Author: Adam

Kuali Student Services System

The kuali foundation has been around for a few years now but only recently (since July 2007) started work on a student services system backed by 7 N. American institutions to the tune of more than $25 million over a 5-year period, with Carnegie Mellon and MIT digging deepest. At an initial glance, this initiative looks like yet another well-funded open source initiative and the sceptics will not be expecting much long term impact, although it is worth noting that the kuali board does include Brad Wheeler, co-founder of the reasonably-successful sakai project.

The timeline gives an idea of the scope and indicates that “curriculum development” is an early priority, something that is elaborated upon in the outline functional description. This seems a logical starting point but is far from trivial, being the subject of several recent and active JISC projects such as COVARM, XCRI and COVa. How well the kuali team can deliver in this area should be an acid test of the viability of the initiative.

A glance at the Technical Architecture Principles is worthwhile, both for what it says as well as what it doesn’t. It is very committed to SOAP web services and service orientation. On this front, I think the student services system effort will be useful even if it fails miserably: a contributor and proving ground for the principles of the eFramework. On the what-it-doesn’t-say front, the list of open standards didn’t mention IMS or other educationally-oriented group outside the US; only PESC was mentioned.

One to watch!

BPEL – Opening Up Applications for Digital Repositories?

 At a recent workshop for the eFramework there  was a session discussing experiences in using “BPEL” (Business Process Execution Language) in the RepoMMan project. BPEL allowed the RepoMMan to use web services provided by the Fedora Repository to bring the workflow process out of the repository in a configurable manner. Significantly, the engine that runs the BPEL instructions can also access other web services.

This seems like an important development, demonstrating how a generic repository could be wrapped within a more application-specific workflow, so avoiding the need for an application-specific repository. This makes a good companion to the ability of a repository to support multiple metadata application profiles.  Assessment “item banks” could be a good test case for the utility of BPEL. For example, a recent report on item bank requirements for repositories identifies the policies around access and capture/storage/access of usage data. Projects such as Minibix are developing the state of the art in item banking and should provide the baseline for further exploration, but we should not expect the full service oriented vision, BPEL and all, to be upon us soon. But you never know…

Its Been 21 Years Since My First IM

The thought struck me last night, on seeing a film with some students around a green-screen monitor that it was in 1986 that I first used instant messaging. The command was called “notify” but the cognoscenti created an alias, “n”, in their logon script (personalisation) and another alias to see which of their friends were on line. I think it was an IBM 3084 mainframe behind the scenes and very “dumb” and impersonal terminals at the front end. It is sometimes hard to remember the days before the internet (it was probably 1992 when I first experienced TCP/IP and was jolly pleased to remote control my SUN workstation in Durham while in Boulder Colorado) but in 1986 we were pretty impressed by CUDN, the Cambridge University Data Network and being able to IM each other from our respective places of study and dwelling.

How often we “notified” I shudder to recall, for the monikers of my friends are still memorable: SJC10, BTS11, GAN10 …

Funny how some things change while other’s don’t…

Google Co-op

Google Co-op caught my attention recently as a welcome addition to the social bookmarking family, although it is not really being sold as this by Google; the collaborative/collective elements are not brought to the fore. I don’t wish to suggest that Google Co-op is the “answer”, but it has a complementary approach  built more around an interest than a person.  It’s well worth a look, there is a browser add-in to send URLs to your “custom search” and a facility to download your URLs (essential if you are going to invest much time). A little disappointing is that there hasn’t been any attempt to include any categorisation or tagging within a “custom search”.

Some Interesting Work in eAssessment

In Southampton today there was a get together jointly between some people from the HE Academy and JISC CETIS communities to look at developments and directions in eAssessment in the context of the physical sciences and mathematics. A couple of “quite interesting” pieces of work caught my attention…

Martin Greenhow showed us some examples of his new Mathletics (you need to scroll down to find this) and talked about some of the technology (MathML, SVG and an impressive JavaScript library) and challenges in setting maths and statistics objective questions with quite deep elements of randomisation and apparent smartness. It all looked rather well done and the questions work in QM Perception. This really looked like something deserving of much wider use and appreciation as well as an exploration of how some of the underlying technology could be tool-kit-ised and exploited in wider circles.

Aside from the technical perspective, Martin talked about the positive student response and the way the rich feedback had effectively turned what was ostensibly an assessment into a learning resource – mal-rules based on common student errors and misconceptions were linked to feedback in a particularly effective way. Students were keen to work through a surprising volume of feedback material. One of the problems Martin identified is students gaming the system, something that makes the question setting quite challenging, but maybe this student strategy, once understood, can be exploited too.

Frank Margrave gave us a preview demonstration of LinuxGym, which isn’t officially released yet. This is a nice example of assessment in context, in this case linking a live linux box with a set of questions so that students would execute the commands in response to a question and have the consequence of the commands checked against various rules for success.

REST-o-rant

I’m a bit of a sceptic when it comes to REST and the SOAP vs REST “debate”. There seems to be a tinge of hatred of corporate IT and dogmatic promotion of niche programming languages. There has been quite a lot of REST-o-ranting on the topic and I am not going to go there now.

REST does interest me in the influence it has on service design; indeed I have been a little worried that a lot of alleged SOA is not as loosely coupled as the philosophers would like and that, even when using document-style (rather than rpc style) SOAP, it is too easy to be led astray into, for example: OO thinking, CRUD. REST is both attractive in lending itself to loose coupling and unattractive in drawing one towards class-oriented design rather than business process oriented design so I thought I would have an attempt at a REST approach to IMS Enterprise Services v1. Scott Wilson had a look at a “REST-style” approach to IMS Enterprise Services some time ago but this isn’t in the same style that I look at, which I think is more fundamentalist resource-centric. I don’t think either of us is actually saying that real implementations of IMS Enterprise would never want to use the WS-* stack; it offers a heap of valuable heavy-weight capabilities.

