[ILUG] EU Rejects Patent Software Law - For Now

[Paul Jakma]

On Fri, 18 Feb 2005, Michael Conry wrote:

> design presented to you.  The fact that you are "imagining" you have
> no imagination makes it all a bit strange.

;)

> at their jobs, it's always going to be very difficult to judge the 
> obviousness criterion.

Yep.

> Something that became very stark for me after this conversation was 
> the immense fallibility of the system, and the implications this 
> has for patent holders.  Just because you are awarded a patent does 
> not guarantee it is going to stand up if challenged.  If you have a 
> very compromised patent system (as many claim the US system is) 
> that adopts a policy of "if in doubt, issue the patent" and then 
> waits for the courts to sort it out, you end up with a very 
> devalued currency (many patents may be open to challenge).

Yep.

There possibly is an obvious fix. Change the system to *not* examine 
patent applications (least not for the difficult stuff, eg prior art 
searches, obviousness test) - just make the 'grant' a rubber stamping 
that the application's documentation is in order. This should allow 
cost of application to be lowered (preparation of the application 
will still be as expensive, but the patent-offices fee should be 
significantly lower). Then examine patents properly only when they 
are disputed, with recourse to the courts if needs be.

A lot of patents are stupid, most never see litigation. Advantage is 
you dont have to waste precious examiner resources on examining all 
applications, you can concentrate them on examining only patents 
which someone actually cares about - hence, that examination can be 
more thorough (on which the case for such reform rests).

Ie, formalise the situation we have now, where the real test of a 
patent (and only a tiny subset of patents) is /after/ application, 
when a patent is disputed - except instead of using the courts to 
examine validity, use (presumably cheaper) patent office examination, 
with courts as a last resort (and possibly some kind of weighting 
against the party which 'lost' during patent office examination, if 
they chose to go to court).

Soft pats are evil simply because the patent system is so broken (and 
9 times out of 10 software doesn't require huge resources to 
develop[1], a fairly key assumption for the patent system). If the 
patent system were reformed to *not* be broken, software patents 
might indeed be reasonable, but till then..

> Michael

1. Software can be 'innovated' with no more resources than a year or 
two of time, possibly spare time, possibly not even requiring a 
computer, just pen and paper. The patent system's protections hark to 
times where recouping costs of 'innovation' required massive capital 
investment. Lumping software with same patent protections does not 
recognise this.

regards,
-- 
Paul Jakma	paul at clubi.ie	paul at jakma.org	Key ID: 64A2FF6A
Fortune:
"Mr. Spock succumbs to a powerful mating urge and nearly kills Captain Kirk."
-- TV Guide, describing the Star Trek episode _Amok_Time_