[Commons-Law] "Bill Gates and other communists" article by RMS (Anivar Aravind)

[Hasit seth]

Hello,

I apologize for mispelling Aravind's name as Arvind in last post.  My
sincere regrets.

I have a great respect for Richard Stallman (RMS). Not just as a great
programmer (GNU and Emacs) but also as a visionary who envisioned that
free software is a practical idea.

There is enough buzz about anti-software patents that it is hard to
answer so many concerns.  Why are patents a problem only in two areas
of software and pharma-biotech?  Patents over mechanical, chemical,
electronic, electrical technologies seem to have no problem for those
who oppose software and pharma-biotech patents.  In theory a mechanic
with screws, nut-bolts, weilding machine, sheet metal and a forging
press should be able to create aeroplanes, cars and what not.  Such a
mechanic would be infringing thousands of patents.   Take a chemist,
who has a periodic table and bottles of all the elements listed there,
could theoretically make every possible patented medicine.  But this
has never been a concern because the innovation in these areas is
complex.  Common sense tells us that making an argument for such a
hypothetical super-mechanic or chemist is naive, since inventing is
hardly so obvious as to hamper other inventors innovations.

The computer you are using has probably hundreds of patented IC
circuits and the Intel/AMD/VIA processor inside it has hundreds of
patents protecting it.  Some circuit designers may be hampered by
these patented processors in their goal to design a new processor. One
solution is that they can license the patents. But other solution is
what the patent system really wants  you to do  - invent around it,
design around it, make your own invention.  This is the key, but being
"obstructionist" in apprerance, patents really encourage more
innovation by forcing inventors to find new ways towards the same
goal.

 In a sense this is economic waste, why reinvent the wheel? But wheel
since the ancient time has not remained the same, think that Goodyear
was not allowed to make wooden wheels due to a hypotehtical patent and
that led him to develop the rubber process that ultimately led to
pneumatic rubber tyres - that would be innovation though the imaginary
wheel patent was "obstructionist".  Story goes that in 19th century
one US patent office recommended that the patent office be closed
since all that was to be invented was already done and nothing more
remained.  But innovation and invention is an never ending source.  We
humans are masters of finding new ways of solving problems.  Plus,
some alternative that was developed due to a patent hurdle may be
useful in some other field, who knows?  Since innovation is
incremental, more patent disclosures (now within 18 months a patent is
filed it gets published for others to learn) provide scientists and
engineers even greater base to invent more.

Now software programmers do feel does that mean that I have to read a
thousand patents before I write a 1000 line program?  Compare software
innovation with the two microprocessor and wheel situations, a lot of
answers are there - though not all.

Pharma-biotech patents concerns morality and social issues (arguments
of type: should a dying person be not able to get medicine just
because it is expensive and patented).  It is a question of economics
of invention and social policy about health, hence they are out of
context when considering software patents.

But software patents are something else.  Here is my take on this
topic.  First, because of internet and cheap computers, any one can
create even a complex operating system.  Well Bill Gates did it
(MS-DOS) and so did Linus Torvalds  (Linux).  The entry barrier to
creating a complex software invention is very low, and hence there is
a valid argument for freedom to write software without worrying about
patents.  If I alone can do it why should I not be able to do it?

Second, the whole argument got momentum because of a stupid job done
by U.S. Patent Office in examining the inventions and granting a lot
of inital patents without looking at prior art.  The fault lies in
examining software patents before issuance, rather than anything
innately wrong with software patents.  Otherwise, all sorts of
mechanical and chemical patents too would be troublesome to inventors
in those fields and to the public at large.

Third, it is also a question of time. Software sprang up in last 20-30
years at a pace unmatched by any other technology. This meant that
initially the patent offices had very little prior patents to rely on
in rejecting the early patent applications and offices didn't have
trained examiners who knew a lot of software technical literature to
reject those patent applications.  Over time, a large pool of prior
patents will become available and those in patent practice are already
finding it terribly hard to get very broad patents issued.  More the
prior art, narrower the patents, and more is the inventive freedom to
design around it.

