[comp.edu] Net Textbooks

fred@bu-ma.bu.edu (Fred Blundell) (05/17/89)

If hypermedia textbooks and automatic quiz generating software were
available at universities, students could study at their own pace. The
opportunity to be retested repeatedly would motivate them to go back
and correct their misunderstandings. One can imagine video
presentations at the quality level of the PBS series NOVA, followed by
illustrated text and exercises. A student would be required to
correctly solve several problems of every kind, at his own pace. Small
group discussion sections with faculty or graduate students could
facilitate interactive learning. The need for textbooks could be
reduced. The cost of both lecturers and textbooks could be greatly
reduced.

Traditional lectures in large auditoriums are inadequate because so
many professors are unskilled in educational techniques, and because
multimedia presentations rivet our attention more effectively than a
distant voice. Also, a student often needs the opportunity to stop the
flow of information to think for a few minutes, which is impossible in
a lecture hall. 

In subjects of highly standardized and technical content like freshman
calculus and computer programming, nationwide competence exams should
be available. That way, people who cannot afford a college education
and people who study technical subjects independently after graduating
in the humanities could certify their qualifications.  Self-paced
online computerized education, and national competence tests could
make technical education available to age groups and income groups
that have traditionally been excluded.

In my hometown of Huntsville, Alabama, it seems that the public 
library has almost become day care center, with many students from
the several nearby public schools waiting there for their parents
in the afternoons. Many of them play with the personal computers
that are available there. The public libraries could be an ideal 
setting for computerized educational systems.

The suggestion has been made that a free textbook foundation should
be established in the image of the famous Free Software Foundation,
to make electronic manuscripts available over the network. This is
a good idea, but productions of the depth and sophistication needed
to attract large numbers of users would require several hundred
man-years of highly skilled labor. It would seem to me that the
national interest could be served by some involvement by the Department
of Education. The philosophy that has relegated all educational
projects to the states is inadequate if our goal is to fully exploit
the potential of nationwide computation and communication networks
for informing people who are isolated from the flow of information
for reasons of age, health, economics, or geography.

--Fred Blundell

manis@faculty.cs.ubc.ca (Vincent Manis) (05/17/89)

In article <611348814.15174@bu-ma.bu.edu> fred@bu-ma.bu.edu (Fred
Blundell) writes:

>Traditional lectures in large auditoriums are inadequate because so
>many professors are unskilled in educational techniques, and because
>multimedia presentations rivet our attention more effectively than a
>distant voice. Also, a student often needs the opportunity to stop the
>flow of information to think for a few minutes, which is impossible in
>a lecture hall. 
As one who (quite unwillingly) gives lectures in large auditoria (I won't
comment on whether I'm skilled or not!), I have mixed feelings about this
statement, and the implied consequence that the solution lies specifically
in multimedia.

First of all, there is no doubt that the lecture is the least
efficient method known of imparting information. Psychologists have
for decades tested various ways of rote learning, and have found that
the lecture does most poorly. 

But do we really want university classes to concentrate upon the
imparting of information? I think not. Universities are supposed to be
places where students learn concepts and their applications, not to
mention clear and critical thinking. The primary use of a lecture
should be to motivate students to analyse what they're studying, to
give examples which illustrate specific concepts, and so on. 

It can of course be argued that having 200 people in a room offers
little or no chance for dialogue. This is of course correct. However,
the solution involves decreasing class sizes more than anything else. 

Second, multimedia have their own problems. It costs a fortune to
develop good multimedia, which means that the primary development will
occur on the part of those which have the financial wherewithal to
carry out such large projects. The financial justification has to
involve amortizing the cost over many years: hence such materials
won't be responsive to changes in paradigms or pedagogy over that
time. And, as textbooks amply demonstrate, flashy packaging is no
guarantee that the contents are worthwhile. Studies of texts in
various areas (life sciences are one example; a recent article in CACM
on textbook treatments of random number generators is another) show
that authors often lift treatments bodily from other books, not
bothering to determine whether the source is correct. There is no real
reason to believe that multimedia will be any different.

A better way of looking at multimedia is as another way of teaching,
added to our existing repertoire which includes lectures,
laboratories, assignments, books, audio- and videotapes, and so on. 

>In subjects of highly standardized and technical content like freshman
>calculus and computer programming, nationwide competence exams should
>be available. 
Here we come to a philosophical divide. Why are universities teaching
material of `highly standardized and technical content'? You can buy
good books on programming in my local supermarket. Why should a
university offer such material?

