Re: [Patents] Software Patents: The Patent Trap

[Bouthier Christophe <Christophe.Bouthier@loria.fr>]

On Mon, Feb 25, 2002 at 10:09:17AM -0500, Gregory Aharonian wrote:
> You need to do a lot more research about these issues.

That's why I'm trying to discuss on the mailing list, so that people
that have the information (or just links to the information) can guide
me on the right way.

> Otherwise, you are jumping straight to a conclusion making it pointless
> for me to spend time answering your question.

Sorry, this was not my goal. Feel free to forget the conclusion.
I would really like to have your point of view on my questions.

> First, while the intent of IP laws is to influence economics, the
> wording of IP laws has nothing to do with economics.  It doesn't matter
> what some EC directive says or starts out with, even they have to deal
> with the law as it is written.


I'm not sure to follow your point here, or to really understand your
idea. Please correct me if I misinterpret your meaning.

Are you talking about the content of the laws ? The articles themselfs ?
Am I right to think that your point is to say that in the articles 
of IP laws, there is nothing written about economics concerns ?
And so, that the fact that software distribution costs could be null is
not a significant difference (with respect to others engineering fields)
concerning its patentability ?

Once again, I just try to understand the real meaning of your point.


It could help to remember the start of the discussion.
You write, in older mails :
" Software is just another field of engineering - nothing noble about it,
  and no reason its activities are any different than any other field of
  engineering."
And in another mail :
" There is nothing special or holy about software -
  just another technology to be patented.
  Problems? Sure. Problems with patent quality, problems with the costs
  of patent litigation.  But problems shared by biologists, chemists,
  electrical engineers, software engineers, etc."

Then I tryed to find differences between software and other engineering
fields, differences that could legitimate differents patenting laws 
between software and others technology.
One of the argument is that the distributing cost of a software could be
null.

And you replyed that "IP laws, patents and copyrights, mostly ignore
such economic considerations".

And I answered that the introduction of the new european proposition
was all about economic consideration.


So, I hope that I have well understood your meaning. If not, 
feel free to correct me. 


Here :
http://www.europa.eu.int/comm/internal_market/en/indprop/02-32.htm
it is written, under the "Would the proposed Directive restrict the
interoperability of computer programs?" section :

"In fact, Member States patent laws, while not fully harmonised, do not
 in general extend to acts performed privately and for non-commercial 
 purposes."

So, I agree that I haven't seen in any law any concern about the 
distribution cost of a product, either in copyright or in patent law. 
I haven't seen in the European Patent Convention any mention of the fact
that if an invention has no distribution cost, then it shouldn't be 
patented. I agree with you on this point.

Still, it seems to me that IP laws are made, as you tell, with the
intent to influence economics. And there seems to be, as the
"non-commercial purposes" citation paragraph suggest, economics
consideration in the application texts of IP laws.

(I'm prudent here because even if it's written on the page of the
European Commission (URL page given), I am not a specialist
(yet ! :-)) in patent law, and I admit that I don't know this law,
or in which Member state it applies. I would be very interested by
any pointer to it :-) ).

Am I right ? 


> ANd as the law is written around the world, for the most part you can
> only patent process and method, and can't copyright process and method.
> That is the law of Europe.

I agree with that. 
Especially with the "you _can_ only patent process and method" (emphasis
added).


> SInce software is all process and method, no
> matter what it's cost, the laws as written make it subject to patent
> law.

Of which law are you talking here ? Of the new directive ? Of
the European Patent Convention ? Of an international treaty (TRIPS
maybe) ?

I thought (but feel free to prove me wrong) that the only law that
defined what is patentable and what is not is the European Patent
Convention, with the (un)famous Article 52.
If it is the case, can you tell me where it is told that every process
and method are patentable ? I didn't see it, but then I could have
missed it.

Moreover, the fact that a law told us "this is patentable and this is
not" is not really the point here, it it ? I thought that the point of
the discussion is :
Is there any differences between software and other engineering field
that make its activities different and thus call for different patent
laws ?

