Hello all,
I've been lurking here for a couple of weeks since Sarah told me about the group. I thought I would follow up Sarah's posting with some further information.
My principal interest is in designing and applying controlled vocabularies to improving information discovery - specifically, improving navigation and search mechanisms via application of vocabularies to content and to information architecture components.
During the course of my most recent project I and my colleagues developed a range of vocabularies, tagged an enormous amount of educational content and developed a website that made some innovative use of these features. For anyone who may be interested in some of the background, I've included [1] some historical (strange to use a word like that about a project less than three years old, but that's the agile practitioner speaking I suppose!) insight into my current interest in information discovery as it pertains to educational resources.
Towards the end of the project, I realised that wider use of controlled vocabularies was held back by a number of factors, including:
o Lack of easily available high-quality vocabularies
o Lack of easy to use tools for creating and managing vocabularies
o The difficulty of putting all of the pieces together (OK, so you decide to tag your content. Even if you find good relevant vocabularies, even if you fiigure out how to use the existing tools to create your own, exactly how do you then tag your content to a high enough quality, and how do you then build systems that can discover that nicely-categorised content?).
So for the last year or so I have been working on a small project to address some of these problems. The result of this project is the Open Vocabularies Service and the Visual Vocabulary Tools. The DfE and the agency that ran the National Strategies project have generously allowed me, when the project came to an end at the end of March 2011, to put these into the public domain. I have now set up the basic OVS here:
http://openvocabs.org. This is a repository for storing and managing controlled vocabularies, complete with skos import and export mechanisms. The Visual Vocabulary Tools comprise:
o A very basic visual browser for controlled vocabularies. This provides a (I hope) visually attractive rendition of an information taxonomy as a hypertree, giving a nice impression of multiple levels of hierarchy in a single view. Browsing through the hierarchy is a matter of point and click on the term of interest.
o A fairly basic visual editor. Once logged in a user can choose a term from a vocabulary and edit or delete it. In the same way new terms can be added, whole branches can be pruned and grafted and new vocabularies can be created or imported (so far we have built import filters for skos and Excel).
o The beginnings of a browser-based tagging mechanism. We have built a Google Chrome extension that, once installed, appears in the toolbar on any editable content pages (in Drupal). Clicking on the toolbar loads the current list of tags applied to that content and allows the list to be edited. Naturally, this extension was built for a specific Drupal project and was therefore not a general-purpose tagging extension, though I believe more useful web-based tagging tools could be developed.
What I would like to do
Clearly, neither the OVS nor the VVT are the finished items. I believe that with some development effort they could vastly improved. I will shortly be releasing the VVT into open source in the hope that others will be sufficiently interested to pick them up and develop them further. I would like to seed such an open-source project by running a mini-project to enhance the current service and tools. If funding is potentially available through the OER call then I think this could be a catalyst for a wider collaborative effort to make the OVS and VVT into a free, freely-available repository, editing tools and tagging mechanism to help in the building of better information services.
There is a wider aim here. The eagle-eyed observer will have seen from a cursory glance at the OVS that the terms in the various vocabularies are identified by URIs. The design of the system ensures that every term across every vocabulary has a unique URI. It hasn't escaped my notice that this is a step along the way to information services with content tagged against a common set of vocabularies in a remote repository. The implications of this for meshing up new services based on aggregating information from a variety of sources are, I think, very interesting, but lie a long way beyond any current mini-project.
Anyway, that's a lot more than I intended to write when I sat down an hour ago! I am keen to explore the idea of some funding for an open project through the current mini-project call, and I hope to be able to put together an offering in advance of the deadline. Obviously I'd be interested to read any initial responses to the ideas above.
Regards,
Ian.
--
Ian Piper
Tellura Information Services - the web, document and information people
Registered in England and Wales: 5076715, VAT Number: 874 2060 29
Author of "Learn Xcode Tools for Mac OS X and iPhone Development", Apress, December 2009
01926 811574 | 07973 156616
--
[1] That historical background
As Sarah mentioned, we worked together on some very interesting information related projects for the National Strategies. I was particularly pleased to work on this project because it allowed me to do something that I had wanted to do for a long time. That was, to explore the extent to which it is possible to build information systems in which the very mechanisms of information discovery were centred on tagging of content against a family of controlled vocabularies.