We all know that the point of teaching calculus to the average person
is not to have him/her know any mathematics. Rather, it's to teach her/him
how to solve various problems about ladders and bumblebees flying
between trains. This will presumably be useful in some undefined way
later on. The average university Calculus I course has very
little contact with epsilons and deltas these days (and the failure
rate is still around 25%!). 

Rather than considering the best way to condition people to recognise
L'Hopital's rule, maybe we ought to stop and ask why we're teaching
this stuff. In what way does it contribute to learning, as opposed to
satisfying a math requirement so somebody can get a degree as
expeditiously as possible? 

>The philosophy that has relegated all educational
>projects to the states is inadequate if our goal is to fully exploit
>the potential of nationwide computation and communication networks
>for informing people who are isolated from the flow of information
>for reasons of age, health, economics, or geography.

I don't wish to discuss US Federal arrangements (our own Canadian ones
are strange enough!) but as well as `information' one might want to
consider `ideas' and `opinions', other things which most definitely
belong in a university (and not just in mushy humanities-land, but
even, dare I say it? in computer science). If we really want to educate
people, rather than just training them, maybe we should stop and
decide upon our objectives before we decide how we're going to
implement those objectives.



____________  Vincent Manis                    | manis@cs.ubc.ca
___ \  _____  The Invisible City of Kitezh     | manis@cs.ubc.cdn
____ \  ____  Department of Computer Science   | manis%cs.ubc@relay.cs.net
___  /\  ___  University of British Columbia   | uunet!ubc-cs!manis
__  /  \  __  Vancouver, BC, Canada V6T 1W5    | (604) 228-2394
_  / __ \  _ 
____________  "I'm not making this up, you know!" -- Anna Russell
             
              

darin@nova.laic.uucp (Darin Johnson) (05/18/89)

>First of all, there is no doubt that the lecture is the least
>efficient method known of imparting information. Psychologists have
>for decades tested various ways of rote learning, and have found that
>the lecture does most poorly. 
>....
>It can of course be argued that having 200 people in a room offers
>little or no chance for dialogue. This is of course correct. However,
>the solution involves decreasing class sizes more than anything else. 

When I was in school, I found that sections (which were designed for
discussions and questions, etc.) had far fewer discussions than in
lectures.  Part of the problem was that students went to sections
expecting to go over answers to the last test, etc.  Also, there are
not very many students willing to speak up.  In a small section, that
usually amounts to no-one wanting to enter a discussion.  In a
lecture, there are enough people that discussions are started easier.
This was in CS, where there is plenty to discuss.  In math and
physics, lectures involved laying down the "way things are", and
discussions (if any) centered around "We don't get the same answers,
what are we doing wrong".  Often the students are so overwhelmed that
discussion is impractical.  This may have changed in upper division
classes since I had few of those in math.

>Second, multimedia have their own problems.

I haven't seen courses like this.  But in a couple beginning CS
courses, we had "self-paced" instruction.  This can in some ways
correspond to self-paced multi-media instruction.  There were lectures
once a week, but were optional after the first (and were attended
mostly by people who were lost or felt that missing a lecture was a
sin).  Grades were based on how much you finished.  After each
section, you would have an automated quiz, involving questions or
programming.  You also had to have a proctor verify that you had done
the assignments before taking the quiz.  This was all well and good,
but it was too easy to let that class slip while concentrating on
other classes.  You also had to do the sections in order.  If you
were on an early assignment and had an annoying bug that you couldn't
find, you had to stick to it, even though you may have already known
solutions to later problems; if you could take partial or no credit
for that part, you could then work ahead and get a better grade than
otherwise).  Multi-media outside of college would help, since you
could take as long as you wanted.

I also had a 2 unit foreign-language class once (since I had to have
12 units to be full-time, and I couldn't take upper division classes
until I finished ALL the lower division requirements) that was
self-paced and involved listening to language tapes, etc.  There was
no lecture (because you could choose any language there were tapes for,
and I had Danish :-) but 2 quizzes determined pass/fail.  This class
and the beginning programming classes had a common problem - you had
to schedule your time to the times of the lab, and could only get a few
hours a week.  The programming class also had pre-sign ups, so you
couldn't drop in at anytime (and woe if you aren't there when next weeks
sign up is posted).

