|
|
|
|
 |
Why is there no clear leader among CMS / LMS / VLEs?Created by Catherine Howell (La Trobe University) on August 8, 2006
OSS Watch, the UK's non-partisan advice service on open-source software and open standards, have just published a survey report last week about software use in UK higher education. The survey, carried out in February-March 2006, reports on the take-up of open-source software (OSS) in UK Higher Education (HE) and Further Education (FE) institutions. By default, it also gives a picture of the NON-takeup of open-source solutions. The survey is based on data collected from nearly 1 in 5 UK institutions (18%), making it a useful information baseline from which to extract trends and make future predictions. There's a lot of good reading in the report, but the thing that interests me most is that it shows that there is currently no clear leader amongst Content Management Systems (CMS) in UK HE. Why is this? Stuart and Paul will probably have opinions about this :-) What follows is my 2p worth. (NB: Stuart responded before I could upload the rest of my post -- however I think we agree on many points). One argument for diversity in e-learning systems and services for higher education is that instutions are themselves diverse, and have specific needs – particularly in relation to pedagogy. But is this really true? Are the needs of HEIs actually that diverse? If we look at the range of software in use within HE, we see that within particular domains, a handful of commercial providers dominates the landscape: such as Oracle’s PeopleSoft within administrative services; or Microsoft’s Windows Explorer for staff/student web access. Indeed, the OSS survey report shows that Microsoft Office and Internet Explorer are deployed by all institutions on most desktops. And this software is not necessarily restricted to administrative functions: for example, web browsers and email applications are commonly used within teaching and research practice, too. Given this perspective, the diversity of LMS software in use starts to look less like a result of the actual needs of HEIs, and more like a function of the “early stage” marketplace – with commercial vendors offering a range of competing “enterprise solutions”. Where LMS and their ilk are concerned, there is no killer app. Yet. From the above discussion, that would seem to point to the need for an approach to software design that attempts to create “permeable” software; by which I mean, flexible software, that is adaptable by users to their needs. Of course, it’s perfectly possible that we just don’t yet know what a killer LMS would look like. Are our LMS too structured; too hierarchical; too, well, “enterprisey”? That may be, in part, a legacy of the tendency to use LMS to manage administration-related tasks. More importantly, it may also be that, at present, we don’t really know how to reflect pedagogical aims and needs accurately within educational software design. Significantly, the OSS survey shows that Moodle is now in use as a VLE within 56% of UK Further Education institutions. Moodle’s much-praised flexibility and ease of use have always weighed in its favour within the FE sector– HE developers, please take note! Most LMS are designed either to be “pedagogically neutral” (hence, designers / vendors hope that their product will reach a maximum of customers) or else to conform to a specific and/or fashionable pedagogical perspective (e.g. Moodle is designed to reflect principles of constructivist learning). We can argue over which of these approaches is the correct one, but the fundamental issue to my mind is that pedagogy is, by its nature, difficult to define, and thus is difficult to capture. Pedagogy is culturally embedded within institutions; it is “situated action”; for even the most reflective practitioners (including students, as well as staff) it functions tacitly, as well as explicitly. Our current methods for requirements analysis and systems specifications require precise articulation. Inevitably, they are a form of translation, which is another way of saying, a form of approximation. “Stuff” always gets lost in the process, and that “stuff” may well be crucial. We need a system for the specification of e-learning systems that can work in a more intuitive, or more tacit, way – or, more radically, that could enable users to specify their own system.
|
 |
|
 |
 |
||
| Unless otherwise noted, EDUCAUSE holds the copyright on all materials published by the association, whether in print or electronic form. In certain cases the work remains the intellectual property of the individual author(s) (see Special Circumstances). Content from conference speeches, presentations, blogs, wikis and feeds reflect the opinions of the author, and not necessarily those of EDUCAUSE or its members. | |||
I'll rise to this challenge...
I, personally, am of the opinion that there are a still a number ofvery different VLEs (Virtual Learning Environments) / LMSs (LearningManagement Systems)/ CMSs (Content Management Systems) becausepeople want them to do different things.
In the open source world you can contrast Moodle and SAKAI. Moodle reflects a specificpedagogy (socialconstructivism) and is incredibly learning-focused. It is designedto be installed and configured with a single click. SAKAI, on theother hand, is more process-driven and assumes that you have a team ofweb designers, programmers and support people to install and configureit.
As for CMSs,these are more more broadly applicable again, with an even vaguerremit. Like most software, any CMS can be adapted as a VLE or LMS, butthe degree of difficulty and the degree of success will vary.
If you're thinking of adopting an VLE, LMS or CMSs, the first thingto do is to think about your needs, requirements andresources. The second thing to do is talk to your peers, because theyare the people whose needs, requirements and resources are closest toyour own.
As it happens, OSS Watchhave a very useful guide to selectingsoftware it's written for open source software, but the principalsare the same, no matter what the licence is.
