NEIL BROWN

INDEPENDENT SOFTWARE ENGINEER

MR. BROWN:  Well, I have a lot to say. First of all, I'd like  to
thank you all for being here.  I have a lot to say and I appreci-
ate the opportunity to say it to you directly.  It's been a  long
time  since  I've been on stL sorry.  My name is Neil Brown.  I'm
an independent software engineer.  I work  as  a  contractor  for
software development companies and I do development of my own.  I
did programming for fifteen years and I've been getting paid  for
it only in the last six.

Although I can't speak for all of us, I can speak for some of us,
and  I can certainly speak for all the friends I've talked to who
feel very similar to the way  I  do.  I  represent  the  ultimate
source  of  all  revenue  for  every  person who profits from the
software industry, however indirectly, the developers;  if  some-
body  wasn't  writing  the  software,  there wouldn't be a single
software patent lawyer that could possibly make a penny.  This is
why I __ what I, what the League for Programming Freedom and what
the Free Software Foundation, neither of which I'm a member, have
to  say  is  important.  We power this industry; we're the dynamo
which causes it to exist, and in order to keep our jobs, in order
to  continue  marketing  software,  we  must  make  clear what we
needPore its too late.

I'd like to address the questions that you've asked, as  I  indi-
cated  from  not listing affiliations, I'm not speaking on behalf
of anyone other than myself and those in  the  software  industry
whom I have found to agree with me.

On the question, Topic A, Question 1, Example A,  "What  part  of
mathematical  algorithm implemented on a general purpose computer
can be patented?"  My response:  None whatsoever.

The technique of long division where one writes the number to  be
divided  down,  puts a little bar under it, and writes the number
to divide into it, next  to  it,  writes  notations  and  partial
answers, gradually arriving at a more complete answer, can clear-
ly be described in a fashion executable by a computer.  It's  not
at  all  hard  for most developers to write a graphical front end
for long division, so this is a useful, I mean  the  question  of
whether or not division is useful is, is not worth debating. It's
a useful tool to accomplish a useful goal, and real money is made
from  using  it, but wTwould happen to your education?  How would
you have learned division if the school that was trying to  teach
you  that  suddenly found itself being attacked by another school
who claimed to own that very method?

A mathematical algorithm performed on a special_purpose computer;
can  you patent a calculator?  I do seem to remember that the be-
ginning of the digital revolution was really noticed when  pocket
calculators  starting  causing  slide rules to disappear. If only
one company had been able to produce calculators, would the price
have  dropped from four hundred dollars to fifteen in only a cou-
ple of years?  If only one company could own the legal  right  to
build a machine to perform mathematical calculations, where would
the software industry have gotten its start?

Topic A, Question 1, Example B, sorry, Example C  and  C_2.  "Can
you  patent  the  disk  on  which  a computer program is stored?"
Again, with the calculator, if it were  possible  to  patent  the
concept  of  a  calculator,  if  it  were  possible to patent the
cXlator that has the ability to execute more  than  one  program,
where  would  the  software industry be today?  How do you define
what a program is?  Do you define it as being able  to  push  the
sine  key  and  get  the  sine  of  the  number?  Do you define a
separate program as one in which you can press the cosine key and
get the cosine of that number?

Question 2. "What impact, negative or positive, have you or  your
organization  experienced  from patent issues on software_related
inventions?  On several occasions I have found myself  being  un-
sure  of whether or not I was able to use a particular algorithm,
specifically the compression algorithm embedded  in  the  program
known  as  Compress,  within software. There have been many ques-
tions raised and lots of time spent chasing after whether or  not
the company could somehow use this and escape any royalty obliga-
tions.