There was another beginning programming class which was the more
traditional lecture style that you could take instead of the self-paced
class.  The catalog hinted that it was a tougher course and should not
be taken by freshmen.  I proctored in both classes and found the the
students in the lecture-style class had a much better understanding of
what was going on and did better in later classes.  Of course, this may
have been due to other factors - only better students took it, it worked
in spite of the lecture-style, you covered all the material, not just
what you completed, you had to work harder, etc.

Darin Johnson (leadsv!laic!darin@pyramid.pyramid.com)
	We now return you to your regularly scheduled program.

rang@cpsin3.cps.msu.edu (Anton Rang) (05/18/89)

In article <556@laic.UUCP> darin@nova.laic.uucp (Darin Johnson) writes:

   When I was in school, I found that sections (which were designed for
   discussions and questions, etc.) had far fewer discussions than in
   lectures.  Part of the problem was that students went to sections
   expecting to go over answers to the last test, etc.  Also, there are
   not very many students willing to speak up.  In a small section, that
   usually amounts to no-one wanting to enter a discussion.

Depends.  I went to a small undergrad school (about 4500 students),
and our senior-level CS courses typically had 8-12 students.  We had a
*lot* of discussion, and covered much more material than equivalent
courses at larger schools (at least, the two or three large schools
I've visited) did.
  Then again, it probably helped that everybody in the CS department
knew each other by that time.  I don't know how well having, say,
15-person sections would work when the people don't know each
other....

  Followups to comp.edu.

		Anton

+---------------------------+------------------------+-------------------+
| Anton Rang (grad student) | "VMS Forever!"         | VOTE on	         |
| Michigan State University | rang@cpswh.cps.msu.edu | rec.music.newage! |
+---------------------------+------------------------+-------------------+
| Send votes for/against rec.music.newage to "rang@cpswh.cps.msu.edu".   |
+---------------------------+------------------------+-------------------+

cooper%vlab.dec.com@decwrl.dec.com (g.d.cooper in the shadowlands) (06/14/89)

In article <2045@ubc-cs.UUCP>, manis@faculty.cs.ubc.ca (Vincent Manis) writes...
>In article <611348814.15174@bu-ma.bu.edu> fred@bu-ma.bu.edu (Fred
>Blundell) writes:
>>Traditional lectures in large auditoriums are inadequate because so
>>many professors are unskilled in educational techniques, and because
>>multimedia presentations rivet our attention more effectively than a
>>distant voice. Also, a student often needs the opportunity to stop the
>>flow of information to think for a few minutes, which is impossible in
>>a lecture hall. 

>As one who (quite unwillingly) gives lectures in large auditoria (I won't
>comment on whether I'm skilled or not!), I have mixed feelings about this
>statement, and the implied consequence that the solution lies specifically
>in multimedia.

A multimedia presentation may not be the optimal solution but it does
have a number of advantages over the straight lecture format.  A
combination of the two may be a better solution.  However a taped
lecture with corroborative material provides the lecture format and
remains a multimedia presentation in the hypertext type of
environment.  Implementation is the key here in regards to
effectiveness.

>First of all, there is no doubt that the lecture is the least
>efficient method known of imparting information. Psychologists have
>for decades tested various ways of rote learning, and have found that
>the lecture does most poorly. 

>But do we really want university classes to concentrate upon the
>imparting of information? I think not. Universities are supposed to be
>places where students learn concepts and their applications, not to
>mention clear and critical thinking. The primary use of a lecture
>should be to motivate students to analyse what they're studying, to
>give examples which illustrate specific concepts, and so on. 

You are making an unjustified assumption about what and how material
could be presented in a multimedia cai course.  Even if such a course
was limited to factual rather than to conceptual manipulations it
would still be able to replace much of the introductory level
instruction in a variety of fields.  The `hard' sciences are obvious
places for factual based instruction.  Even in such `soft' fields as
drama there are applicable techniques; a play on a video disc that can
be run in parallel with a commentary.

>It can of course be argued that having 200 people in a room offers
>little or no chance for dialogue. This is of course correct. However,
>the solution involves decreasing class sizes more than anything else.

If a significant fraction of the initial information can be accrued
outside of the lecture hall then the use of lecture time can be
improved.  Class size affecting informational transfer is true
predominatly when basic data must be recapitulated; lowest common
denominator limiting.

