Software Patents/Fran

A Discourse upon the Arguments Surrounding the Granting of Patents Covering Software

This discourse will concern itself with the issues surrounding the granting of
software patents, the dialogue will take place between two antagonists the
privateers, a group of people who are arguing for the granting of software
patents and the widerstandskämpfern, a group of people who are arguing against.
The choice of these names is bound to unmask my inherent bias, although I hope
that I am able to confine my bias to the names and present as far as possible, a
fair argument. In this discourse the term "patent" will be taken as meaning;

A patent for an invention is granted by government to the inventor, giving the
inventor the right for a limited period to stop others from making, using or
selling the invention without the permission of the inventor. When a patent is
granted, the invention becomes the property of the inventor, which - like any
other form of property or business asset - can be bought, sold, rented or hired.
[0]

Privateers

Considering the extent of the confusion surrounding the issue of software
patents, we propose that the definition of software as a program or a part of a
program that runs on a computer, and that software patents are defined as
patents that are enforceable on software. Let us also stress that we are not
talking about merely patenting single products or programs when talk about
software patents, we are talking about patenting algorithms, processes, in a
word, ideas!

Widerstandskämpfern

This sounds to us a very reasonable definition and are pleased that you have
made the distinction between patenting individual products, as is generally the
case in the pharmaceutical industry and the patenting of individual features or
portions of a software product, such as a compression algorithm [1] or a way of
distinguishing between used and unused memory [2]. We would also like to attempt
to constrain the argument to discussing economic factors as we find the
discussion of the "moral" or "ethical" factors to be too full of the kind of
emotional rhetoric that one expects of politicians and other such filibusters,
and after all, governments have been known to ignore patents when progress or
safety has been in question! [3]

Privateers

A fair point indeed, but surely you must agree that inventors have the right to
live off their successful inventions! Take for example the case of the
individual inventor or small company that invents a new and brilliant design,
without patents they would be mercilessly crushed by the competition of the
market, a bigger company would come along and with the benefit of economies of
scale, be able to copy their design with impunity and produce it at lower cost,
quite likely removing any chance the individual or company had of profiting from
their idea.

Widerstandskämpfern

An argument like this is held up in every discussion of this nature, we were
hoping to avoid it, but we will provide a brief response as it is evident that
you won't let it lie. This myth of the "small inventor benefiting from patents"
is a modern one, and quite insidious. The fact is that the entities that benefit
most from patents, and software patents in particular, are those entities that
are large enough to build up patent portfolios. When a big company wants to use
ideas that have been patented by a smaller company, normally, instead of
licensing the patent, it just threatens to take the smaller company to court for
infringing one or more of the patents in its patent portfolio. The smaller
company cannot usually afford the cost of going to court and so signs a deal to
the effect that the bigger company can use its patented idea if the bigger
company agrees not to take them to court.Thus, the current patent system works
against the individual inventor, not for him! This is quite apart from the cost
involved in making a patent.

Privateers

Simply not true! Small companies win patent claims against big companies all the
time, we cite the case of Palm vs. NCR [4], Microtune vs. Broadcom [5],
InterTrust vs. Microsoft [6] and Sendo vs. Microsoft [7].

Widerstandskämpfern

It's interesting that you cite these particular cases as it proves our point
quite nicely. First of all, none of the plaintifs you cite could really be
classed as "small" companies, smaller maybe, particularly when talking in
relation to Microsoft, but certainly not small. Secondly, all of those companies
already have patent portfolios, none of them is trying to fight with just one
patent, and thirdly not all of those cases are resolved, which goes to show how
drawn out and lengthy the legal process is when talking about patents, and we
haven't even arrived at the issue of cross licensing yet!

Privateers

Fine then, but firstly we demand that you define your terms, patent portfolio
and cross licensing, after all, one can't conduct a proper discussion without
defining ones terms.

Widerstandskämpfern

Certainly, we will gladly agree to that request! First of all, a patent
portfolio is a collection of patents filed by a company in order to use them
strategically against their competitors. In practice, the more patents a company
has in their portfolio, the less effect the patents of others have on them. So,
as we explained earlier, the big company can always threaten to take the small
company to court for infringing many of its patents, something the truely small
company can rarely, if ever afford. Cross licensing is where one company
licenses some or all of its patents for use by another company in return for a
similar license, this is actually what came to pass in the InterTrust case you
cited earlier. We trust with this explanation that we can get off the "moral"
and "ethical" issues and get on with discussing the damage to innovation and
progress that software patents are causing and will cause if introduced in a
global scale.

Privateers

We will assent, but don't for a minute think we have conceded our point. Well
then, let us make first the point that software patents are vital to protect
investment in innovation in the software market, and the fact that without
patents companies will have no incentive to innovate as they cannot be sure of a
return on investment. Surely this will slow progress, why bother to develop
something if you can't be sure make money from it?

Widerstandskämpfern

This is a common view point, and initially seems to be obvious. However an
economic study by members of the Federal Reserve Bank and the Boston University
School of Law [8] shows that instead of encouraging research and development,
software patents actually discourage it! The study goes into quite some detail,
but we will provide some of our own explanations as to why software patents
hinder innovation. Software cannot be thought of as just one part, as we have
already made clear. An innovative piece of software can be one that is not just
a produce featuring a new idea, but a product that is a new combination of old
ideas. This kind of innovation is particularly badly effected by software
patents, as developers, and especially small developers can never be completely
sure that one of these ideas that they are using, which could seem quite obvious
to them has not been patented. The patent system is notoriously difficult to
navigate for numerous reasons, among them, not all patents are available for
public viewing, some patents pending are kept secret, there are so many patents
over 100,000 [9] and no way of accurately or centrally searching them,
furthermore if one does manage to find a relevant patent it is likely to be
completely incomprehensible to the average software engineer due to the
obstructionist legal language it is phrased in. Considering the low
time-to-market of software products, in general around 18 months, compared to
pharmaceuticals, where a new drug can take 10 years or more to develop [10], it
is clear to see that the chances of finding a patent violation are much lower in
the software market compared to the pharmaceutical market.