In a nutshell, the design idea underlying the National Strategies site was to take a different approach from the original site design, which was a standard, traditional CMS-driven site. There was a fixed site structure and completely editorialised content. This brought all of the information discovery problems that you might expect: content was hard to find, content was often in unexpected places in the structure, and the editorial workflow resulted in content taking ages to appear on the site.
The new site was to be a complete departure. Rather than editorialised content in a fixed structure, the new approach was:
o All content is equal. The content repository has no internal structure, being simply a collection of content objects
o All content is tagged by a human expert according to their understanding of what the content is about (with no assessment of where it belongs)
o Content is tagged using a collection of discrete vocabularies, each one focused on a concept (such as role, curriculum subject, content type and so on)
o The site information architecture includes a hierarchical navigation structure in which the navigation points contain queries
o Each query (using the CQL query language) defines the profile of tags that content must have in order to surface at that point
This approach conferred considerable flexibility on the surfacing of information on the site. By modifying either the tagging of content or the query assigned to a navigation point it was
possible to tailor the content that appeared across the site. Because the whole process was dynamic, any changes required to the site structure were immediately reflected in the content that appeared.
That's probably all I need to say about the background for now. Happy to answer questions.
On 5 Apr 2011, at 11:41, Sarah Currier wrote:
Hello everyone,
Friday is looming: the deadline for sharing our proposals for the 3 JISC OER technical mini-projects. I've been wondering how others are approaching this. I'm facing this innovative open approach to bidding with some trepidation- for me, it's my livelihood, so it seems counter-intuitive to put ideas out there where the people I might be be competing with for funding can see them. A tension that echoes right across the open education domain I guess.
On the other hand, mini projects (1) and (2) are so dear to my heart that I can't but press forward! To me the intent behind them represents the culmination of work that many of us have been trying to do for some years now. Well, it's a culmination and, I hope, the start of a new, more productive level of research in this area. If *someone* does it and does it well, I'll be very happy even if I'm momentarily grumpy that it's not me!
I am talking with another independent consultant, Dr. Ian Piper at Tellura, about possible approaches to all 3 mini-projects. Ian and I have been working together for the past year (and Ian was around a lot longer than that) on a large-scale schools-level initiative developing 10s of 1000s of openly available resources for teachers.
We are both very keen on building on that project's work in open vocabularies; quality assurance for metadata; and content and metadata frameworks that are linked data- and Semantic Web-friendly.
Ian's history is (among other things) within the English schools and FE sector, while I've been involved for some years in HE educational metadata, as well as being involved in the Dublin Core Education Community. In DC-Ed we tried gathering use cases to see what folk were doing with educational metadata on the ground (not just OER use cases), but it was clear to me then that a larger-scale survey like mini-project (1) would be useful.
I know Ian has a great idea for the 3rd, open mini-project: to further develop his openvocabs tools: http://openvocabs.org/ - in order to ensure they meet the requirements of those developing and working with OERs. (Ian is at a meeting today so he agreed I could mention him in passing- he's on this list and will be able to answer questions himself). However, we're both sure that there will be a bunch of excellent ideas and tools coming forward for the 3rd mini-project.
For me, I am keen on the first 2 mini-projects, and Ian has the technical tools and expertise to help me with that side of those (I'm the semantic analysis person). On the other hand, there could be someone out there planning an excellent approach to the first 2 projects, and it would be more feasible for me to collaborate with them, if needed?
Anyway, I just thought it would be an idea to put this out there and see what comes back. Is everyone else preparing perfectly formed bids that they are going to post on this list on Friday? Is anyone looking for a collaborator or two? Or is everyone thinking "I hope someone else will bid for this because this work needs doing?".
In any case, I look forward to discussion on this over the next couple of weeks once proposals are in.
Best wishes,
Sarah