>Second, multimedia have their own problems. It costs a fortune to
>develop good multimedia, which means that the primary development will
>occur on the part of those which have the financial wherewithal to
>carry out such large projects. The financial justification has to
>involve amortizing the cost over many years: hence such materials
>won't be responsive to changes in paradigms or pedagogy over that
>time. And, as textbooks amply demonstrate, flashy packaging is no
>guarantee that the contents are worthwhile. Studies of texts in
>various areas (life sciences are one example; a recent article in CACM
>on textbook treatments of random number generators is another) show
>that authors often lift treatments bodily from other books, not
>bothering to determine whether the source is correct. There is no real
>reason to believe that multimedia will be any different.

Cost is not a reasonable limiting factor because of obsolescence.  If
the implementation is done in an extensible fashion then as new
material becomes available it is added to the package.  This
*increases* the amount of time over which amortization can be
calculated.

It is not valid to abjure a new technology because of the failures of
the old.  If bad information is disseminated in a textbook it is
extremely difficult to updated only the faulty pages.  In an
upgradable CAI package modification of erroneous material would be a
simple procedure.

>A better way of looking at multimedia is as another way of teaching,
>added to our existing repertoire which includes lectures,
>laboratories, assignments, books, audio- and videotapes, and so on. 

If you have all of this in a hypertext type of environment then presto
change-o multimedia CAI.

>>In subjects of highly standardized and technical content like freshman
>>calculus and computer programming, nationwide competence exams should
>>be available. 

>Here we come to a philosophical divide. Why are universities teaching
>material of `highly standardized and technical content'? You can buy
>good books on programming in my local supermarket. Why should a
>university offer such material?

A university needs to offer material such as this to provide the
uniform background necessary for higher level courses.  Basic
information such as the foundations of calculus does not change from
text to text only the presentation changes.  If a good basic calculus
CAI package was implemented then as it evolved it would provide a
suitable uniform environment for learning and would guarantee that all
students who finished had a certain level of understanding of the
subject.

>We all know that the point of teaching calculus to the average person
>is not to have him/her know any mathematics. Rather, it's to teach her/him
>how to solve various problems about ladders and bumblebees flying
>between trains. This will presumably be useful in some undefined way
>later on. The average university Calculus I course has very
>little contact with epsilons and deltas these days (and the failure
>rate is still around 25%!). 

This doesn't have anything to do with reasons to not develop such a
system.  If a CAI program teaches problem solving then it can replace
the instructing professor.  Also the failure bears out the fact that
the traditional methods of teaching the subject are *not*
satisfactory.

>Rather than considering the best way to condition people to recognise
>L'Hopital's rule, maybe we ought to stop and ask why we're teaching
>this stuff. In what way does it contribute to learning, as opposed to
>satisfying a math requirement so somebody can get a degree as
>expeditiously as possible? 

This doesn't appear to have any correlation to whether or not a
subject could be taught in a different manner.  It is totally outside
the domain of the problem.  What is required to be learned and why it
is required is a question to be asked before requiring a course.  It
is not related to how the course will be taught.  Typically courses
are requirements because they hold some necessary information that
will be used either to learn more about their subject or in the use of
their subject.  Programming FFTs is fairly difficult if you failed
differential equations. 

>>The philosophy that has relegated all educational
>>projects to the states is inadequate if our goal is to fully exploit
>>the potential of nationwide computation and communication networks 
>>for informing people who are isolated from the flow of information 
>>for reasons of age, health, economics, or geography.

>I don't wish to discuss US Federal arrangements (our own Canadian ones
>are strange enough!) but as well as `information' one might want to
>consider `ideas' and `opinions', other things which most definitely
>belong in a university (and not just in mushy humanities-land, but
>even, dare I say it? in computer science). If we really want to educate
>people, rather than just training them, maybe we should stop and
>decide upon our objectives before we decide how we're going to
>implement those objectives.

What subjects/styles in which a student is trained/educated is a
separate issue from what tools are used to train/educate that student.
If the decision is to train in critical analysis of ideas then the
tools are oriented towards that approach.  If the concensus is to
train towards fixed manipulations then they are appropriately changed.
Selecting objectives does not invalidate the concept of multimedia
CAI.  The selection process is a criteria in the development.

	 CAI to teach the basics, professors to do the rest,

				shades
============================================================================
| But I that am not shaped for sport- | Geoffrey D. Cooper                 | 
| ive trick, nor formed to court an   | cooper%vlab.dec.com@decwrl.dec.com |
| amorous looking glass...            | business (508) 467-3678            |
|                                     | home (617) 925-1099                |
============================================================================
Note: I don't work for DEC I'm a consultant.  My opinions are just that MINE!