Learning from Logo - 1986 LME
Abstract
Liddy Nevile
Barson Research
335 Johnston Street
Abbotsford, 3183
Victoria, Australia
While there is a reasonable amount of literature about the design of microworlds in Logo emerging, there seems to be little written about what makes one microworld more appealing, more powerful, than another. This aspect of microworlds has aesthetic qualities, and as such is perhaps least explicable in Computer science terms. Nevertheless, I believe it is the most important , and that the aesthetics of microworlds should be discussed if only so they become describable.
In creating a number of microworlds to support the learning of such concepts as three-dimensional representations of space, isometric and perspective, tessalations in terms of reflections and transformations, data-bases as intelligent collections of inter-active data, the concept of recursion itself, and so on, we have confronted above all the question of what metaphor will best establish a link between the user and the knowledge in the system?
----------------------
Learning from Logo
When the first Logo papers emerged with dreams of children creating meaningful microworlds for their own purposes, and incidently building powerful learning environments in which they could absorb knowledge by osmosis, the imagination of many was fired.
The reality, fifteen years later, is that teachers (as a gooup) do not know how to encouyrage this activity on a long term basis, the the learning does not have obvious enough transfr into the standard school syllabus, that programming skills fall far short of the needs of the programmer, and generally that what was promised is being doubted.
We, in Victoria, have been teaching and using Logo for more than 10 years, and yet do not feel any urgency to abandon the task. Instead, we battle harderto broaded the cultural support for using Logo, and argue that even when there is not a lot of hard-data research report which prove the effectiveness of Logo, there is potential which should and could be realised. This is so, we say, despite the fact that for many, the use of Logo is exhausted, after the 'using it for computer awareness' stage.
For the last year I have worked hard to discover some metaphoirs which I believ are to a large extent the missing links in many Logo users' knowledge base.
The need for somple metaphoric representations of Logo itself have been responded to more formally in recent years. For instance, Brian Harvey has written his chapter on recursion four times in one book, to facilitate the reader's comprehension of this process (i).
In 1985 a number of papers appeared which offered suggestions about how micrworlds could be designed in Logo. These were not to be used for teaching programming techniques as such, but as environments for learning about something in particular, e.g. balance. The knowledge which the user would be expected to bring to the activity might vary, and so might the learning which would take place.
In reaction to what appeared to be a 'soft option' compared wiht those an environment such as Logo could offer, I undertook some work to produce microworlds for school use which would bridge the gap in what I felt was an appropriate way.
The basic design criteria were roughed out as follows (by myself and Colin Fox who has worked on this project with me):
- that any microworld should hhave a low threshold rentry point in terms of Logo programming skills, even if the programmes themselves use "advanced" techniques;
- that any microworld should have a subject content which would benefit from the support of an environment such as Loog could provide;
- that the use of the microworlds should be two-fold: while offering interesting contexts for using and developing progrsmming skills and activities, they should also offer stimulatinf subject-specific environments which could be used independently of programming skills;
- that while each microworld should be seen as a unit in one sense, a number of increasongly sophisticated sets of procedures should be provided for interaction with the microworld;
- that while all programmes should consist of easily examined and altered procedures, the procedures which control the system should be easily idenitified and modified for personal use;
- that as far as possible, Logo and it s associated epistemology of learning developing from a fragmented base of knowledge, should be presented in an exemplary fashion.
The last of these design ctriteria is of course the most elusive.
How can a topic be represented so that a number of people with differing backgrounds and styles of learning can benefit from the same environment?
Andi diSessa suggests that in time there will be sufficient knowlkedge about the learning process and the knowledge which people bring to a learning task, for us to design microworlds which can respond to the knowledge of the learner (ii). SWhat seems to be the problem then, is whether, with this knowledge, the microworld designers will choose accuracy or appropriateness as the inclusion criteria for their microworld.
It seems that the stumbling block for many microworld users, is not the smallness and relevance of the knowledge embedded in the microwold, but the way it is presented, accessed, modified, and thius learned.
A simple microworld example might be one with set of procedures which merely extract an element form a set and print that element. For instance, given a list of nouns, a procedure called NOUN might pront one of the nouns, chosen randomly. Some such sets may be subject specific.
While I was using these procedures with a group of curriculum developers in New Zealnd who preferred cricket to computers, their attention was captured when they related their work to cricket. Within a short imtw, the cricket microworld (which had lost its identity as my set of procedures) contained an extensive data base of onformation about cricket. There were lists of the acceptable actions which take place on the field, of the positions in which cricketers are founf, of the scoring which results from certain events, of illegitimate cricket acts, and even of the comments which Aussiue commentators make about New Zealand cricketers.
What were the critical elements of this microworld?
I believe the original starting point, a very simple sentence of the type
DOG BITES BILL
and which became
NOUN VERB NOUN
was the key to the exercise.
In a number of ways this was a metaphor for the collectin of cricket information, fo a data base. The information about what happens was easily gathered, and recognized as such, under the heading ACTION. Further information stores could be made under the heading NOUN but more headings soon needed to be added. All the time the metaphor was used to support the activity: each time new classifications were developed , sentences using the new information were run to test the material The relationship of one piece of information to another was testable according to the "does it mae a sensible sentence?" test. In the same way, the information it self could be tested accordong to whether the sentence conveyed a meaningful cricketing situation, or not.
This is not a new, or unusual, use of Logo but the example is often trivialised unnecessarily, I believe.
When computer users are struggling to use computers to work with information using other data schemas, they are often confused by the result they achieve. I beleive that moving from pencil and paper collections of data to computer-based ones is often a wasteful exercise because the computer is expected to be just a bigger electronic collection modelled on the paper and pencil version. The power of the collection of data in Logo is of another dimension, and until the user can create a metaphoric representation which can be used to consider what is being done by the computer, the computer's power will be neglected.
Mike Sharple's boxes are a good example of this (iii). He has created a microworld which uses the metaphor of a box as the container into which some information can be out. The problem is, however, that the information which goes into one box may well be a box in itself, and other boxes may be expecting access to the information in that element/box too. this the model of this data collection is difficult: it is not a hierarchically organised set of informtaion , it is not a set of records which is related by simple matching, and so on. It is not just a computer version of what can be done with paper and pencil. In fact, the model has the same sorts of attributes as the recursion one which Brian Harvey was dealing with in his book. It is, in a sense, merely an extension of the NOUN type example above. But the difference is that the extra dimension which is added is not trivial or easily explained.
I believe that to a large extent the value of a microworld depends upon the appropriateness of the metaphors upon which it is designed. The sentence is so simple a metaphor that it is often not recognised for its role. It is so powerful that its role can be redefined time and again according to a vry wide range of situations, styles, and users.
I hope that in tim we will be able to learn to look for good metaphors, and so be able to offer some more satisfactory answers to those who question the value of Logo in education.
(i) diSessa, A>< KNowledge in Pieces, Address to th Fifteenth Annual Symposium of the Jean Piaget Society, 1985.
(ii) Harvey, B., Computer Science - Logo Style, McGraw-Hill, 1985.
(iii) Sharples, M., Cognition, Computers and Creative Writing, John Wiley and Sons, Chichester, 1985.