The Definition Debate

A CONFERENCE ROOM FULL OF PEOPLE

R: Welcome everyone, and thank you for coming.

O: Happy to come!

V: Thank you for inviting us, R.

R: Should we start by saying why we're here? (Everyone nods in agreement). Well, I study instructional applications of technology for a living. There are a handful of key issues around learning objects that I'm interested in exploring, and I thought getting several people together for a food and conversation would be a wonderful - and enjoyable - way to gather some first-hand data about what's happening in the world of learning objects.

D: Speaking of food...

R: C agreed to cover food for our inaugural meeting. I think you chose Thai food, C?

C: Right. It should be here any minute.

R: Ok. Thanks. We may do food again or we may not. We'll see how it goes tonight. So... introductions. (Looking at S)

S: Me next? Ok. Well, I have my head buried in Learning Technology standards almost all day, every day - literally. I guess we'll talk about it later, but lately it feels like I spend more time thinking about XML bindings than learning, and I wanted to see what everyone else is doing.

D: I know what you mean! I work as a programmer for EvilLMS and spend half my time trying to understand the standards and make our software conform. In my spare time I build systems I think are (frankly) a little more innovative, but even when I ignore the standards there are still a thousand questions about how to put these things together so that somebody will actually learn something.

O: I'll do my best to avoid disparaging EvilLMS during our conversations. Of course I was super disappointed when the state system chose EvilLMS instead of Sakai...

R: (Interrupting) Let's try to stick to the topic. Why are you here, O?

O: Well, learning objects are digital, aren't they? That means that, once they've been authored, we can copy and distribute them at basically no cost.

V: (Interrupting) You can't just ignore the costs of the authoring process, though. How are you going to handle those costs?

R: (Interrupting, again) You're in a minute, V. (Back to O) You were saying, O?

O: Well, when you can copy and distribute educational materials at basically no cost, that starts to sound like a formula for universal education, doesn't it? I guess that's why I'm interested in learning objects, though, not why I'm here. I guess I'm here just to keep up with the work everybody else is doing.

V: Sorry, O. (To everyone) Learning objects are a key part of our e-learning platform. I'm responsible for our learning object strategy and the directions our products take. Frankly, I'm just curiuos to see if we're behind, up to speed, or a little ahead of everyone else.

R: How about you, C?

C: Honestly? My Product Manager told me to come. (Everyone laughs) I'm supposed to be finding out what everyone else is doing, what the latest techniques are, etc.

R: And what about you, T?

T: My principal sent me. I'm supposed to be the expert at our school on learning objects, but frankly, I don't have time to be searching for objects that I can use in my classroom or assign to students outside of a classroom environment. Heck, I barely have time to get tomorrow's lesson done, so I don't see what all the hullabuloo is about learning objects.

I: Well T, I'm coming from a very different perspective, but I still think I can help you. Within your classroom there are at least two different ways you can be using learning objects. First, you have a standard curriculum, correct?

T: Yes, we have to follow the National Science Education Standards.

I: Well then, if you and your fellow teachers were willing to organize a little, you could record the learning objects that worked for each lesson plan. It would require a little more work up front, but would reduce your workload in subsequent years. Also, if you have students that are either behind or ahead of the rest of the class, learning objects could be used to help bring them up to speed or give them additional resources, without draining you anymore than you are obviously drained.

R: Thank you for addressing T's concerns. I'm sure we'll be able to help address them more over the course of our conversations. In the meantime, would you like to introduce yourself I?

I: Sure, I'm a graduate student who thinks that learning objects will be a primary component of the automated, individualized instructional system of the future. I know that's a mouthful, but basically, I believe that we can identify patterns in learner progress that will facilitate providing optimised instruction for each individual learner. At the moment, I am studying the semantic webbecause I think that it will play a key role in such a system.

R: Well, I'm sure we'll all get to know each other better over the coming weeks. (Shuffles through several pages of notes) Should we open the first topic?

S: Can we really proceed without a standard definition of "learning object" to make sure we're all on the same page?

R: Actually, that's our first topic. I don't know that we can hope to reach consensus on a definition, but we should all at least talk a little about what we think a learning object is.

S: There's already consensus! According to the IEEE Learning Technology Standards Committee, which creates internationally recognized standards for key pieces of the learning technology puzzle, learning objects are defines as:

"'Any entity, digital or non-digital, which can be used, re-used or referenced during technology supported learning'"

D: And that's about the most worthless definition ever produced by any body of people. How long did that take to come up with?

S: Have you ever tried to negotiate consensus between representatives of dozens of countries and large corporations, some of whom don't even speak English as a first language?

