COMMENT ⊗   VALID 00012 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00003 00002				   Computing Made Easy
C00008 00003	And now let me show you two quotations.  These quotations present two
C00012 00004	Now for a little historical perspective.
C00015 00005	Early history
C00022 00006	We come not to the early 19th century and to Charles Babbage.  Let me show
C00030 00007	And now it is time to get back to our story about Charles Babbage and Lady
C00040 00008	But why use computers?  Let me explain this by the next slide.
C00048 00009	So now let us look at some of these machines.
C00064 00010	Now I am sure that some of you are just waiting for an opportunity to ask
C00070 00011	We are now more or less up to the present time.
C00072 00012			SLIDES
C00075 ENDMK
C⊗;
			   Computing Made Easy
				    or
		   What You Should Know about Computers

I. Introduction.

The papers are full of articles extolling the power of the computer and
telling of all the marvelous thinks the Computer is doing and will be
doing in the near future.  Seldom, if ever, do the articles tell how the
computer does these things.  Is it any wonder that the general public gets
the idea that the computer possesses some sort of magic and that an
understanding of how it does these things is quite beyond the layman.  This
is simply not true.

In spite of all of the false information about computers that is floating
around, computers are quite easy to understand.  Oh, of course, you will
not understand all of the inner workings of the computer but after all how
many of you understand the inner workings of your car and this does not
prevent you from using it.  What you want to have is a functioning
understanding, that is, you would like to know what the computer can and
can not do.  Perhaps I can help you to get this sort of understanding.

You will have to give up any idea that the computer possesses some sort of
magical power and you may even have to do a little careful thinking.

Oh, of course, if you really want to understand all the details of inner
workings of the computer, you will have to go back to school and study
Electrical Engineering and Computer Science.  If you already know Calculus
perhaps a two year course might suffice.  How many of you really
understand how your television set works?  Never-the-less you feel quite
comfortable in using it.  You do not fear it and suscribe to the belief
that it is a threat to mankind.  After all television can do harm through
too much emphasis on violence and crime.  We do not blame the television
for this, but we do blame the people who choose the programs.  So please
do not blame the computer for some of the wrong things that people program
it to do.

Will it help, if I tell you that the basic principles of the digital
computer were first suggested by an Englishman Charles Babbage in 1833,
yes, 1833 not 1933.  Furthermore, the first really clear explanation of
these ideas was written in 1844 by Lady Lovelace, the 28 year old daughter
of Lord Byron, the English poet.  If Lady Lovelace could understand the
principles of the modern computer some 100 years before it was built,
surely you should be able to do this today.  Let me show you a picture of
this early Victorian lass.

SLIDE 1.  Lady Lovelace

Just remember that the principles of the modern digital computer are
really quite old and they are perfectly simple to understand.
And now let me show you two quotations.  These quotations present two
quite different points of view, each true if read carefully but each
subject to misunderstanding by the unthinking.

SLIDE 2.

    "The Analytical Engine [the Computer] has no pretensions
    whatever to originate anything.  It can do whatever we know
    how to order it to perform."  Lady Lovelace in 1844

This is as true today as when it was written.  It seems to set a serious
limit to what the computer can do. It also puts the blame for any failure
of the computer to perform some desired task directly upon man's lack of
knowledge as to how to order it to do the desired task.  I will have more
to say about this later but let's now look at the second quote.

    "Man has in a single generation found himself sharing
    the world with a strange new species: the computer and
    computer-like machines."  Marvin Minsky in 1967

Also true, if one does not ascribe human characteristics to this new
species.  Of course, the computer is not animate and to call it a species
implies that it is, but it is true that the computer can and does perform
many functions that formerly were within the exclusive domain of the
animate.  Let's follow out this rather whimsical idea and look at the
capabilities of this new species.