Fourth, lexicon is a problem with software.  It is hard to express
software elements in terms of language. Today, every programmer can
recognize terms like parsers, browsers, protocol, kernel etc., and one
cannot even think of getting a patent on these.  But software develops
at rapid pace, and words do not exist most of the time to describe a
lot of software. This ends up in a patent office as a broadly
described software patent which claims the result but not the process 
by which it is achieved.  This is just a matter of time, just like
mechanical and chemical fields, terminology will become standardized
and we will see a lot of less broad software patents.

Fifth, the level of disclosure detail required by patent office.  US
Patent office required inventors till early 20th century to submit
actual models of their invention.   They discontinued this practice
since mechanical inventions got narrower and narrower.  For biotech
still sequence listing are required to be deposited.  I wish they
required this for software too, so that it is not claimed at the level
of an idea but an actual method backed by progam listing or
pseudo-code to back it up.

Sixith, physical nature of software is unlike a chemical or a machine.
It is expressed in writing, which as RMS mentions should be protected
by copyright.  Though it is writing, it is a set of instructions that
makes a computer behave in particular way. Hence, it is also
functional, and copyright does not extend to functional stuff, but
expressions.

Seventh, so much of open source software is being developed without
software patent posing as problems.  SCO v. IBM battle is hardly about
specific software patents, but more about copyright/code theft.  Java
virtual machine architecture from Sun exists and so does virtual
machines from Microsoft's .NET virutal machines.  Though empirical,
this does point out that the problem is not an epidemic that has
paralyzed software development around the world.

Eight, tons of stuff can be said about the way the patents are
enforced.  Getting a patent is  just the beginnning.  To enforce it is
a bigger problem, perhaps than defending it.  Keeping aside the high
American legal fee issues, patents with broad to narrow scopes keep
getting invalidated in courts all the time.  An issued patent when
litigated gets attacked by a million pieces of prior art (which can
include even an obscure Ph.D thesis dated before the invention in a
small library of a tiny college in some little country barely visible
on a map)   and not just those which were available to the examiner. 
In short, enforcing a patent is a herculean task, and it is no wonder
that many broad patent enforcement suits died before being harmful in
anyway to innovation.

There is a lot more to be said about this. Some other time.

