Subject: Re: Learning Objects
From: Clark Quinn (cquinn@knowledgeplanet.com)
Date: Wed 16 Feb 2000 - 18:43:18 MET
Date: Wed, 16 Feb 2000 09:43:18 -0800 From: Clark Quinn <cquinn@knowledgeplanet.com> Subject: Re: Learning Objects
List address to send message to everyone: ifets-discuss@LISTSERV.READADP.COM
Details of current discussion: http://ifets.ieee.org/discussions/discuss.html
---------------------------------------------------------------------------
>Clark wrote "For example, a learning object that discusses how autos behave
>differently with and without anti-lock brakes might be used in several
>different educational domains: the physics of friction, automotive design,
>or insurance liability." Yes, but it must be very difficult to construct a
>single learning object that would be appropriate for such different
>contexts.
Envision a simple chart of different speeds, and for each different speed the stopping distance on a car with and without ABS brakes. Then it's the context that makes it relevant to the different contexts; preceding with information about actuarial tables on different types of cars, or with formulas for relative friction coefficients, or with diagrams of antilock brake mechanisms. Or followed by.
>When I try to think about this whole issue, I see things very
>differently if the learning objects are passive (text or hypertext) or are
>real software components. If they are passive then an instructor can edit
>together different pieces - though it would still be a big job to adapt some
>text to any of these educational domains. If they are active then a
>sufficiently sophisticated object could adapt to different circumstances
>including educational domains, background of students, preferred learning
>styles and pedagogy, etc.
What I can't see is trying to have too smart an object. The requirements for the information to be passed and acted upon would demand much higher level agreement on standards than I think the other approach requires.
I see two models for an intelligently adaptive system: one is where the objects are relatively simple but richly tagged, and each system operates on them with a lot of it's own knowledge. Another where the objects are very smart, and the system has to pass them specific information, and each object has to adhere to a particularly stipulated set of input/output parameters. I guess I think the former is easier to achieve than the latter, as the individual system can add their own information about objects, but you can't have generic systems passing unique information to different objects.
>One worry I have is that software components in general have been widely
>promoted for all the reasons that Clark mentions (re-use, selection of "best
>of breed", etc.) and yet have had rather limited success. Instead of seeing
>a word processor as a collection of components for text entry, spell
>checking, formating, printing, etc the marketplace has gone in the opposite
>direction and most people buy office suites in which a word processor is
>tightly integrated with other programs (spreadsheets, presentation software,
>databases, etc.). Why are educational components different from the general
>case?
I think this is at the wrong level of discussion. The analogy is not to large components such as editors and paint programs, but to OO components with which to build such things: libraries for GUI widgets like scroll bars and sliders. For instance, OSs like MacOS and Windows have routines you call to do things rather than write it all yourself. It's harder to initially write good objects, but the payoff is high in long term productivity.
BTW, my recollection is that Microsoft bought a spell tool and integrated it, and the same with some MacOS functions. So it's hidden but there are component approaches.
Of course, I'm not convinced that the failure of Open Doc had to do with it's inherent benefits rather than marketing battles. But that's a separate issue.
On a similar irrelevant tack, there's there's a movement called 'design patterns' in software development that I think may have some relation to instructional design models (Merrill springs to mind) of saying that for certain types of learning, certain approaches make sense. I'm also mindful of the tools Schank's group builds that support certain learning tasks.
>My guess is that general software components have had only limitted success
>is that most uses are pretty standard. No one wants a word processor without
>a spell checker. So Microsoft and others make a very small number of
>standard assemblages of pieces and sell only the assemblages. And there are
>many advantages to the user to having tightly integrated collections of
>tools. Why is education so different? Is there a great need to compose
>custom courses?
Actually, in the training business, the answer is very much yes!
>Isn't the demand higher for tightly integrated materials
>than a loose collection of pieces? And if the marketplace for learning
>objects ends up only a place for large granularity (course size) learning
>objects, then what is the new idea here?
Well, I reject the large learning object picture, I've definitely argued for the smaller granularity. However, one idea I had was that a company that produces materials might produce objects about joining that a subsequent company that uses those materials might customize to do training on it's construction with those materials (this actually came from a talk at the Adaptive Learning Systems session at the NIST funding workshop). Say a company makes plastics that have to be cut this way, and bonded that way. They produce objects about this. Then a company that uses these plastics to make certain objects would customize those objects to produce training for their assemblers. Then that company might further customize those objects to provide home assembly instructions and troubleshooting/repair instructions. As the materials chain gets longer, the benefits accrue.
-- Clark
-- Clark Quinn KnowledgePlanet.com (510) 768-2408 cquinn@knowledgeplanet.com--------------------------------------------------------- Forum website: http://ifets.ieee.org/ Forum's contact person: kinshuk@massey.ac.nz Info on Join/Leave List: http://ifets.ieee.org/maillist.html ---------------------------------------------------------
This archive was generated by hypermail 2a24 : Thu 17 Feb 2000 - 01:27:52 MET