SLIDE 3.
                This Strange New Species

                1. The ideal servant
                2. No will of its own
                3. Unfailingly obedient
                4. Very fast and accurate
                5. Never sleeps, never tires
                6. Does only what is ordered
                7. No imagination, no emotions
                8. Perfect memory, instant recall

With a list of characteristics such as this, it is hard to see what all of
this public furor is about.  The computer can hardly be the threat to man
that some people seem to think it is.  You have all read or heard the
story of the Sorcerer's Apprentice, so it behoves us to be careful how we
use this ideal servant, but the servant is not to be blame for our
misdeeds.
Now for a little historical perspective.

Perhaps I should first present an outline of what I hope to cover in this
talk.  I am going to have to be very selective and limit myself to only
what I think that you should know.  If I leave things out that you think
that I should cover, or if I fail to make myself clear, do feel free to
interrupt me at any time.  It was with some misgivings that I decided to
use slides because this tends to introduce an element of formality to 
the talk. This is a very informal talk with you my friends and you are
expected to talk back. 

SLIDE 4.

        What You should know about Computers

        1. Early history
        2. Computers vs. calculators
        3. The parts of any computer
        4. Why we need computers
        5. Their speed and accuracy
        6. Programming (writing orders)
        7. What computers are now doing
        8. What computers will be doing soon
        9. The long range future

You will note that The first item is written in Old English to remind you
that the computer is not so modern, at least not in terms of basic ideas.
The last item, the long range future is also very hard to read. (On the
slide this is written in very small type.)

I have made this outline primarily so that I will not forget some parts
of my story, but I would be very happy if you would interrupt and ask
questions.  It is more important for you to understand something about
computers than it is for me to finish a set speech.

So now back to early history.

Charles Babbage is usually credited with the invention of the digital
computer.  However, its origins go back much further, back even to the
dawn of recorded history.
Early history

The problem of keeping records is perhaps some of the dullest repetative
work that falls the lot of the so-called white-collar worker in an office.
This has always been so, and it has always lead to the development of
mechanized aids.

In fact, if one goes back far enough, it seems that the invention of
writing itself was prompted by the need to keep accounts.  The oldest
surviving written records, dating from some 5000 years ago, were primitive
books of accounts inscribed on clay tablets by the Sumarians.  Curiously
one of the earliest written records that was not of this type contained a
lament for the "good old days" and a statement that the world was going to
the dogs because children no longer obeyed their parents.

While the Linear A writings of the early Minoans has not yet been
decyphered, all of the surviving records in their Linear B language were
bookkeeping records, and this well before 1470 B.C.

Even in Egypt where most of the records that have survived are on temple
walls and are praises of the Pharohs and of the Gods, the real reason for
the invention of their form of writing is thought to be the need to
preserve survey records of land holdings during the annual flooding of the
Nile.  The preservation of historical and cultural information could well
be left to oral transmission, but when it came to settling disputes
between adjacent land owners, nothing took the place of a written record.

Writing can be looked upon as being a mechanical means of supplementing
and even of replacing man's memory.

Similarly there has been a long history of attempts to make mechanical
devices to supplement and replace man' ability to do arithmetic.

Perhaps the earliest device that has survived to this day is the abacus. I
have one here.

The abacus came to us from the orient and was not introduced into Europe
until around 1000 A.D.  Its design is based on the method of representing
numbers known as "positional notation", a notation that was introduce by
the Arabs and that you all know but not by this name.  I am sure you have
all heard of the man who discovered rather late in life that he could
speak in prose.

To understand the importance of a positional notation consider the problem
of multiplying two numbers using Roman notation.  Try, for example, to
multiply MDCXVIII by MDCCCCLII.  By rewriting these numbers in the more
familiar "positional notation" form the problem is greatly simplified, but
multilpication is still a more difficult operation than is simple
addition.

The invention of logarithms by Napier in the 16th century, made it
possible by the use of tables to multiply two numbers by adding their
logarithms.  This was a long step in the simplification of computations.
Unfortunately, the general public has never become conversant with
logarithms and so this is still within the domain of the scientist and the
engineer.

But to return to mechanical aids.  The abacus is still primarily a memory
device in that it remembers the initial and intermediate states of a
calculation.  To develop any speed in using it must still be able to add
up to 9 plus 9.  Many attempts were made through the centuries to make
mechanical adding machines; Pascal and Leibnitz, in the 17th century made
notable contributions along these lines but it has only been within the
19th century that the desk calculator came into general use.