Regards,
Hasit Seth

On Wed, 16 Feb 2005 12:00:06 +0100 (CET),
commons-law-request at sarai.net <commons-law-request at sarai.net> wrote:
> Send commons-law mailing list submissions to
>         commons-law at sarai.net
> 
> To subscribe or unsubscribe via the World Wide Web, visit
>         https://mail.sarai.net/mailman/listinfo/commons-law
> or, via email, send a message with subject or body 'help' to
>         commons-law-request at sarai.net
> 
> You can reach the person managing the list at
>         commons-law-owner at sarai.net
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of commons-law digest..."
> 
> Today's Topics:
> 
>    1. Fwd: 1970 patent law quick review (Anivar Aravind)
>    2. "Bill Gates and other communists" article by RMS (Anivar Aravind)
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Wed, 16 Feb 2005 07:18:53 +0530
> From: Anivar Aravind <anivar.aravind at gmail.com>
> Subject: [Commons-Law] Fwd: 1970 patent law quick review
> To: commons-law at sarai.net
> Message-ID: <35f96d47050215174852aff4b at mail.gmail.com>
> Content-Type: text/plain; charset="ISO-8859-1"
> 
> -------- Original Message --------
> From:   "André Rebentisch" <arebenti at web.de>
> To:     in-parl at ffii.org
> Subject:        1970 patent law quick review
> List-Id:        FFII India <in-parl.ffii.org>
> 
> I just read the 1970th. Very detailed.
> 
> For a good patent law it is also useful to describe what a patent actually is,
> what it is indended for, why it is granted by state and so on.
> 
> This is not important for the patent granted under a patent law itself
> but for patent law's  further development.
> Often attorneys adopt their normative theories from the law.
> 
> 1. It has to be stressed that patent law is an incentive system, an
> instrument which a state applies to
> reach certain effects. I miss something like: Why is there a patent law at all?
> 
> 2. The so called property theory has to be rejected which causes so
> much confusion.
> 
> The problem with the 1970 act is that it says these inventions are not
> patentable rather than these issues are no inventions
> and thus not patentable.
> 
> "CHAPTER II
> INVENTIONS NOT PATENTABLE"
> 
> where 3. What are not inventions
> 
> The following are not inventions within the meaning of this Act, -
> a. an invention which is frivolous or which claims anything obvious contrary
> to well established natural laws;
> b. an invention the primary or intended use of which would be contrary to law
> or morality or injurious to public health;
> c. the mere discovery of a scientific principle or the formulation of
> an abstract
> theory;
> ....
> 
> It is somehow illogical to say an invention is a not invention. This
> can only be understood when you stress the "within the meaning..."
> So there are two subsets "inventions under Indian Patent law" and "
> other inventions"
> 
> Undefined however is what's a non-invention! The wording is dangerous
> because even when you say it is not an invention under Indian law
> you agree that it is an "invention", in invention which is not granted
> the right to get patented.
> 
> Further it is better to say that these are not inventions and
> non-patentable, because there might be "other acts" under which it
> could be patentable.
> 
> a) and b)  are better under 4), because these are exeptions for other
> reasons and systematically break the Article.
> 
> "an invention which is frivolous or which claims anything obvious
> contrary to well established natural laws;
> b. an invention the primary or intended use of which would be contrary
> to law or morality or injurious to public health;"
> 
> It is difficult for me to understand what inpact this actually has so
> I assume it is the same you find in all patent laws.
> 
> 6) "First to Invent" rule, no first to file!
> 
> "p. "person" includes the Government;"  --- ???
> 
> Person is not defined, but I believe that only natural persons may apply here.
> 
> Intresting:
> - no language regulation so far I can see
> - no specification what can be done with a patent,
> no real enforcement regulation. You get a patent ... and what then?
> - how long is it granted?
> - regulation of revocation is weak.
> - general clause 3 a/b can be used in pratice to withdraw a patent
> - are there patents "by the Government"?
> 
> Very intresting for us: The 2002 industry definition
> 
> (ac) "capable of industrial application", in relation to an invention,
> means that the invention is capable of being made or used in an
> industry.
> 
> However it remains somehow recursive. What is industry? Here the
> economic sector theory could apply. Good.
> 
> Ah, 2002 3a) was touched.
> 
> Well, from my perspective the law looks more or less chaotic. It is
> not that bad. It seemed to be subject of serious struggle.
> The systematics is somehow broken. It has to be reorganised.
> 
> _______________________________________________
> In-parl maillist
> Subscribe through the project system
> http://www.ffii.org/assoc/knecht/proj/, fine-tune via
> http://lists.ffii.org/mailman/listinfo/in-parl
> http://lists.ffii.org/mailman/listinfo/in-parl
> 
> ---
> Anivar Aravind
