Internal And External Context
A CONFERENCE ROOM FULL OF PEOPLE
R: And here we are, yet again. And everyone's even on time tonight!
S: And apparently there are going to be some visuals tonight? (Gestures to a computer and projector on the conference room table)
R: We've been talking about context and how important context is for a while now, but we have yet to really talk about context specifically.
O: Oh, good! I've been looking forward to this for a while now.
V: We've said that more context can make instruction more effective. We've also said that more context can get in the way of reusability. Just what is context?
R: I have some ideas on the topic, of course, but I don't want this to turn into a lecture.
D: There's little of danger of that happening. (General laughter) So go on.
R: Well, let's start with this: context is what gives things meaning.
O: I think I'm going to need some context around that definition - you know, something to give it some meaning?
R: I'll begin by saying that there are much more formal definitions than the one we're going to develop tonight. I skip those because I'm more interested in developing an intuitive, functional understanding of context than I am in helping you memorize a few phrases you could regurgitate on an exam.
C: (Cutting in) Here, here!
R: (Continuing) I simply want to develop the idea as it pertains to learning objects and reuse. Has everyone seen the inkblot pictures psychologists use?
D: Sure. They're called Rorschach tests. Psychologists use them to get inside people's heads.
C: How can a blot of ink get inside your head?
V: The blots of ink are purposfully ambiguous in their appearance. They look like different things to different people. What you see in the inkblot - the way your mind interprets the shape - supposedly tells the psychologist something about what's going on in your mind. For example, if S saw butterflies, hearts, and bunnies in three inkblots, but you saw guns, knives, and explosions in the same three, that would be rather telling.
(S looks scandalized)
V: They've been broadly critizied, though, so I'm not sure they're the best example to use, R.
R: Whether or not they actually provide a window into the soul is moot for our purposes. Why is it that exactly the same image can appear so radically different to different people? It's the same image... why doesn't it appear exactly the same?
V: Well, if S is the type who spends lots of time thinking about bunnies...
S: (Interrupting angrily) Okay now...
R: Yes, come on V.
V: Okay, okay. Anyway, what you see in it is a reflection of yourself and your surroundings. If you grew up near the ocean and I grew up in the mountains, and neither of us had ever seen the other's landscape, that could influence what we saw in the ink blots.
O: In other words, your personal history and experience determine what you see?
D: Well to some extent, yes. People who lived in the 1500s would never have said an ink blot looked like a car or plane or rocket. Those things are completely outside the realm of their experience, and that would make it impossible for them to see those things in the splotches.
S: I had an experience like this once in China. I had flown to Beijing for some standards meetings and was spending the weekend after the meetings in a slightly more rural part of the country with a Chinese friend of mine who was interpreting for me. Anyway, during lunch I commented that a passing cloud looked like Jimmy Durante. My friend saw the resemblance right away but the people we were eating with did not. Turns out they'd never heard of him before.
O: Makes it a little hard to judge whether a cloud looks like him or not, then.
V: But what do ink blots and clouds have to do with learning objects?
R: I think they help us see that context contributes significantly to meaning. In all the encouters you have day to day with familar objects like chairs and computer mice and door knobs, you draw on your own history and experiences (your internal context) and your immediate surroundings (the external context) in order to make meaning of those things and decide how to interact with them. Let's have another look at the inkblot in the Wikipedia entry. What's the first thing that comes to your mind? Don't say them out loud, write them down on these slips of paper and give them back to me. (Pauses while people scribble on their cards and hand them back) So let's see... a bat, an anchor, a pelvic bone, a manta ray, a wolf's face, and... oh dear.
O: What!? Read it?
R: (Reluctantly) Butterflies, hearts, and bunnies.
(Raucous laughter from all but S)
R: (Shaking her head) Ok. I suppose I walked into that one. Now I'm going to show you one more, and I want you to wait again before saying what you see in it. (Pauses while people look). Did anyone not see a face at first?
D: Of course not.
R: Why not?
C: We didn't have much choice, did we?
D: Why not? What happened that made a previously abstract image look exactly like the same thing to all of us?