The next slide will show you the Pascal calculator which was built in
1642.  On it, you can see the forerunner of the modern telephone dial.
Many modern devices are not nearly so modern as modern man is apt to
assume.

SLIDE 5. The Pascal calculator (1642)

We will have to skip the next 200 years, not that there were not
developments during this period but time does not permit me to dwell on
these.
We come not to the early 19th century and to Charles Babbage.  Let me show
you his picture.

SLIDE 6. Charles Babbage

Yes, he was just as sour as he looked and he could never explain his ideas
properly, but he was a man well before his time. I wish that I had the
time to tell you more about this interesting man and of Lady Lovelace's
part in promoting his ideas.

Babbage first comes to our attention with the invention of what he called
the Difference Engine as shown on the next slide.

SLIDE 7. The Difference Engine (1822)

This device was actually what we would call today a calculator, not a
computer but it did do a great deal more than the small vest pocket
calculators that we have today. Babbage spent most of his life in trying
to build such devices.  However in 1833, when work on the Difference
Engine was halted for a year because of lack of funds, Babbage conceived
of his Analytical Engine.

An now, I suppose, it is time to point out the differences between a desk
calculator or its more modern form the pocket calculator and the digital
computer.  The most apparent distinction is one of size and cost but this
distinction is becoming blurred as more and more elaborate calculators
come on the market and as computers are becoming smaller.  Some pocket
calculators are, in fact, really simplified computers, but to clarify your
thinking I want to draw a sharp distinction.  I have tabulated the
principal differences on the next slide.

SLIDE 8.

Computers vs Calculators

        1. Size and cost
        2. Permanent files
        3. The stored program
        4. Multiplicity of steps
        5. Alternate program paths

We have disposed of the first item.

When one uses a simple pocket calculator, one enters the numbers into the
calculator as they are to be used and one causes the calculator to perform
one operation, that is one addition, subtraction, multiplication or
division at a time.  If one uses a calculator,for example, to figure one's
income tax, one must keep track of the sequence of operations that are
required and one must enter the needed numbers and depress the needed
operation keys in the proper order if one expects to arrive at the correct
answer.

When one uses a computer, one supplies it with all of the numbers that it
will need ahead of time, and in fact many times the numbers are already
stored in the computers permanent files.  This is item 2 on our list.

Then one supplies a list of the operations that are to be done.  The items
on this list are called instructions but they would be more correctly
identified if they were called orders.  The list is called a "program".
Whhen everything is ready, the computer is given the signal to start,
where-upon the entire series of operations is done without further human
invention.  This is the third item.

Just as one must do the required operations in the proper sequence when
using a desk calculator, so one must have the desired operations specified
in the proper sequence in the program that one gives to the machine.
There is one big difference however.  If one writes a program to compute
ones income tax then this same program can be used again and again to
compute the income tax for other people.  The program remains the same and
only the input numbers change.  To handle the many different situations
that can arrise for different people, the program can and does get to be
quite complicated and thus contains many instructions.

Now actually computing one's income tax is much more complicated than just
performing a number of additions, subtractions, and multiplications, as
all of you know.  There are all sorts of decisions to be made, usually
based upon the comparisons of two numbers, and I quote, 
	 "If line 14 shows a loss
	 Enter one of the following amounts
	 1 if line 5 is zero or a net gain
	   enter 5% of line 14
	 2 if line 13 is zero or a net gain
	   enter line 14, or
	 3 if line 5 and line 13 are net losses
	   enter amount of line 5 added to 50%
	   of the amount on line 13".

When you use a pocket calculator you must make these comparisons and you
must do the right thing.  When you use a computer, it must make these
comparisons and it must then do the right thing.  So the program is a good
deal more complicated than being simply a list of mathematical operations
that are to be performed.  It must also contain commands that will cause
the computer to make comparisons and to jump to different places in the
list of instructions for its next instruction depending upon the outcome
of these comparisons.

