would contribute to that negotiation process.

April 17, 2007

Describing your del.icio.us tags

I'm a bit slow on the uptake with this item, but I noticed a few days ago that my favourite social bookmarking service (del.icio.us) has introduced a feature called "tag descriptions", and I just started to play around with it a bit this morning. This feature allows you to add some descriptive information (title and free-text description) of a tag that you use i.e. to create some metadata "about" the tag and your intended "meaning" for the tag.

For example, I apply a tag "REST" in del.icio.us entries for resources related in some way to the Representational State Transfer architectural style. So to indicate that my tag is related to that particular meaning of "rest" rather than the "having forty winks" notion, I created a "tag description" for my "REST" tag which provides the expansion as a title, and a brief textual description. (I included a couple of URIs of "defining" resources too, but that's probably overkill given that I've already created and tagged entries for those same resources!). That description is displayed at the head of the page which lists my del.icio.us entries using that particular tag:

http://del.icio.us/PeteJ/REST

and it's also included in the "channel" metadata for the corresponding RSS feed:

http://del.icio.us/rss/PeteJ/REST

The feature offers a useful mechanism for individual taggers to articulate and to disclose to other users the intent behind their use of a tag, at least in human-readable form.

Pulling together the different descriptions provided by different taggers for the same tag (which I'm not sure del.ici.ous allows me to do, at the moment anyway) might be an interesting exercise in highlighting where a single tag is applied by different taggers with contrasting "meanings". As far as I can see from a cursory scan of the first few pages, the vast majority of del.icio.us entries tagged with "REST" are indeed related to Representational State Transfer. I guess some of the other tags I apply ("Turtle", "TAG") are rather more likely to be applied with quite different intent by other users, and "tag descriptions" provide one way of highlighting such divergences/collisions.

 

When persistence has a sell-by date

I note that Nicole Harris at JISC has started the JISC Access Management Team Blog... good stuff and a welcome addition to the UK HE blog landscape.  In her posting entitled The Accountability Question she notes that The Rules of the UK Federation (section 6.4.2) state that:

where unique persistent Attributes (e.g. eduPersonTargetedID or eduPersonPrincipalName) are associated with an End User, the End User Organisation must ensure that these Attribute values are not re-issued to another End User for at least 24 months;

I remember reading this guidance during the comment period on the various policy documents that came out at the start of the UK Federation - it struck me then as rather odd.  Any sentence that starts with 'unique peristent' and ends with 'not re-issued ... for at least 24 months' has got to ring alarm bells somewhere hasn't it?

Why 24 months?  Less than the period for which most students are at university!  The problem, or so it seems to me, is that any service provider that wants to make use of these attributes can't rely on them being persistent even for as long as the student is typically at university.  As a result, service providers will presumably have to find some other way of guaranteeing that the person they are dealing with today is the same of the person they were dealing with yesterday, at least for any unique persistent attribute that is nearing its second birthday :-(

I'm tempted to ask why any time limit is suggested?  Why not simply say that these attributes must never be re-used?  Presumably some institutions have problems ensuring that they do not re-use their local usernames and so on.  But so what?!  Generate a truly unique persistent handle for the user in some way (a UUID or something) and associate it with the local username thru some kind of look-up table.

That way you can easily guarantee that these identifiers will never be re-used.  Am I missing something obvious here?

April 14, 2007

Revised DCAM and DC-RDF available

Before the intervention of the Easter holiday weekend and my return to a pile of Eduserv Research Grant proposals to review, I had intended to post a quick note to say that DCMI has made new versions of the DCMI Abstract Model and Expressing Dublin Core Metadata using RDF available for public comment. The main changes in this version of the DCAM reflect the requirement which emerged from discussion of the previous version to distinguish unambiguously between literal and non-literal values. Those changes have added some complexity to the description set model, but I think the consequence is that the mapping to RDF graphs described in the second document is now much clearer. Anyway, comments welcome - to the dc-architecture mailing list rather than here though, please.

My next job on that front is to try to remember where we had got to six months ago and draft a new version of the DC-XML format (or formats) reflecting the updates to the DCAM. A few recent(-ish) items (sort of) in this space:

I'm off to Manchester on Monday to talk about the DCAM to a meeting of the JISC CETIS Metadata and Digital Repositories Special Interest Group. I'll put some slides on Slideshare, err, when I finish them....

About

Search

Loading
eFoundations is powered by TypePad