> 
> ------------------------------
> 
> Message: 2
> Date: Wed, 16 Feb 2005 12:09:02 +0530
> From: Anivar Aravind <anivar.aravind at gmail.com>
> Subject: [Commons-Law] "Bill Gates and other communists" article by
>         RMS
> To: bytesforall_readers at yahoogroups.com, reader-list at sarai.net,
>         commons-law at sarai.net
> Message-ID: <35f96d47050215223966875816 at mail.gmail.com>
> Content-Type: text/plain; charset="US-ASCII"
> 
> This is a Followup of  CNET interview with Bill Gates
> http://news.com.com/Gates+taking+a+seat+in+your+den/2008-1041_3-5514121.html?tag=nl
> 
> >From Slash dot:
> soloport writes "C|Net has published an article, written by RMS, in
> which Stallman points out that Gates is merely calling the kettle
> communist. Toward the end of the article, Stallman strengthens his
> point by feeding Bill his own words. Back in 1991, Bill said, in an
> internal memo: 'If people had understood how patents would be granted
> when most of today's ideas were invented and had taken out patents,
> the industry would be at a complete standstill today...A future
> start-up with no patents of its own will be forced to pay whatever
> price the giants choose to impose.' Now, if only Bill were as
> clear-minded on the subjects of Innovation and Interoperability."
> 
> Bill Gates and other communists
> ===========================
>  February 15, 2005, 3:55 AM PT
> By Richard Stallman
> 
> http://tinyurl.com/5ouyy
> 
> When CNET News.com asked Bill Gates about software patents, he shifted
> the subject to "intellectual property," blurring the issue with
> various other laws.
> 
> Then he said anyone who won't give blanket support to all these laws
> is a communist. Since I'm not a communist but I have criticized
> software patents, I got to thinking this might be aimed at me.
> 
> When someone uses the term "intellectual property," typically he's
> either confused himself, or trying to confuse you. The term is used to
> lump together copyright law, patent law and various other laws, whose
> requirements and effects are entirely different. Why is Mr. Gates
> lumping these issues together? Let's study the differences he has
> chosen to obscure.
> 
> Software developers are not up in arms against copyright law, because
> the developer of a program holds the copyright on the program; as long
> as the programmers wrote the code themselves, no one else has a
> copyright on their code. There is no danger that strangers could have
> a valid case of copyright infringement against them.
> 
> Thanks to Mr. Gates, we now know that an open Internet with protocols
> anyone can implement is communism.
> Patents are a different story. Software patents don't cover programs
> or code; they cover ideas (methods, techniques, features, algorithms,
> etc.). Developing a large program entails combining thousands of
> ideas, and even if a few of them are new, the rest needs must have
> come from other software the developer has seen. If each of these
> ideas could be patented by someone, every large program would likely
> infringe hundreds of patents. Developing a large program means laying
> oneself open to hundreds of potential lawsuits. Software patents are
> menaces to software developers, and to the users, who can also be
> sued.
> 
> A few fortunate software developers avoid most of the danger. These
> are the megacorporations, which typically have thousands of patents
> each, and cross-license with each other. This gives them an advantage
> over smaller rivals not in a position to do likewise. That's why it is
> generally the megacorporations that lobby for software patents.
> 
> Today's Microsoft is a megacorporation with thousands of patents.
> Microsoft said in court that the main competition for MS Windows is
> "Linux," meaning the free software GNU/Linux operating system. Leaked
> internal documents say that Microsoft aims to use software patents to
> stop the development of GNU/Linux.
> 
> When Mr. Gates started hyping his solution to the problem of spam, I
> suspected this was a plan to use patents to grab control of the Net.
> Sure enough, in 2004 Microsoft asked the IETF (Internet Engineering
> Task Force) to approve a mail protocol that Microsoft was trying to
> patent. The license policy for the protocol was designed to forbid
> free software entirely. No program supporting this mail protocol could
> be released as free software--not under the GNU GPL (General Public
> License), or the MPL (Mozilla Public License), or the Apache license,
> or either of the BSD licenses, or any other.
> 
> The IETF rejected Microsoft's protocol, but Microsoft said it would
> try to convince major ISPs to use it anyway. Thanks to Mr. Gates, we
> now know that an open Internet with protocols anyone can implement is
> communism; it was set up by that famous communist agent, the U.S.
> Department of Defense.
> 
> ------------------------------
> 
> _______________________________________________
> commons-law mailing list
> commons-law at sarai.net
> https://mail.sarai.net/mailman/listinfo/commons-law
> 
> End of commons-law Digest, Vol 19, Issue 19
> *******************************************
>