This then is the remaining, and certainly the most inportant, difference
between a calculator and a computer.
And now it is time to get back to our story about Charles Babbage and Lady
Lovelace.  Babbage's Analytical Engine was to have all of the
characteristics of the modern computer. as we have just enumerated them.
And remember, this was in 1833.  Babbage was quite unable to explain his
ideas at all clearly.  Fortunately, Lady Lovelace became interested in
Babbage's Analytical Engine and was able to explain it.  As she described
Babbage's machine, it was to consist of 4 parts, as listed on the next
slide.

SLIDE 9.
        The four parts of a computer

        1. The arithmetic unit
        2. Storage unit or units
        3. Input-output mechanisms
        4. Comparison and switching

As an interesting aside, I might tell you of the subterfuge that Lady
Lovelace used to present these ideas.  It happened that Babbage was asked
to deliver a lecture in Italy on his machine.  An Italian, who later
became one of Geribaldi' generals, wrote a paper reporting on this talk.
Writing technical papers was very unladylike in Victorian England, but
translating was very gentile so Lady Lovelace translated this paper into
English.  Since the paper suffered from defects that were largely due to
Babbage's awkward way of expressing himself, Lady Lovelace added numerous
footnotes, more voluminous in fact than was the paper itself, in which she
explained what it was all about.  These footnotes are the best explanation
that exists of Babbages's ideas.  Lady Lovelace's explanations are so much
more lucid than are Babbage's writings that I sometime wonder row much of
what we attribute to Babbage really should be credited to her.

Now let me describe Babbage's Analytical Engine and by analogy the modern
computer in much the same terms that Lady Lovelace used in 1844.

If a machine is to perform the functions of a human computer it must
possess four distinct parts.  These are:

1) An arithmetic unit, capable of performong the normal operations of
arithmetic.  Babbage called this unit the mill.  A pocket calculator has
such a unit.  In the modern computer this is a part of the CPU (central
processing unit).

2) A storage unit, that is, a mechanism which will retain numbers and also
the instructions that will be needed.  Babbage called this part of machine
the Store.  In modern computer terminology it is called the Memory.  When
a person does a calculation by hand he uses pieces of paper to supplement
his own memory, and by analogy we could call these pieces of paper a
memory.

3) Input-output mechanisms which allow the operator to put numbers and
instructions into the machine and to extract the results of the
calculations from the machine.  Babbage proposed to use punched cards as
input, and this was 1833 remember. and proposed to have the maching set up
its results in type where necessary.  We have many different kinds of
input and output devices available today, from typewriter like devices to
machines which automatically read type or magnetically recorded records.
Your dividend checks are one form of printed output which the modern
computer produces.

4) A comparison and switching unit, that is, a built in ability to compare
two specified numbers (that may have beeen computed) and to take different
courses of action depending upon the result of this comparison, that, is to
go to a different place in the list of instructions (the program) for the
next instruction.  People sometimes say that the computer "decides" what
to do next, but this is ascribing human capabilities to the computer.  The
decision was actually made by the person who wrote the program and the
computer is only doing what it was told to do.  This part of the computer
is a comparison and switching device only.  It does not decide anything,
it only does what it is ordered to do.

While I am on the subject, let me take a moment to rail against the
all-too-common failing of computer scientists to use anthropomorphic terms
in describing computer behavior.  It is all too easy to call the storage
device a memory, to say that the computer remembers when some needed
information is retrieved from the storage device, to say that the computer
decides etc.  When one puts two unequal weights into the two pans of an
old fashion balance and the beam tips toward the heaver side we could with
equal reason say that the beam decided to tip this way as to say that the
computer ever decides to do anything.  Unfortunately it has become the
fashion to talk in this way, I do it myself, and I may even have done it
during this talk.  But always remember that the computer is only a
collection of inantimate devices and that these devices are only obeying
the physical laws that govern all inatimate objects.  If there has been
any deciding this was done by the person who wrote the program.  Of
course, he may not have known how the calculation was to come out and his
decisions may have been conditional: If the first number is larger than
the second, do this, if they are equal do some thing else and if the
second is the larger do yet a third thing, etc.