Isn't this the point of the discussion ?
(We should definitively agree on what we are discussing, if we hope
 to have good arguments :-) ).


> That some European laws say software is copyrightable (as do some
> US laws) doesn't mean that software should be copyrightable, but rather
> that politicians are too lazy and ignorant to remove these contradictions
> from the law.  Courts on both sides of the Atlantic have ruled over the
> decades that about the only protectable about computer software is the
> distributed executable, and nothing else (because everything else in the
> source code is structure and function prohibited from copyright because
> it is a process).

I agree with you on the fact that executable form is protected by
copyright. 
I would rather add that executable form seems somehow efficiently
protected, as the BSA told us :-)

Concerning the source code, I'm a little bit less sure.
I remember stories of Apple and Be legal problems because they were
using a GPL source code without respecting the license
(redistributing...). I think it was about a network component or
something like this... And another story about Microsoft that had used 
source code of another companie...

I'm sorry, I didn't found web links to thoses stories.
I just found the link to the Microsoft story, but it's in French...
Here it is :
http://www.legalis.net/legalnet/actualite/oeuvres_num.htm#flash11

But I think that I understand what you want to tell : you just have to 
tweak a little bit source code, changing the overall aspect (we are not
speacking about font and colors, but about variables, loops, objects,
...) and you can get something really different (from the source point
of view) but almost exactly the same from the functionnality or GUI
point of view.

Have I well understood ?


> So it is sad to see people like Hartmut advocating copyright for Open
> Source because for the most part that leaves Open Source unprotected
> (which is why many Open Source companies, despite their public stances,
> are patenting their software).

Well, for the source code, you cannot reuse it just like it is.
This is (still) protected, I think. Have you link to courts ruling that
the same source code, with no modificiation at all, is not protected ?
I would rather think (but please prove me wrong) that even with small
modifications (for example, just changing the names of variables and 
functions), the source code could still be considered as copied.

It seems to have been the case here (sorry, still in French) :
http://www.legalis.net/legalnet/jurislog.htm#contrefaon

m I wrong to think that ? On this page :
http://www.legalis.net/legalnet/index.htm
there seems to have several example of such things.

I have also read the link on your page that you give me last time :
http://www.bustpatents.com/aharonian/softcopy.htm

You have a big part about "CASES RULING ARGUING [...] SOFTWARE I
IS COPYRIGHTABLE" and a small part, at the end, about
"CASES RULING ARGUING [...] SOFTWARE NOT COPYRIGHTABLE".
(Well, you know, you written it). As the first part is a lot more bigger
and filled than the second part, the first impression for a new reader
is that you are proving that software is copyrightable.

I haven't read all the case in the first section (pro-copyright). But I
have read the shorter second part. As I understood it, what was decided
not protected by copyright are :
- small part of a program (flow and structure of a subroutine).
Was the reject because the part is too small ?
- instruction set of a processor
- structure of a database (It was in 1992, before the European 
Directive of 1994. Would such a jugement still be held now ?)
- purpose or function of a program (Is it not the purpose of patent ?)
- user interface
- data file
- menus structures (I would put it in the same bag than user interface. 
Do you agree ?)
- data structure.

None of thoses is source code (except the subroutine one. I would be
interested by more information on this case).

So, do you have rulings that say that someone took the source code of
someone else, modified it, and was not ruled guilty of copyright
infrigement and didn't have to pay fees or something like this ?

Such ruling could be then analysed to see their differences between
the French ruling given above 
(http://www.legalis.net/legalnet/index.htm)


> The cost issue to me is irrelevant.  I don't care how much it costs to
> manufacture and distribute something, especially since modern
> manufacturing is making this easier and easier to do.

Yes, I agree that manufacturing is easier and easier to do.
Still, when you build, even really easily, say a microchip, you cannot
give it for free. Not to a lot of people.

When I do a program, I can (and do) give it to anyone that want to use
it. And for free. And I'm satified because I think that I helped 
other peoples. And I think that this is one of my right to help other
people and to let them use what I have done because it cost me nothing
to help them. And I think that this is my right to distribute what I
have done, as I have the right to distribute on my web page what I have
written.

