[Sheflug] Software Patents and small businesses

[Bob Holland <bobholland [at] xxxxxxxxxxxx>]

Some of you will no doubt have received emails regarding EU patent changes 
for software products.

I note that Club UK Online has a workshop in Hull next week. In view of the 
EU vote regarding software patents (now shceduled for September) I am mindful 
to go along & would like to try to get the concensus of opinion of this 
group, particularly in terms of what is best from the point of view of the 
South Yorkshire economy, its citizens, local communities and small businesses.

Are software patents a good thing?

The general argument for patents is that they help protect investment in 
innovation. My own feeling is that in a global marketplace patents in general 
protect large company interests to the exclusion of small companies and 
thereby stifle true innovation. Of course, patented work is 'published' 
(don't believe anyone if they say to you 'this is a secret patented process' 
- something I had to point out recently to the authors of a Which? report). 
The counter argument to this is that patents can be used to protect 
the work done by small companies. However, since it is the inventor's 
responsibility to...

(a) identify any pertinent 'prior art';
(b) produce the patent application in the appropriate 'patent speak';
(c) defend the patent in the courts against infringement;
(d) pay renewal fees;

...very deep pockets are required. I guess we are talking today about a 
budget of somewhat more than £30k plus renewal fees to obtain patent 
protection in the major industrialised territories. In a global markeplace 
there would appear to be little point going for a UK patent. There are not 
too many small companies I know who can do this.

When it comes to (c) and a big(ger) player is involved, they will probably 
sell out rather than defend their independence through the courts - sounds 
like a recipe for anti-competitive practice to me.

An example from my own background...

Back in 1979, working for a small local company I designed a 'programmable 
controller' for use in the baking industry. This incorporated what would be 
known today as an 'integrated development environment'. This system was sold 
to a number of clients and incorporated in a range of bakery process 
machinery. A high speed dough mixer incorporating the system was launched in 
Las Vegas in 1980 and was widely used during the 1980s.

A year or so ago I was reviewing my CV and noted how commonplace one of the 
techniques I had used in this product had become. Something 
prompted me to do a patent search using a couple of key words. I very quickly 
located a patent awarded to a large US company (familiar to all but not an 
obvious player today in the computer game). The patent was clearly important 
(it had been filed in many parts of the world) and had been licensed to the 
likes of Borland, Microsoft et al. The date on the patent was 1983. Whilst I 
have no wish to claim 'invention' of the technique (as others unknown to me 
had probably beaten me to it), I can probably prove that the patent is 
invalid - but heck, I am on my own and would be fighting against a large US 
corporation and the patent has (almost) expired anyway.

There are several points raised by this...

1) It is almost impossible to identify relevant prior art - the US patent 
department seems particularly bad at doing this (another story) and 
frequently rely solely on the statements of the applicant;
2) The technique I developed was merely a side issue to the main purpose of 
the product and was therefore not considered especially significant (by me);
3) Software patents were not allowed in the UK anyway, so the technique would 
not have been identified as patentable.

This experience seems to suggest that small companies should be more 'patent 
aware' and that patent protection should be allowed for software. It also 
suggests that it would be a good idea for 'inventors' to retain an 
experienced patent agent who would review all developments both for 
patentable content and potential infringement. Having worked with a number of 
patent agents over the years it seems inventors are particularly bad at 
identifying 'patentable' aspects of their work.

Given that a patent system already exists and is likely to be extended to 
software then I would tend to agree that a good patent agent becomes a 
necessity - good news for patent agents but an impossibly large cost for an 
individual and prolific inventor. Together with (1) above, it is why I 
believe that the existing patent system discourages real innovation in small 
companies and why further extension should be resisted.

Given the worldwide explosion in technology since the concepts of patents 
originated, is it not time for the world to have a complete rethink?

Open source's concept as the equivalent of publishing scientific papers runs 
somehwat parallel to the patent system. Why should individuals not be free to 
publish their work if they wish? Providing the 'prior art' is researched 
correctly, patents will not be granted for work already in the public domain 
and open source will have achieved its aims. The difficulty for innovation is 
that individuals may be discouraged from coming up with really new ideas in 
the public domain in case they are in infringement of a patent somewhere.

Many 'patents' which have demonstrated the properties of 'industrial 
application' and 'an inventive step' will of course have been built on 
published research. The publishing process encourages inventiveness and 
rewards inventors through salaries, research grants and the 'kudos' of  
having their work recognised in peer-reviewed journals.

The patent system rewards those in turn who may incur significant up-front 
costs and commercial risks to turn 'ideas' into the real products which 
provide benefit for humanity. This is often forgotten in the discussion on 
software but is very real in other areas of technology.

Does extending the scope of patents in reality have any effect on this?
 
I view the open-source approach as essential for innovation in my own work - 
it enables me to 'stand on the shoulders of giants' as it were and to earn a 
reasonable living whilst in return acknowledging the contributions of others.

Perhaps government-funded open-source projects operating in the same way as 
other government-funded research will enable authors to be adequately 
rewarded for their work. Existing funding agencies don't appear to understand 
this concept.

Any views on this? Perhaps an 'Open (source) University' should be 
established and we can all go back to wearing kipper ties? (Hey - I bags the 
copyright on that one!).

Must go for now,

Bob Holland


 




___________________________________________________________________

Sheffield Linux User's Group -
http://www.sheflug.co.uk/mailfaq.html

  GNU the choice of a complete generation.