This tendency to use anthropomorphic terms misleads the public and causes
them to think that they can not understand computers.

That's all there is to a computer, just four quite prosaic parts, prosaic
at least in terms of their function, but, of course, they are marvels of
engineering perfection in terms of the way in which they are built.

Now in case this discussion has still failed to disaffirm your belief in
the mystical powers of the computer, let me state catagorically that the
computer can not do anything that a person cannot do with a pencil and
piece of paper.  The computer can only do it faster and with less danger
of making an error.  Lady Lovelace was right.
But why use computers?  Let me explain this by the next slide.

SLIDE 10.
        Why do we need Computers?

        1. People are too slow
        2. People make mistakes
        3. People are too expensive
        4. There are not enough people
        5. Computers create more jobs
        6. Many tasks impossible without

So now we have all of the ideas as to the functioning of a modern computer
and we have the need, this all in 1833, but Babbage was never able to
complete his machine because the technology of the time was not up to the
task.  Babbage's machine was to be entirely mechanical, electronics, and
indeed even the electron was unknown.

We have then a lapse of roughly 100 years.  In fact it was not until the
computational demands of the second world war arose that a serious attempt
was made to realize Babbage's ideas and it was not until the war's end
that working models were actually available.

Viewed from the present vantage point, it is hard to see why the computer
was so slow in coming into being.  As early as the mid 1920 the necessary
technology was in hand.

Perhaps one of the reason was the fact that attention had turned toward
the so-called Analog Computer and that substantial progress was being made
with this type of computer, notably by Professor Bush at MIT with a
particular type of analog machine known as the Differential Analyser.  I
was fortunate in being able to work on this machine when I entered MIT as
a student in 1923.  The Analog Computer represents numbers by physical
quantities, perhaps a quantity of electricity or the number of rotations
of a wheel or a screw.  Numbers are added by simply rotating the wheel an
amount to represent the first number and then an additional amount to
represent the second number and then observing the total amount of
rotation of the wheel.  The accuracy that can be obtained with the Analog
Machine is, of course, limited to the accuracy with which the rotation of
the wheel can be measured.  This accuracy at the time when I worked on the
project, was about 2 percent.

Even by this time, the military wanted to use the computer for the
computation of the trajectories of shells and so to replace the need for
expensive test firings of the guns, and in fact we were successful in
developing a method which involved making a simplified calculation,
ignoring certain effects, such as the variation in air resistance with
speed and with elevation, the variation of gravity with elevation and the
Coriolis effect.  This gave a solution with an error of perhaps 5 percent.
Then one used the Analog computer to compute the difference between the
true solution and this approximate solution with an error of 2 percent,
yielding a final solution with error of 2 percent of 5 percent or 1 tenth
of a percent which was just tolerable.

During the period from the early 20's to the start of the war the Bush
machine was greatly improved and it was on machines of this type that
almost all of the compution during the war years was done.

One of these Bush Machines was at the Aberdeen proving ground where a
young army lieutenant Herman Goldstein was working. When the work load
became to large for this machine some of the work load was transfered to
the University of Pennsylvania where two young graduate students, J.
Presper Eckert and Jonn Mauchly were working with another copy of the Bush
machine.  Eckert and Mauchly, aided and abetted by Herman Goldstein
induced the government to finance the development of a digital computer
which would circumvent the accuracy constraints of the Bush machine. This
machine turned into an ungainly monster called the Eniac that employed
over 1800 vacuum tubes, completely filled a 30 by 50 foot room and cost
400,000 dollars to build.  This was the first machine that actually
embodied most of the ideas that Babbage suggested in 1833 and it was not
completed in time to be of any use during the war years.

Professor Aiken at Harvard University also addressed himself to this
general.  He asked IBM to work with him and togather they developed the
Mark 1 computer which also was completed at about the war's end.