Hi Stuart, I agree that institutions may want to do different things with their software, although I think that once a sufficiently flexible system is developed, it will be greeted with wider (and potentially, cross-sector) adoption. Perhaps another way of describing the difference between Moodle and Sakai is to say that they have different approaches to the problem of "sufficient flexibility". Moodle tries to achieve flexibility by addressing the needs of learners; Sakai tries to achieve flexibility via its component-based architecture. In terms of usability, I think Moodle wins hands down; but then I think Sakai is better at supporting distributed research communities :-) I do think that developers need to find a way to involve users more in the process of systems specification. At the moment, users tend to get consulted at only two stages of the design process: at the initial needs analysis stage ("tell us how you work") and at the evaluation / end user stage (often restricted to bug-fixing, or at best, "show us how you use it"). I worry that users get left out of the formal specifications bit, which to my mind is the crucial part.
I was thinking about what I guess is something like this a few days ago in the context of learning objects. I think there certainly seem to be self-stories about the un-productizability (sort of related to enterpriseyness) of HE and also the uniqueness of indiviudal institutions. I suppose that they're necessary stories to give out because it justifies the existence of lots of organisations.I don't mean that they're not true, I think they probably are to a reasonable extent, if they were you'd need to tell those facts anyway, to justify yourself to people with power over you, but the same would be the case if they weren't true! So I'm not sure that there's much to be gained from analysing things that would be asserted whether or not their apparent meaning were true or not, if that makes any sense, :)?I guess this comment's a bit tangential, but I guess I mean that the existence of a killer app could potentially be seen as a threat in itself?
Dan, I guess my take on this is that institutions tell many different stories about themselves and about eachother. Sometimes these stories emphasise the similarities between HEIs, sometimes the differences. I suppose there is a sense in which a "killer VLE" might be perceived as a threat to institutional diversity, precisely because VLEs are part of promotional discourses about institutional specificity -- and internal ones, too. But surely institutions would have to identify very strongly indeed with their chosen VLE for that to be the case. Interestingly, Moodle's success is often attributed to the fact that its users and institutions seem able to "own" it and identify with it in a way that more "enterprisey", corporate-ish systems don't allow... so you have the case of collective, yet somehow diverse, identification with a single product.
"You can tell which floor a person works on by the way they dress." I recently overheard this remark at a staff canteen of another department - one buiding, same department, but different modes of dress. There's a strong cultural element and I think the same applies in various ways here; it cannot be easily discounted.
Killer apps tend to do one thing very well, they're usually niche, not generalist. According to Wikipedia you've already cited some of the most well known examples - web browser and office applications. [Btw, is there no alternative to this term? 'Must have app'? sqn (sine qua non) app? alas probably not; have tried lifesaver app, but that means something slightly different)]. Furthermore, there could be a case made for the predominance of certain packages in particular domains being due to reasons other than common needs and also for this situation breaking up in future.
In contrast, Content Management Systems (to emphasise its distinction from Course Management Systems) are exceedingly broad in their requirements and cross multiple domains - in the 'content' data; in the relationships between the data; and in the functionality needed from the system. (I don't think they should be equated with VLE or LMS).
So I'm really not surprised by the plethora of solutions adopted. Even if you took a certain class of CMS there may be considerable variations within an institution. When investigating solutions for a teaching resource management system for the Theology Faculty at Oxford we (HCDT/ACDT ) considered a very sophisticated system developed for a Science Faculty, but it's data model was fundamentally unable to satisfy the essential many-many relationships between lecture courses and [examination] papers. It provided a lot of other functionality that was not needed or rendered inappropriately, so overall it was unsuitable. Eventually a simpler system was found that did support the required relationships and could be adapted. It used a completely different set of technologies.
Getting the right level of academic input is quite an art. From my experience as a developer academics are seldom available, especially over a sustained period, unless time is formally allocated for them to work on such projects. Once they start getting more involved on an ongoing basis, their expectations may increase along with their enthusiasm and wonder why saying, "This looks quite good, but we also need this," is not met with instant nods of agreement - it can be hard to avoid 'feature creep'. Even so, I think we should indeed try more for situated requirements specification, if that's the right phrase, but how can this be accommodated in terms of resources...?
Regarding systems themselves, there are many other issues (and cultures) to consider. There are institutional cultures - some smaller Universities may feel they have to centralise all operations and run with standard operating systems and applications across the board, simply because of cost. On the other hand, other Universities with greater resources are naturally more heterogeneous. Even IT departments and teams have their own cultures - within one team, PHP and MySQL is a natural combination for Web applications, whereas at others they are far happier with CGI/Perl or Python plus PostgreSQL.
One challenge of increasing flexibility is maintaining coherence - one might argue that flexibility would mean choosing a large selection of small tools that focus on particular functionality, but how do these work as a whole and not become fragmented? It's a challenge faced by developers of Personalised Learning Environments - some have tried to take a system-oriented approach, pooling and aggregating data from multiple sources, sending out to other systems, all within one application. However, the demonstrator systems that I've seen really lack much in the way of pedagogical structures.
But maybe Sakoodledora will come along and deliver what we all need .. :-)