O: I expect we're about to hear that this blot of ink has some additional context.
R: What does that mean?
D: It means the artist tried to bias the way we would see the blotch.
R: Yes they did. But how?
V: By placing other pictures around the ink blot.
R: Yes. I want to make two points here. The first is that by placing additional images in the same frame, the author changed the context of the second ink blot. (Nods to O) I might have achieved the same effect by saying the word "face" as I projected the image on the wall. The first point is that the context of an ink blot - or a learning object for that matter - is the collection of things you encounter close to it in space (like the other images printed around the image in the frame) or in time (like saying the word and then projecting the image on the screen). The second point...
C: (Cutting in) Before you make your next point, I just have to say that I saw a skull and not a face. Sorry to ruin your first point.
R: Not at all. My second point is that contexts _suggest_ meanings and interpretations, but do not determine them, as C has just pointed out. (To C) Thanks for the segue. Context can provide strong suggestions as to how we should understand a thing, but we are free to make our own interpretations nonetheless. For example, when she was so young she couldn't even put her own coat on, my youngest daughter used to carry her jacket to my husband and plead, "put this on!", to which my husband would respond, "no, dear, that jacket's much too small for me." Now the context of the tiny girl holding up the jacket definitely suggested an interpretation. But to be silly (and to help her learn to be more precise in her speaking) my husband would willfully ignore the meaning suggested by the context and choose to make another interpretation. Most of the humor in popular sitcoms, or the jokes in Reader's Digest, operate on this principle of missing a meaning clearly suggested by a context.
D: I get it. The guys that wrote the movie Airplane! were geniuses at that kind of humor. Like the now famous "Don't call me Shirley" bit.
R: (Interrupting) Yes, I would highly recommend watching the film to understand how ignoring the suggestions made by context can lead to hilarity. Let's come back to the topic, shall we?
C: So are you going to say that people do the same thing with learning objects? That they misunderstand either willfully or for other context-related reasons? That can't be so... learning objects use words and images - not ink blots. They're not as ambiguous.
R: Words and images *do* seem to be less ambiguous, don't they? However, the Airplane! examples show that people can misunderstand even when they're deep in the context.
V: But the Airplane! examples are carefully scripted by writers, so I don't think it's fair to say they represent reality.
S: (Almost to himself) I've worked with a couple of people who talk like they came out of the movie...
R: I've prepared a few examples to show that the meaning of even the simplest words depends highly on its context of use. (Handing out slips of paper) Here are some slips of paper. Printed on each is a single sentence. C and D's make a simple dialogue, as do mine and O's, and S and V's. The second sentence in each of the three dialogues is the word "no" followed by an exclamation point. Let's begin by reading the dialogues aloud. C and D?
D: (Whining) Come on Mom, can't I please go out with my friends tonight?
R: Okay. Thanks. Now for me and O. (In an official sounding voice) I'm sorry to call so late maam, but there's been a serious accident involving your husband.
R: Thanks, O. S and V?
V: (In a news anchor voice) A report of a multi-billion dollar study released by the federal government today suggests that eating less and exercising more can result in significant weight loss.
R: Now each of these dialogues ended with exactly the same word, punctuated in exactly the same way.
O: But with completely different meanings.
C: When I said "No!" it meant "absolutely not, and don't ask me again unless you'd like to be grounded for a week."
O: When I said "No!" it meant "oh please, don't let it be true."
S: And when I said "No!" it meant "What idiot couldn't have figured that out without wasting my tax dollars?"
R: And how do you know what "No!" meant in each case?
V and D: (Talking over each other) By the words that came before.
R: So this single, commonplace word can have multiple meanings, and the only way we can determine which meaning we are supposed to understand with any given use is by the context of the conversation in which the use occurs.
C: (Nearing exasperation) So what does this have to do with learning objects?
V: Is it that the meaning of a fairly straightforward learning object may change depending on the other learning objects it is used together with?
(Everyone lokoks at each other in silence for a few moments)
D: But that would be a nightmare! Isn't there some way to make the suggestion behind our intended meaning a little less subtle? To leave no doubt about which intepretation we're hoping the learner will make?
