Is Uniglu what I need?

A warning upfront: I think this post may expose my ignorance, and there may be a 'duh, we've been working on that for ages', type reaction. But based on previous experience, I've found that my ignorance is often shared by others, I'll forge ahead.

I'll give you the background: On Friday I had a quick play with video in Googlemail. Naturally it works fine, so I put out a rather facetious tweet about why do we bother to design software specially for education when this stuff is just there. Niall responded saying:

"can it be linked to student registration systems for automated population of tutor groups etc? "

This is a good point, and one that people like Scott and co have kicked around a few times.

The IMS Tools Interoperability Spec seeks to do some of this, stating:

"The IMS Tools Interoperability (TI) approach addresses the growing demand for a reusable mechanism for integrating third-party tools with core LMS platforms."

And Chuck Severance pointed me at something he is playing with called Simple Learning Tools Interoperability.

I may be wrong, but these don't quite seem to be what I'm after. What I want is the tools equivalent of eduglu, which for the sake of not bothering to come up with an original name, we'll call Uniglu. Here's my user scenario, so you can tell me if these will do what I want:

Tom is a lecturer at Strummer University in French Language. He wants his students to get used to engaging in online audio/video discussions, capturing mobile input. He has distance as well as campus based students so wants something asynchronous. Seesmic seems to be ideal. But rather than tell his students to sign up there and swap ids, he just wants them to be able to use it as a suite of tools. If they decide to go there, it will be already configured for them.
So he speaks to his IT services dept. who tell him that Seesmic is one of the 2138 applications that are listed in the Uniglu directory. This means they are trusted and are capable of accepting university data. They pass the secure data to the Seesmic API which automatically creates accounts for all the students, using their OpenID and preferred user names. It adds in all the students in the cohort as Followers and Following.
Similarly, in the course wiki, Tom is using Wetpaint. This also adds in all the students, and also sets up Tom and the two course assistants as moderators. The access to the wiki is set to registered only (ie the course cohort only, not public) which is the default option when it is available.

It seems to me that we aren't far off doing this, and that a combination of existing data standards would get us some way there, for instance OpenID, OAuth and OpenSocial?

I accept there would be issues around data protection and privacy, but one could imagine universities being declared trusted data owners and that applications have to be vetted to be in the Uniglu directory. The difference here with the IMS spec is that we are sending data out to these apps, not requiring them to come into our systems. I'm not sure if that's significant or not. The standard is probably something that is in the domain of IT Services, not individual educators, but it would need to be easy enough to implement for application providers if want the rich tapestry of apps to choose from, and of course, from the user's perspective it must be seamless.

So my questions are:

1. Does this make sense?
2. Is it desirable?
3. Does it exist already?
4. If not, could we do it easily?

And here I am asking the question in Seesmic:

UnigluWhat I want is Uniglu>

Comments

[AJ Cann]

Decho: http://www.readwriteweb.com/archives/decho_to_offer_api_access_to_a.php

[John Connell]

Niall's 'duh' comment is precisely the rationale I have been using to justify the development of Glow in Scotland for the past half dozen years or so to anyone willing to listen to me. As things stand, tools such as GoogleDocs and the myriad other free Web 2.0 tools around offer amazing functionality to individuals or to small groups. But they do not scale in the sense that Niall indicates.

A national development such as Glow, entirely web-based, integrated, single-sign-on - and even though it includes what might be seen as a Web 1.0 VLE (although this would be wrong given the VLE's complete integration into the broader web environment) - requires at its heart a national system of authentication. With this, Glow users have both a level of inbuilt security AND the ability to cut the total user-base in a million different ways to create communities of interest. Such 'Glow Groups' (to given them their localised name within the environment) offer untold levels of flexibility to users. Anyone within the system (so long as they have the requisite permissions - and that is too often the rub for some of the more authoritarian local authorities in Scotland in the way they manage Glow) can create any Glow Group they wish: a group of 2, 20, 200, 2000 or 20,000!

And once created, the group can then use the full panoply of applications within Glow to interact, learn together, share resources, share ideas, etc - all without the necessity of central control or intervention.

Unfortunately, in Scotland, there are some third-rate academics (http://www.tes.co.uk/article.aspx?storycode=6005181) who simply cannot see past their own prejudices to recognise the potential for distributed learning using an environment such as Glow.

[Patrick]

Having read the post I was going to make some points about us having been here before while working on the JISC Joinin project (creating tutor groups cohorts using Learning Design, LAMS and Moodle). If you remember, Martin, this was the one where "the line Patrick adds which makes it really difficult" was that we would use IMS Enterprise to pass the user information around. It was all a lot more difficult than it should have been, and then when we had done it no-one wanted that bit of the work anyway.

So this is something that there are quite a few pretend solutions to, and perhaps some real solutions that are too complex in practice. I don't think though that it is just about authorisation - in which case OpenID starts to sound good enough. It is about identity. OpenU does operate an openid server and I have used it but it actually doesn't give me very much and I am not sure of its status as a sort of secret way to be open. What I would value more is somewhere that can really deal with more of the identity issues that can be passed around and grouped rather than just the authorisation side.

Should not be impossible and I think a light enough version could be put together - but then be swamped by extra features.

Anyway - I was going to write something like this - but can't now that I watched the Seesmic video and know that Barbie talks just like a certain Professor of Educational Technology! That is just too disturbing.

[Richard Treves]

I agree something like Uniglu would be a good thing, there are ever more wonderful apps we should have access to for our students.

However, there is an important downside when you consider the typical landscape of the UK University. You, me and your blog readers are power users and champions of IT, we all hunt out new tools and experiment with them on our own and are capable in training themselves in their use. However below that in the IT strata there is the rump of averagely skilled tutors and a whole lot of laggard tutors who want as little to do with the VLE as possible.

Even if you could take away all the technical work involved in supporting a new app (as Uninglu would) you still have a significant cost in terms of training and supporting the new functionality with the average and laggard Tutors. You also have to consider the cost in time to the student in learning to use the new tool. If its only going to be you using it in your course they will have mastered the tool in course time, a precious resource for a skill they won't use again.

Which is not to say I'm a luddite, just that the true cost to the institution and its students need to be considered.