Privateers

Very interesting, but all of the arguments you give are not strictly arguments
against the idea of software patents in themselves, but in the implementation of
the patent system. You give no reason for thinking that if there was to be put
in place an accurate, fast and complete way of searching and indexing all
patents and if they were worded in a more clear manner that the patent system
would be a bad thing. In short, you are arguing against the system, not the
theory.

Widerstandskämpfern

Point taken, consider then if you will the length of patents and how that would
effect innovation in the software market. In the UK and the USA, in line with
the TRIPS agreement [11], the term of a patent is 20 years. Now, 20 years is not
a long time when your product development lifecycle is 10 years, but when your
new product turn over is as fast as it is in the software market it becomes
quite ridiculous! Can you seriously imagine having to develop new software
today, using only ideas that were in existence in 1985?

Privateers

Again, this is a problem with the implementation of patents with regard to
software, not the theory behind it! What if the term of a patent was lowered for
software?

Widerstandskämpfern

Enough! It is impossible to separate the two so finely, not all patents that
happen to cover software are specifically written as "software patents" [12],
many are generalised to cover any implementation of an idea, whether it be in
hardware or software, and so how could it be said that a patent term is so long
for this product, but so long for that product. It doesn't make sense and it is
unworkable. To move on, how does the existence of companies such as SCO fit into
the framework of "patents promoting innovation"? Here is a company that has
gone from being a company that innovates and produces new software to one that
is now attempting to survive almost entirely on litigation and patent lawsuits
against truely innovative software. [13]

Privateers

SCO is an abberation, although the figures aren't as bad as you suggest, one
third of SCOs income comes from licensing [14] and indeed, some companies are
allowing this very same "truely innovative software" to use their patents at no
cost [15]. You can't make an argument against patents on the basis of the
behaviour of one corporation.

Widerstandskämpfern

Granted, it was merely an example. Then back to the theoretical argument, the
fact remains that it is impossible to have innovation in software if all new
ideas are subject to restrictions, can you imagine what the state of desktop
environments would be like if a patent was granted for each and every new idea
that came into them? You might as well say goodbye to the idea of a desktop
environment, not to mention interoperability.

Privateers

But surely interoperability is in the interest of companies, how else could they
persuade customers away from the product they are already using to their
product?

Widerstandskämpfern

If the market was competitive that would be true, but the fact is that with
software patents the market cannot be competitive as they irrevocably tilt the
field in favour of the large companies, if a small developer wants to put out a
new product with an innovative feature that competes with a product from a
larger company already on the market, he must necessarily surrender this new
innovative feature to the large companies as they will undoubtably have
"infringed" on one of the patents in the larger companies patent portfolio. It
is simply not practicable for a small company to or individual developer to
build up a patent portfolio of their own due to the extortionate costs involved
as we have already explained!

Privateers

Well, we can't say that your arguments have entirely convinced us, after all,
the USA has been granting patents on software since the 1960s and there seems to
be no slow down of innovation there.

Widerstandskämpfern

That generalisation is invalid, you can't reduce a very complex subject into a
simple statement, sure the USA might appear to be suffering no slow down of
innovation, but if you had taken the time to read the study we cited earlier
[9], you would find that there are a great deal of people in the USA who have
found that the increased enforcement of software patents is damaging innovation.
They found that most software patents are owned by large companies who don't
use them to prevent the imitation of products but for a strategic purpose as we
described earlier. They also found that software patents have by and large
resulted in a transfer of resources from research and development to pursuing
patent infringements. It was also commented that in fields where innovation is
incremental, like in software and unlike in pharmaceuticals the granting of
patents instead of encouraging innovation hinder it. In conclusion our position
remains; Abolish the granting of software patents where they exist and fight
tooth and nail to prevent their introduction anywhere else.

Bibliography

0. http://www.patent.gov.uk/patent/whatis/definition.htm
1. http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=/netahtml/srchnum.htm&r=1&f=G&l=50&s1=4,558,302.WKU.&OS=PN/4,558,302&RS=PN/4,558,302
2. http://swpat.ffii.org/patents/samples/ep359815/ep0359815.txt
3. King v. Irving Air Chute Co. Inc (1949) (http://www.gowlings.com/resources/publications.asp?pubid=782)
4. http://news.com.com/PalmOne+wins+patent-infringement+case/2110-1047_3-5515731.html
5. http://dallas.bizjournals.com/dallas/stories/2003/03/17/daily51.html?page=1
6. http://www.microsoft.com/presspass/press/2004/apr04/04-12MSInterTrustPR.asp
7. http://www.microsoft.com/presspass/press/2004/sep04/09-13sendopr.asp
8. http://www.researchoninnovation.org/swpat.pdf
9. http://www.patent.gov.uk/media/keyfacts/filingstats.htm
10. http://www.cptech.org/ip/health/econ/howmuch.html
11. http://www.wto.org/english/tratop_e/trips_e/trips_e.htm
12. http://en.wikipedia.org/wiki/Doctrine_of_equivalents
13. http://ir.sco.com/ReleaseDetail.cfm?ReleaseID=115527
14. http://news.com.com/2100-1014_3-1010607.html
15. http://news.bbc.co.uk/2/hi/technology/4163975.stm

Software patents – Obstacles to software development
http://www.cl.cam.ac.uk/~mgk25/stallman-patents.html

Software patents on the Wikipedia
http://en.wikipedia.org/wiki/Software_patent