R: Think back to the reusability paradox.
O: Let's just think back to the examples we used a few minutes ago. Instead of saying "No!" my character could have just said "Oh please don't let it be true!" That would certainly have left little room for interpretation.
R: And why is that?
V: Not to be a simpleton, but I believe it has to do with the number of words. In fact, I would guess that there is a relationship here - the more words used, the less room is left open for interpretation.
S: Unless you're a politician, in which case your sole purpose in life may be to violate that principle.
O: But I think V has a point. It's like the way the extra image of the coat, hat, and head narrowed the interpretations we could make of that second ink blot. Even though C saw a skull and not a face, that's a greatly restricted interpretive range compared to bats, anchors, and (grins, turning to S)
S: Don't say it. I'll point out, while we're here, that crossword puzzles work on the same principle. Each clue includes a word or two with multiple meanings, and you have to figure out which meaning is the one they really intend.
D: I'm having flashbacks of freshman literature - trying to guess which answer the teacher wanted on those ridiculous essays.
R: The difficulties, puzzles, and comedies created by lack of context do turn out to be rather universal once you know what they look like. But what is the relationship between what we are talking about now and the reusability paradox?
V: The more words you use, the more context you add, the more specific you make the suggested meaning, the harder it may be to get that highly-specified, concretized thing to feel like it really fits with any other piece of content?
S: I can tell you this much - over specification can be bad for other reasons, too. Look at the standards docs. They're hundreds of pages long. We can't get anyone to read them.
O: It's like my old high school English teacher used to say about the appropriate length for an essay. Whenever anyone would ask how long a paper had to be, she would get this kind of funny grin on her face and say, "Good writing is the length of woman's skirt. Long enough to cover the subject, and short enough to keep it interesting."
C: Nice! I haven't heard that one before!
R: Of course, we're not trying to keep things "interesting," per se, but I think her little witicism does express some of the tension between being concise to the point of ambiguity and being expressive to the point of inhibiting reuse.
D: Antoine de Saint-Exupery said something similar to your teacher. "Perfection (in design) is achieved not when there is nothing more to add, but rather when there is nothing more to take away."
R: Do you begin to understand why the answer to the question "how big should a learning object be?" is "it depends?"
O: Isn't the reality, though, that if we take a more open view of what learning objects are, that we're going to find things in a variety of states, and have a lot of stitching together to do?
R: (Imitating a bell) Ding! We have a transition point! Let's wrap this up for tonight and pick up here next time. There is, of course, a strong relationship between learning objects' sizes and the ways they can be put together. But let's do that next time.
V: You're not going to forget to summarize the night's questions are you?
R: I'd like to make a summary statement, instead, if I may.
S: You have our permission!
R: When we deal with learning objects, we have to deal with two kinds of context: "internal context" and "external context." What's going on inside an object is its "internal context." This internal context suggests interpretations of what the object "means." The internal context also exhibits a characteristic we might call "degree of suggestion," the extent to which the context suggests a single interpretation. And this degree of suggestion is generally corelated with the number of words and other media (like graphics or sounds) the object contains. Everyone with me so far?
R: The "external context" of a learning object is the group of other objects you want to use it together with. Because of the way context works to suggest meanings, the meaning of a learning object may change depending on which external context you place it in. Also, when a learning object's internal context strongly suggests a meaning unrelated to the meaning suggested by the group of objects that make up the external context, we can say that object does not "fit" in the collection of learning objects. (Looking around) Clear as mud?
D: I think I understood that. It is interesting to think that one object's internal context, in part, defines the external context of anything that uses it--pseudo recursion.
O: Works for me.
R: So the things to take away from tonight's conversation aren't questions themselves, but a vocabulary that will hopefully allow you to ask better questions, here are the terms:
- internal context
- external context
- degree of suggestion
Also, I'd like to recommend Wiley's extended rendition of The Reusability Paradox to you all. Several of the ideas we developed tonight came from that document. It's a little difficult to read, but our discussion should have prepared you for it.
V: Ok. I really have to unplug my brain for a while.
S: Ditto. Let's pick up here again next week.
R. Ok. See you then!