[ILUG] Enda Kenny and Patents

[Paul Jakma]

On Sun, 11 Jul 2004, kevin lyda wrote:

> the concept of a patent is to give inventors a short-term monopoly to
> recoup their research and development efforts.

Right, and why shouldn't that apply to software "innovations" too?

> none of those are patented.

Which says nothing of the argument against software patents.

> therefore, the most compelling argument against software patents is 
> that they are not required.

not really.

> a small number of companies would like them in order to dampen 
> competition

But again, that is not software specific. As I learned from Discovery 
channel, Trevithick ran into patent problems with his improvements to 
the steam engine, where he greatly raised the efficiency by using 
high-pressure boilers. Watt and Newcomen held key patents (on the 
condensor?) which meant Trevithick had to wait x years for Watt's 
patents to expire before being able to develop his ideas - though he 
did manage to find a way to bypass Watt's patented ideas (but not 
before the patent had expired anyway).

Preventing competition is fundamental to the concept of patents, 
hence that property of itself is not an argument against them.

> - after all, that's what a monopoly is designed to do.  however 
> history has shown that in the software industry, that monopoly is 
> not required for steady pace of invention and innovation.

But the innovations you have cited were all developed in the public 
sector and hence not subject to competitive pressure anyway, or not 
exactly patentable (word processor would be hard to patent) or 
actually *were* patented (ie the spreadsheet, see 
http://en.wikipedia.org/wiki/Spreadsheet, indeed that helped set 
precedent for software patents).

> patents are a useful tool in other industries.

Why wouldnt they be useful in the software industry then?

> and the software industry has already benefitted greatly for other 
> areas of government - from the arpa funding that led to the 
> internet, to government research grants and all the way to 
> copyright laws.  that government involvement was useful, beneficial 
> and even handed - from sole developers all the way to major 
> corporations.

That's an argument for state funding of innovation in software. Never 
going to fly in the 1st world :). The whole point of patents is to 
make investment in R&D attractive to the /private/ sector, and not 
have to have state funding.

Universities, especially in US, make a lot of money by way of the 
patent system, either by direct patent licensing or by spinning off 
successful research into commercial entities or by acting as "R&D for 
hire" for private sector interests (who then reap the rewards of 
patent licencing if the R&D is successful).

> software patents cross the line from beneficial to harmful.  from 
> fair to biased.  from policies that enlarge the industry to a 
> policy that would contract it.

I dont really see why software patents are different. In my opinion 
there is a general problem with the patent system:

- too easy to get a patent (if you have the money)

and/or

- too difficult to knock down patents (even if you have the money)

one or both should be corrected. Patents should either be examined 
far far more vigourously, or, given that that costs money and hence 
is not politically attractive, should be far easier to knock down. 
Eg, there should be an easy process to have patents reexamined and 
struck out, with presumption of patentability being negative.

This is by *far* the biggest problem with patents, especially 
software patents, in my eyes. It seems that you can get patents for 
things that would be obvious to any one with half-clue. Eradicate 
this aspect, or at least make it trivial to knock such patents down 
and patents (software or not) would be a lot more palettable.

- Patent grant periods are too long for the modern, fast-to-market 
economy, and are applied uniformly across all industries, regardless 
of differences in development->market times in differing industries.

Maybe back in the 1800s 17 to 25 years was a reasonable time to 
develop, produce and market a product and hence recoup the costs. 
It's not any more, even pharmaceutical products (with probably some 
of the longest R&D times and regulatory approval times in modern 
industry) probably dont need 25 years to recoup cost and earn money. 
25 years certainly is too long for software, The grant period should 
be *much* shorter and/or possibly commensurate to the costs of the 
industry concerned, eg no more than 5 years for software, no more 
than 10 years for patents in general or whatever.

I dont know, but it seems to me that arguing that software patents 
are somehow different to patents in other fields is a flawed 
position. One ends up using arguments against software patents that 
actually apply to the patent system in general, while maintaining 
that those arguments show that software patents are somehow different 
- which seems an intrinsically flawed argument to me. The logical 
argument would be to use the problems highlighted by patents as 
applied to software to demonstrate the desperate need for patent 
reform in general.

PS: you mentioned that the software industry already has copyright 
anyway. However, in industrial design, one can also apply for design 
protection, which would be somewhat analogous.

> kevin

regards,
-- 
Paul Jakma	paul at clubi.ie	paul at jakma.org	Key ID: 64A2FF6A
 	warning: do not ever send email to spam at dishone.st
Fortune:
Emacs, n.:
 	A slow-moving parody of a text editor.