When the war was over, Eckert and Mauchly formed their own company to
develop the Univac computer, John Von Neumann, who had collaberated in the
University of Pennsylvania work, went to the Institute for Advanced Study
at Princeton and started work on a machine that was to be a prototype for
many different machines both in the United States and elsewhere and the
race was on.  I could tell many stories about the trials and tribulations
of these first few years, but time does not permit.
So now let us look at some of these machines.

SLIDE 11. The Manchester machine

The first picture is of a machine built in Manchester England for which
Fred Williams and Tom Kilburn deserve the principal credit.  Obviously you
are not going to be able to tell anything about it except that it was
large. Prof. Williiams was the first to develop a satisfactory storage
system, which, of course, was soon superceded by others, and Prof. Kilburn
was the first person to suggest a major improvement on the basic ideas
advanced by Babbage.  He proposed what is now called indexing, which is
simply a clever way to alter the effect of a set of instructions by means
of yet other numbers that are stored in the so-called index registers.

SLIDE 12.  The IBM 701

This was the first IBM machine that embodied all of Babbage's ideas.
Actually as originally marketed it did not yet include the Kilburn
improvement.  It was designed for scientific computing, and IBM, with some
misgivings decided to build 18 of these, a number that was thought to be
large enough to take care of all of the computing needs more or less for
all time.  Stanford University, alone has more than 1000 times the
computing capacity that these 18 machines provided.

SLIDE 13.  The IBM 702

This was aanother early machine, designed in this case for business use,
it being assumed at that time that a basicly different machine was needed
for Data Processing as contrasted with the 701 which was designed for
scientific computing.  By the way I want to apologize for favoring the IBM
computers as much as I do but I am naturally more familiar with these and
also it was easier for me to get pictures of their machines.

These first machines all used vacuum tubes rather than transistors as the
transistor was not made public until 1947, I believe, and the first
transistors were entirely too unreliable for use in computers.

I am however going to skip ahead roughly by roughly 10 years and show you
a more modern computer whic had roughly the same capacity as the 701 and
702 computers did but was at least 10 times as fast.  You will observe
that the entire machine had also gotten smaller physically.

SLIDE 14.  The IBM 360/40

And now a modern large computer, the IBM 370/148.

SLIDE 15. The IBM 370/148

And now just one more computer, this time a Hewlett-Packard computer which
illustrates the trend in physical size and appearance.

SLIDE 16.  The H-P

It is in fact hard to show, by over all pictures, what is actually taking
place in computer design, because very little reduction in the size of the
input and output devices has taken place.

I can show what is really taking place however by the next slide.

SLIDE 17. Comparison in size

The various pieces of equipment show the trend in size of equipment to do
a simple function in the computer.  The vacuun tube circuit was used in
the IBM 701.  This next piece is from one of the first transistorized
computers.  Then two more steps in the trend are shown.  Unfortunately
this slide is a bit old and it does not show the latest.  Actually it is
impossible to show an equivalent computing element because we now build a
great many circuit elements all togather on a single chip of silicon. Just
by way of comparison one such chip will usually do 20 or 30 times as much
as the old tube circuit did and it all occupies a volume about equivalent
to that taken by the nail on the little finger of a rather petete lady.
So the over all reduction in size of the components used in today's
computer is roughly a factor of one thousand.  The end is still not in
sight.

Along with this very dramatic reduction in size has gone an equally
dramatic increase in speed and over all capacity.

So let's look at the next slide, which compares the speed and accuracy of
the modern computer with a person.

SLIDE 18.
        Speed, Accuracy and Reliability

        1. One million times as fast
        2. One million times as accurate
        3. Less "down time" than people
        4. Automatic error checking
        5. Double-entry bookkeeping
        6. Computers blamed falsely

Strangely enough I have been quoting this figure of 1 million for the
relative speed for some years.  However, when formerly I was compariing
the computer with a man using only a pencil and a piece of paper the
comparison can now be made with the man aided by a small calculator.