"What implications, positive or  negative,  can  you  foresee  in
maintaining  or  altering  the standards for patent eligibility?"
This is Question 3. Well, I seet, because every one of  the  hun-
dreds  of  ideas that the developer goes through while writing an
application, every one of the little techniques that he  goes  to
use  or  goes to put together with another, he has to go and call
up Legal to find out if that's been patented or if  it  might  be
covered  by a patent.  The software industry is not going to pro-
gress very rapidly if people like me spend all their time on  the
phone to Legal asking if we can do this.

"Does the framework of patent, copyright or  trade  secret  law,"
Question  4,  "effectively  promote  innovation  in  the field of
software?"  Yes. "Does it provide the appropriate level  of  pro-
tection?"   A  qualified  yes.  The qualification is that it pro-
vides too much protection, potentially.

Question 5. "Do you believe a new form of protection for computer
programs  is  needed?"   My  answer  is  no.   The water is muddy
enough.

On Topic B. I agree with all concerns that access  to  prior  a`s
difficult, or is outmoded.  The difficulty of determining whether
or not two programs are equivalent or similar is extremely diffi-
cult.   I  deem  it intractable.  There are so many languages out
there, there are so many sophisticated ways of  expressing  algo-
rithms  that it's hard enough just to understand one, but compar-
ing two?  Doing this for every program out there  that  seems  to
possibly be related to an application can take forever.  The very
concept of, is there anything out there at  all  that  does  what
this  does,  is extremely difficult to solve, and I deem it to be
intractable for software in general.

Topic B, Question 1. No, I  don't  think  that  the  patents  and
printed publications provide examiners with sufficient collection
of prior art, and as I  said  before,  it  can't.   The  work  on
software interface patents, if you allow patenting of the idea of
having a hammer on a desk and deem it a  different  invention  if
the  hammer is the drawer of the desk, how is somebody using that
desk going to be able to gdheir job done?  How is somebody  going
to  be  able  to design __  How is anyone going to be able to get
the job done if their job is to design a new desk and  they  have
to  go and find everyone that has similar functionality available
at the top of their desk?  The patent which has control  informa-
tion such as page numbering and position on the page and document
being edited available on the screen, the  very  idea  of  having
many  pieces of information available for manipulation of the in-
formation is the whole idea of an interface.  You want to be able
to provide as much ability for the user to manipulate the raw ma-
terial they're working with as apparently is possible.  You  want
all  of  these  tools to be easy to get to and easy to work with,
and if someone comes with a formalism for  making  all  of  these
things  available, then if any means of providing that same func-
tionality is deemed equivalent, then how  can  progress  possibly
happen?

I do have lots more to say, as I said, but __

COMMISSIONER LEHMAN:  Well, mayh can help you to wrap up in  just
asking you, I have a little confusion in your statement to us; in
the answer to Question 4 you  basically  seem  to  say  that  the
present framework is okay, but I have the impression that you ba-
sically don't think that __ that you think there  are  a  lot  of
problems with the patentability of software just generally. Is it
your position that software should not be patentable?

MR. BROWN:  Software patents are a  blight.  They're  a  problem.
They get worse.

COMMISSIONER LEHMAN:  So basically you think that copyright  pro-
tection is __

MR. BROWN:  Copyrights and trade secrets.

COMMISSIONER LEHMAN:  __ is okay, and trade secrets, but that ba-
sically the difficulties from trying to work with the patent sys-
tem applied to this industry are so great that it virtually makes
it  impossible to use it as an effective technique for protection
that developers like yourself can really work with.

MR. BROWN:  Yes. How many houses would you build if every time  a
carpenter went to build a houselery time a carpenter went to take
up a tool he had to pay a point one percent royalty on the  gross
profit on that house, or the gross revenue on that house?

COMMISSIONER LEHMAN:  We appreciate your coming and sharing these
comments. Thank you.

MR. BROWN:  Thank you very much.

COMMISSIONER LEHMAN:  Next I'd like to ask, if he's here,  Gordon
Irlam, representing the League for Programming Freedom, which our
previous witness referred to in his statement.

Back to The San Jose Index

Forward to Gordon Irlam