D: No, which would be why I've never been associated with a definition that stupid, unless you count some of the DTD's I've had to produce for the EvilLMS... (Everyone laughs but S, who looks scandalized)

R: (To D) Ok, now. Let's try to keep it slightly more civil. (To everyone) The LTSC definition *has* been broadly criticized. What is problematic about the definition?

V: Well, for starters it doesn't rule out anything. You may as well have said a learning object is a noun.

C: But that would rule out all the other parts of speech... (Mixed laughter)

O and C: (Singing) I find it quite interesting...

R: (Over the singing) Any contributions other than zingy one liners?

O: As I said above, I think the definition should be constrained to things that are digital. What good does it do to call Mr. Rogers a learning object? Or the Civil War? Or Amsterdam? Or a xylophone?

S: While it may not be apparent at first, there are several good reasons. IEEE LTSC WG 12, the Learning Object Metadata Working Group, are focused on how to index, catalog, and otherwise enable the discovery of things. Ever want to find your teacher in the system? Or another tutor? They're very important to helping you learn, and both can be described in metadata to make them easier to find.

R: Metadata, of course, being "card catalog" information like title, author, publication date...

S: Or office number, or phone number.

C: (Looking somewhat confused) But when my manager tells me to develop reusable learning objects, surely he doesn't mean he wants me developing people, places, and things?

I: But what about using something like RDF and URIs (unique resource identifiers) to create a digital identity for those people, places, and things? (Cut off by an excited D.)

D: Hey! Check out http://www.uwm.edu/Dept/CIE/AOP/LO_what.html.

R: Not all of us have wireless smart phones, D. What's it say?

D: It has a bunch of definitions. One by Wiley - "any digital resource that can be reused to support learning" - which NLII or ELI or whatever its called these days seems to have adopted. There's also a longer one by the Wisconsin Online Resource Center:
 * Learning objects are a new way of thinking about learning content. Traditionally, content comes in a several hour chunk. Learning objects are much smaller units of learning, typically ranging from 2 minutes to 15 minutes.
 * Are self-contained – each learning object can be taken independently
 * Are reusable – a single learning object may be used in multiple contexts for multiple purposes
 * Can be aggregated – learning objects can be grouped into larger collections of content, including traditional course structures
 * Are tagged with metadata – every learning object has descriptive information allowing it to be easily found by a search

C: They define learning objects is terms of minutes?

D: (Re-reading) Looks like it.

R: I'm actually hoping we can save the size conversation for another time, if that's ok.

T: When you have 50 minutes to teach a class, time is very important information to consider, so can we talk about it now.

C: (Mumbles) We define it by how many screens of information a person sees...

V: Is it possible that people's definitions differ because they *should* differ? I mean, is it possible that defining what a learning object "is" is a strategic decision that each organization has to make for itself?

S: It's possible, but it's going to throw interoperability out the window.

O: But it depends on what kind of interoperability you want, right? For me, the primary interoperability rule is "will it run in a browser window?"

S: (Interrupting) Same as SCORM...

O: (Continuing) "On any platform without any plugins (except maybe Flash or Quicktime)?"

V: For example, Cisco decided on a heavily templated approach with Courses being aggregates of Lessons, and Lessons being aggregates of Topics, etc.

R: So V, you're suggesting that the answer may be context specific?

D: (Cutting in) Unless you want the object to be able to talk to anything, like the LMS. How is the management system supposed to record your score on a test if the learning object containing the test doesn't know how to talk to the LMS?

R: I think the vast difference in opinion among this group is proof enough that the answer *is* context specific. D and S want standard interfaces other systems can talk to, O wants learning objects that run in a browser, V wants objects that make strategic sense for his organization...

D: (Cutting in again) And C wants learning objects with five screens of information, plus or minus 2. (Chuckles)

R: I think we will have to accept, for the purpose of our on going conversations anyway, that we each value different things in learning objects and, therefore, we each define them differently. Having said that, with apologies to S, it seems like the group is converging on being *digital* as an important characteristic of "learning object-ness." Unless anyone disagrees...? (Looks around the room)

S: Apologies aside, I still think its a mistake to limit the scope of our discussion to things that are digital.

I: Like I was mentioning earlier, with something like RDF and a URI, something that isn't digital can have a digital presence. That way, when T is teaching a lesson on Roentgen's discovery of X-rays, there can be a digital fill-in for Roentgen.

O: Well it's like R said, isn't it? Your definition of learning object is heavily influenced by what you value. Take me for example. My primary interest in these things is the reuse that comes from something being digital.

V: Ours as well. Because when the resource is digital, we practically eliminate warehousing, manufacturing, and delivery costs.

D: Hold on - I'm confused. If you can store them on a hard drive, I see the warehousing costs being basically zero. And when users can download them over the net, you pay for an internet connection instead of trucks, drivers, insurance, etc. But how does their being digital affect your manufacturing costs? You still have to develop them.