Now as to its accuracy:  A computer seldom makes a mistake.  When I say
seldom I am of course refering to seldom in relation to the number of
operations that it performs.  If a computer makes one mistake a day, and
this number is high for a well maintained computer doing routine work, it
still will have performed well over ten thousand million correct
operations before it makes a single mistake.  Compare this with your own
rate of error when you try to add a string of numbers.

People who work 8 hours a day can, in computer terms be considered to be
down two thirds of the time. For operations requiring continuous manning,
it is customary to have 4, not 3 workers for each operation.  If this much
back up were provided with computers we would never see the time when the
computer was down.

Because of the speed of the computer, it is quite reasonable to
encorporate all sorts of error checking techniques, and then one can use
all of the standard double entry bookkeeping techniques that are used in
manual systems.

Now lest you forget Lady Lovelace's astute remark, let me paraphrase it as
shown on the next slide.

SLIDE 19.
           "The Computer has no ability
        whatever to originate anything.
        It can do only whatever we know
        how to order it to perform."
        paraphrased from Lady Lovelace

        As true today as it was in 1844!


In spite of the fantastic speed and accuracy of today's computer it still
will only do what we tell it to do and we can not tell it to do anything
that we ourselves do not know how to do.

So now I must say a word or two about Programming.  A program is simply a
list of orders.  However, writing this list of orders is not at all easy.
In fact, the power of the computer depends entirely upon the skill of the
programmer.  If the programmer writes a foolish program, the computer will
do foolish things.  Let's look at the next slide.

SLIDE 20.
        Programming

        The programmer must
        1. Fully understand problem
        2. Use a programming language
        3. Anticipate all complications
        4. Write orders in minute detail
        5. Make no typographical errors
        6. Leave nothing to "common sense"
        7. Be completely "literal minded"

Fully understanding the problem is not as simple as it sounds.  It is a
common experience for the programmer to be asked to write a program to
take over some operation only to find out that there is no one single
person in the organization who knows all of the details as to how the
former manual system really worked.  The knowledge is actually diffuse
through out the organization and most of the people in top management
positions actually have wrong ideas as to how it really works.  This does
no particular harm in the manual system because the people who actually
do the work, do the right thing even when told to do something that is
wrong.  But now the programmer is told what to do by this manager and he
does not know that the manager is missinformed so the program is wrong.
The misinformation usually has to do with the handling of special
situations which may not occur every day so the error may go undetected
for some time when suddenly the computer seems to make a mistake.

Another stumbling block has to do with the need to use a special
programming language. The language that the final commands must be written
in is very difficult to use.  So there has developed a number of so-called
higher languages in which programms are usually written.  But such a
program can not be used directly to control the computer. It must be
translated into machine language.  Fortunately, we can use the computer to
make this translation if we write a program in machine language which is a
complete set of commands to the machine ordering it, in a step by step
fashion to do what is necessary to translate the higher language into the
machine languase.  Fortunately, in most cases, some one else has already
taken care of this process and there is such a program already available.
Now to get the desired program to run we must go throuugh a two step
process, we must first run a "compiler program to do the translating and
then we can run the program that this produces.  The problem is even more
complicated than this bsecause we would like to use the computer in the
process of writing the program and the computer can not help in this
directly unless some one has written still another program containing
orders in machine language as to what must be done to assist in this
writing process.  We need what is called an Editor program.  It happens
that I have spent a large part of my time during the last few years in
writing such a program and in improving it to make it do a better and
better job.  But the end is still not in sight.  A computer, without what
is called an Operating System which is simply the name for yet another
program, such a computer would be extremely hard to use and could be used
by only one person at a time. So somebody has had to write an operating
system.  At the AI lab at Stanford we have a very good but very
complicated operating system which allows as many as 49 people to use the
computer at the same time, with each user behaving as if he had the
machine all to himself.  The only difference that he can see is that the
computer seems, at times to be a bit slow.
Now I am sure that some of you are just waiting for an opportunity to ask
me why it is that computers make so many mistakes if they are so fast and
so accurate.  Well, I have a slide to answer that.