Maybe I'm wrong to think that this is a right. What do you think ? Is it
a right or a privilege ? Shoudn't this right came into the balance
between inventor and citizen ?

And are IP laws written to put on the paper such a balance ?
Am I wrong to think that IP laws are made to balance rights between 
inventors and citizen ?
What do you think on this subject ?


> It is apparent you are not an inventor.

Well, nothing I have "invented" is really so usefull. 
I'm still writting my prototype for my PhD thesis.
It will be something that has never been done before, and I hope that it
would be usefull. 


> If you were, you would realize
> that the cost that patent laws are trying to protect are the cost of
> making the invention.  That is what the patent laws are meant to be
> incentive for - the costs (money, time) that the inventor puts in to
> develop his invention. 

For my prototype, the big deal is not about the idea. Althought it's an
interesting idea, that none had before, it's not such a big deal. 
My thesis is about "group awareness". And my idea is to use the context
of the user, build for example with bayesian networks and neural
networks, to adapt the awareness to the user.

The idea is better explained in an article published in the CITE 01
conference. 
(And as it has been published, from what I know about patent law, it
couldn't be patented now. Am I right ?)

But the big deal is to develop the code. Because it use a lot of
component (bayesian networks, neural networks, packet sending, 
information receivers, information representations, ...).

And if another one want to do the same product, it will have to put a
lot of time in it : not to understand the idea, but to code it.
So, my valuable part is not in the idea or the functionnality, but
in the time put in writing the code.
And, I think, this is protected by copyright. Am I wrong ?


> And like every other area of engineering, 
> software takes money and time to invent.

I don't agree on that. I think it takes time to develop a new, say,
mechanical device, or chemical molecule, because you have to test it
to be sure to have the right functionnality.

But in software, the idea is not such a big deal. What takes time is
to write the software. 


> And great software (which
> doesn't describe most Open Source software) takes a lot of time and
> money to develop.

Well, I don't know how you can judge what is a "great" software.
I think that this is still a big research field, to be able to 
judge the quality of a software. There are so many parameters
(encapsulation, evolutivity, bugs, adequations with requirements, ...).

And to make such a statement implies that you have took a look at "most
Open-Source software". With just the 34 500 projects in sourceforge
http://sf.net
that takes a little bit of time, especially if you have to judge if they
are "great".

So please, no flamebait here. I just want to discuss, seriously, with
you, because I found it more valuable to discuss and exchange idea with
somebody that DON'T think like me than with somebody that do.
So, please, let us stay with real arguments and real reflexions.

> But even that is secondary, because the patent laws, AS WRITTEN, only
> care about the nature of the technology to be invented - is it a
> machine, a process, etc.

For the definition of "what is patentable", yes I agree. We can not tell
that software is considered by the law as an exeption because of its
economical caracteristics.

But, if I'm not wrong, as told at the start of this mail, this is not
the point of the discussion.

> SO try inventing, try running a company, etc.  It will give you a 
> different perspective.
> 
> Greg

Well, in fact, some years ago, I tried with some friends of mine, while we
still were in engineering school. It was about a new giga parallel
computer. There was part-hardware and part-software. The great part was 
the BIOS, that should look on multiple PCI buses to find others nodes.
Really fun :-)
Well, we stop before founding the compagny. The leader came into US and 
is now the CEO of Intalio (www.intalio.com), but that's another story.

And yet, in the whole process, we always agree to put whatever sources
we would made on Open-Souces.

So, I don't think that's the fact that you are an inventor really change
the perspective. At least not for me. But I can imagine that it could do
it for others.

Yet, I would like to have your answers on the problems posed on this
mail, hoping that we could continue this really interesting discussion
on the point of view of pro- and anti patent, but, for once, only based
on fact and reflexion, no hysteria or flamebait :-)

C.
-- 
By the grace of God Almighty
And the pressures of the marketplace
The human race has civilized itself.
It's a miracle.                      -Roger Waters-Amused to death-