My first cut is available in PDF: IMS Enterprise v1 in Fundamentalist REST form

Have the Social Bookmarking Sites Got it All Wrong?

There certainly are a lot of social bookmarking sites around at the moment with varying levels of social networking associated. The idea of social bookmarking is splendid and, providing the spammers can be kept at bay, is sure to be here to stay but here are a number of things that make me feel that there remains a lot of scope for evolution. Getting over the bad joke of the way “tagging” is implemented is one but in other ways the whole model seems quite old-hat, almost like the days when web “search” sites were based on human-maintained data.

The model that makes me sigh involves the transfer of my bookmark into a silo. The silos compete for market share and in the mean time there simply isn’t a critical mass of users with more than a few, often semi-geekey or web-trendy, interests in any one silo. This model probably fits quite nicely with the Yahoo business model with its groups, dating, portalesque approach so it is no surprise that they bought del.icio.us. It does seem at odds with the web as a network of connected (linked, related) resources.

FOAF does fit much better with this view of the web and it is appealing to think that there will be an inexorable draw towards distributed and free networks etc when it comes to social bookmarking. The basic idea of FOAF is that your “friend of a friend” file lives where you choose (anywhere a web page can live) and it can be read and links followed to the FOAF files of your friends – i.e. “crawled” or “spidered” (the FOAF geeks use “scuttered” ) just as Google indexes conventional web pages. FOAF does offer some experimental ways to include bookmarks but this capability is not well exercised as yet. Maybe it isn’t right to imply that the processing should be “just as … conventional web pages” for two reasons: the point of using something like FOAF is that a little bit of extra meaningful information can be harvested; the relationships or links between people and bookmarks adds to the equation.

A related initiative is Annotea, coming from the opposite direction to FOAF: resource annotation and bookmarking. For my money, Annotea is just a bit too complicated to get the buy in from the mass-market and maybe FOAF is a little, too. Given a little bit of support in user tools, though, it shouldn’t be too hard to move to a more open and distributed model where bookmarks and their relationships to people are just part of “the web” and where Google or a more agile sucessor can neatly interweave these with their existing crawl/index/query capabilities and sweep away the silos much as many internet “search” sites were swept away several years ago. In the mean time it looks like Google may just buy out a silo to keep up with the Jones’.

Group Size and Satisfaction

I have often observed chaotic meetings or been frustrated by group discussions that seem to lose the plot. I had never really thought about why this happens other than to complain under my breath about a weak chairperson or facilitator or to mutter “stick to the point” now and then. Now, at least, I have found a partial explanation: Christopher Allen has an interesting article, “The Dunbar Number as a Limit to Group Sizes” where he provides some evidence and relates to more fundamental anthopological research.

Work, Expression, Manifestation, Item

It is often easy to miss something that is obvious when pointed out. When identifying resources it is easy to be a little sloppy and fail to discriminate between 4 concepts that have been called “work”, “expression”, “manifestation” and “item” in the rather interestingly-acronymed FRBR. This isn’t an emoto-tronic friend but Functional Requirements for Bibligraphic Records (may be found at http://www.ifla.org/VII/s13/frbr/frbr.pdf).

A work is the intangible product of intellectual or creative endeavour, whereas an item is a tangible thing that I could own. For example (borrowed from FRBR): J. S. Bachs Six suites for unaccompanied cello is a work, the performances by Janos Starker recorded in 1963 and 1965 are an expression and the recordings released on 33 1/3 rpm sound discs in 1965 by Mercury are a manifestation.

If software systems are built to correctly handle these different cases then quite a number of user benefits can be realised. For example, by allowing the location of available resources that are of the same work when all I know is the identity of a manifestation. The manifestations of an expression could differ in the medium, for example microfilm vs print for conventional publications. It is at this level that some accessibility issues could be dealt with.

For an example of where FRBR has inspired an e-system, take a look at ePrints Application Profile.

Argumentative Technology

One of the essential features of learning is argument. There are many ways to unpack this rather bald statement but two takes on “argument” are particularly appealing to me and have a natural expression in educational technologies without being at all futuristic to implement.

Firstly, I want to consider the kind of rigorous exploration of ideas through argument that Socrates was so well known for. The Socratic method may be an over-grand term to use but I think it is possible to use educational technology in some simple ways that engage the learner in the kind of reflective thinking that Socrates forced his subjects into. I particularly like a style of dialogue where the Socrates-character leads their subject into an exploration of the logical consequences of a misconception. Clearly to attempt to programme a computer to be Socrates is a bit far fetched but I wonder how much effort it would take for someone who knew their students to use something like Quandry to take a prepared decision-tree for such a dialogue and create something that would require learners to engage with the ideas. I gather something like this could even be expressed using IMS QTI.

I talked about doing this kind of thing with Andrew Ravenscroft at a conference some years ago and he introduced me to some software hehad been designing for computer aided argumentation, which illustrates my second “take” on argument in this post. I recently stumbled across some work by Andrew and colleagues in a JISC case study to illustrate innovative e-learning practice using what they have called AcademicTalk. AcademicTalk is not at all like Quandry … read the case study to find out more.

There is, maybe, an interesting Russian Doll of argument within argument. What would it be like to place something like AcademicTalk, with synchronous and asynchronous discussion, within the maze-room-like decision points in a Quandry-style “canned” Socratic dialogue? Record the discussions and the creator of the maze has a natural feed of new misconceptions to challenge.

This is all soundling like a multi-user dungeon.