SLIDE 21.
        "Computer" mistakes

        1. Wrong input figures
        2. No manual checking
        3. Special situations
        4. Problem of changes
        5. Fraud protection

Most so called computer mistakes are people mistakes.  In a manual system
there are many opportunities for checking, some of these planned but
others simply because the people operating the system have good common
sense and can frequently sense an error.  Suppose you buy some small item
at a store and charge it.  The sales clerk can easily make a mistake in
writing out the charge slip, maybe making the price one hundred and fifty
dollars instead of one dollar and fifty cents.  If this has to be posted
by hand the clerk entering the item will probably know the approximate
price of the item and call the sales clerks attention to the possible
error. This was not the posting clerks responsibility but just good common
sense. At the end of the month your bill must be prepared, Again a clerk
posting the item may well notice the error.  But now consider a
computerized accounting system.  The sales clerk can still make a mistake,
but once the item is listed in the computer no one ever sees the item
until you get your bill.  In principle, the cost of all items in the store
could be kept within the computer and the program could cause the computer
to check each item as the sales clerk enters the original transaction, but
in a large store with many items, with continuely changing prices, with
special sales etc. this becomes a big job and it is usually not done.  So
the computer makes a mistake.

Then there is the problem of writing a program for the computer that
actually does what the manual system does.  Frequently the managers of the
operation do not know how it really operates.  Perhaps some erroneous
instructions have been issued.  The people on receiving these orders know
enough to realize that they are not correct but rather than calling
attention to themselves by protesting, they simply ignore the instructions
and do the sensable thing.  Never the less when the programmer writes up
the set of orders for the computer there erroneous orders are
incorporateed into the computer program and the computer, being an
obedient servant does exactly what it is told to do.

Then there is the problem of correcting a computer program. If the manager
of a manual system finds an error he need only tell one or two people that
a change will have to be made, and he can actually make a mistake in
telling them what is to be done but again the people receiving the order
interpret it correctly and do the right thing but not the literal thing
that they were told to do. Making this same mistake in correcting a
computer set of commands will perpetuate yet another error. And so it
goes.  The computer demands the same sort of accuracy from the people that
the program itself is capable of giving.

There is still another complicating factor.  In a manual system, many
people are usually involved and fraud is not easy.  In a computer system
one need only change the computer program to make it do all sorts of
illegal things.  So most systems are engineered so as to make it as hard
as possible to make changes.  This is all right as a method of fraud
protection but it then makes it extremely hard to make legitimate changes.
One thus frequently is told that it is impossible to make some trivial
change in the way that a computer system operates.
We are now more or less up to the present time.

And now we come to the future.  What can we expect in the near future.


SLIDE 22.
        The near future

        1. Tele-type writing
        2. Personal computer
        3. Checkless economy
        4. Weather forcasting
        5. Automated factories
        6. Personal newspapers
        7. Direct-access library
        8. Better telephone system

And now the long range future

SLIDE 23.

        The long range future of computers
            (Economics will dictate)

        1.  A mechanical house maid
        2.  Self steering automobiles
        3.  Completely automated factories
        4.  Automatic language translation
        5.  The Information Society

By way of summary I will show one last slide.

SLIDE 24.
        What You know about Computers

        1. Arcaiologia
        2. Computers vs. calculators
        3. The parts of any computer
        4. Why we need computers
        5. Speed and accuracy
        6. Programming (writing orders)
        7. What computers are now doing
        8. What computers will be doing
        9. Long range future of computers
		SLIDES

 1. Lady Lovelace
 2. Quote
 3. This Strange New Species
 4. What You should know about Computers
 5. The Pascal calculator (1642)
 6. Charles Babbage
 7. The Difference Engine (1822)
 8. Computers vs Calculators
 9. The four parts of a computer
 10. Why do we need Computers?
 11. The Manchester machine
 12. The IBM 701
 13. The IBM 702
 14. The IBM 360/40
 15. The IBM 370/148
 16. The H-P
 17. Comparison in size
 18. Speed, Accuracy and Reliability
 19. Lady Lovelace paraphrased
 20. Programming
 21. "Computer" mistakes
 22. The near future
 23. The long range future of computers
 24. What You know about Computers