V: Yes, but we only have to develop them once. If learning objects are digital, we can create an infinite number of copies at no cost. Remember, with a physical product like a car or treadmill, every time you want to sell one you have to pull raw materials out of inventory and assemble them into a new copy of your product. It costs the company something everytime you sell another widget. That's the beauty of being in the software business. Why do you think Microsoft is the richest company on the planet?

T: (Under breath) And save school budgets the expense of purchasing frogs when we can dissect virtual frogs. Not to mention the stress of animal rights activists...

D: (Almost thoughtfully) Huh... I've never done anything but software development... I never thought about the production process that way.

O: And if your goal is to give learning objects away to people in developing countries, then it really helps to have almost nonexistent warehousing, production, and delivery costs. If you can cover the cost of producing them the first time through grants, or recover the cost through sales in developed areas, you're good to go.

V: (Stares in amazement at O)

O: Yes, even the zealots of openness think about business model and sustainability issues.

I: Digital is also important for future developments. For example, while most of us are probably thinking about learning objects being used in a typical desktop or laptop system, we should also think about other options, things like pdas, podcasts, and even immersive simulations.

R: Ok. So "digital" will be part of what we mean when we talk about learning objects. Another charateristic of learning object-ness that causes frequent debate is whether or not the object needs to be explicitly instructional.

C: (In disbelief) You mean there are people who don't think learning objects need to help people learn?

R: (Enthusiastically) Oh, good. Here's the food. (The conversation momentarily stops as every one selects a small styrofoam conatiner from a mountain of very fragrant Thai food and breaks apart wooden chopsticks) Not that they don't need to help people learn, just that they don't have to be instructional.

C: So, what you're getting at is whether or not they have a specific instructional objective?

V: Or whether or not they provide opportunities for practice, or whether or not they give feedback, etc. Why would we call them "learning" objects if they weren't designed to help people learn?

T: I'm using a Wikipedia article to discuss the effect the levees on sedimentation. Are you saying that this article isn't a learning object? It seems to fit Wiley's definition, it is a digital resource and I'm planning on reusing it in each class to support learning. If it isn't a learning object, what is it?

D: Is an article on Wikipedia, CNN or slashdot a learning object, then? In high school and college I had loads of teachers who used newspaper articles and other reference material in their teaching. Are you saying, V, that these types of resources shouldn't be called learning objects? Even when they're digital and available for reuse in online instruction?

V: By themselves, no. Though they'd probably be a substantial piece in a larger aggregation.

S: (Slightly flustered) Do standards have no place in this conversation??

C: What do the standards say on the issue, S? Does a learning object have to be designed specifically to support learning? Does it have to present information, give opportunities for practice with feedback, etc.?

R: Before S answers, I want to call a time out. We're making good progress on getting to consensus on a definition. We've heard pretty convincing arguments that learning objects should be digital. There's been no objection to suggestions that they be reusable. Where we're headed next, architectual or structural issues of how objects are designed, leads very nicely into the topic of our next conversation - what is the right size for a learning object?

S: (More flustered) How can you say we have consensus when I'm in complete disagreement on the issue of learning objects being digital?

R: Because we have consensus on the identity of a key issue that must be considered by anyone who wants to use learning objects - do you limit them to things digital, or not? I don't think the six of us can answer all the important questions in these sessions, but if we can identify several of them, and agree that they really are important issues that deserve considering, then I'm going to call that consensus. And a useful consensus at that.

C: So our goal isn't to put together a definitive set of answers to the important questions?

R: I don't think it can be. We already see a significant difference of opinion on the very first issue we've addressed. I think the best we can do is discuss each issue thoroughly and put together a summary of the questions for others to use in making up their own minds, based on their own contexts and situations, what learning objects are and how they're going to use them.

D: (Almost to self) Such an academic approach...

C: No, wait. This would be really useful. So when my manager comes over to my cube and says "we're doing learning objects," this will actually give me a list of questions he and whoever else is in charge of strategy have to answer before I design for 3 weeks only to have him tell me that I did things wrong...

R: Yes, I imagine our list being used like that. And I think people will find it very useful. So questions from today, all of which we will revisit in more detail before our meetings are finished:

(R writes on the whiteboard)


 * 1) Should only things that are digital be counted as learning objects, or should a variety of other things be included, like people, places, and events?
 * 2) How important is the potential reusability of material in determining whether or not some content should be called a learning object?
 * 3) Does content need to be explicitly instructional in order to be called a learning object?

Did I miss any?

S: I assume we'll come back to the topic of standards?

R: Yes, in just a few weeks.

O: Sounds good to me.

D: Yea! This has been pretty interesting.

R: Ok. Let's get together again next week then. See you then!