D&C GLug - Home Page

[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]

Re: [LUG] If this wasn't so dangerous, you'd laugh.

 

On Friday 11 February 2005 1:01 am, Jon Lawrence wrote:
On Thursday 10 February 2005 17:36, Simon Waters wrote:
Jon Lawrence wrote:
| It would be interesting to see what would happen if someone with
enough money
| made a case against them.
What for?
There is nothing wrong with having a patent for something that was
invented before, you just can't enforce it.
If you tried to enforce it, you might open yourself up to being sued for
bringing pointless legals actions, but if you never try and enforce it
you just have an expensive piece of paper.

Hmmm, I thought it was fraud to
Following that logic, it doesn't matter what a patent says. If there's
prior art, then the patent is unenforceable. If this is the case, what are
we all so bothered about.

Prior art does always disqualify a patent, but there is more to this:

1. A patent can only be disqualified by a court. If you win in a (cheap) 
county court, the litigator can appeal and push it to a higher (expensive) 
court and then again to the House of Lords (in UK) or Strasbourg for the EU. 
Now the costs get extremely serious - but the patent is STILL valid until the 
final appeal is quashed. Funds may be available to do this once or twice, but 
each decision is separate and every single patent would have to be defeated 
in court. There are already thousands of software patents.

2. The courts don't see prior art or patents as we see them. There is good 
reason to believe the Patent Office see software and a technical invention 
expressed in software as two separate things. The technical contribution 
makes software into an invention, not "just" software. The software contains 
an expression of the underlying patentable idea and as such, any software 
that expresses the same idea is infringing that patent.

2a: Example: In code, if you are to compare two objects, you can sum or hash 
their contents and compare the sum or hash. We do this all the time with 
checksums - however the two objects remain the same type. In law, when they 
compare two pieces of software, if one contains a technical contribution, 
they change the *nature* of the object!

Essentially, they allocate a higher priority to 'technical contribution' than 
to the method of expression, i.e. the software. Software is not meant to *be* 
patentable in the EU - the patent laws were specifically written to exclude 
software - yet the UKPO and the EPO have interpreted UK case law to twist 
that definition. They say, in effect, that once software expresses a 
technical contribution, it ceases to BE "just" software and so cannot be 
excluded from what is deemed patentable.

(UKPO: UK Patent Office: EPO European ..)

Their main weapon in this is summarised as:
Imagine company A, call it ACME, that has a superb idea which they want to 
patent. This is a technical idea, it is tested as a proof of concept on a 
printed circuit board in-house. They claim the patent on the idea and 
describe it's implementation in hardware.
Now imagine developer B, who has a superb idea for enhancing his own software 
and expresses that in, say, Java. He releases the software and gets lots of 
praise and acclaim.
Company A thinks that developer B has expressed their idea in software - i.e. 
they allege that the program is a software version of the same conversion or 
processing that they put onto their circuit board. Note that company A never 
had to sell a single circuit board to get a valid patent.

Who wins?

The argument comes down to the nature of the object. If you believe developer 
B, software is not patentable and this overrides any expression of the 
technical contribution in his code. In effect, you are rating 'software' 
higher than 'technical contribution'. The Patent Office and the law courts 
currently disagree with you - 'technical contribution' has a higher priority 
in law than 'software'. Therefore the case by company A is proved because 
even though pure software is not patentable, it expressed a patented idea. 
Developer B is guilty of patent infringement and company A is granted 
punitive damages.

From our perspective, the courts have taken an orange and changed it into a 
lemon.

Note: this is the CURRENT position, without the controversial Directive. It 
has been the current position since the Fujitsu case (below). The status quo 
is NOT acceptable.

Who gives a monkey's if someone wants to patent something that's already
been done - if they can't enforce the patent then as you say, it's just an
expensive piece of paper.

That takes a small fortune to prove and if you simply argue that it is 
software, you WILL lose.

If it's a new invention then they can patent it 
and extract license fees or whatever for using that invention - fair
enough, they invented it - whether or not software code can be classed as
an invention is another matter.

In UK law, software has been classed as an invention - it was done when 
Fujitsu appealed a patent infringement in 1997, I'm trying to get the full 
reference. The Lord Justice at the time has been quoted as saying that 
anything technical is patentable, no matter how the idea is expressed.

The current directive originally tried to 'clarify' the resulting legal 
position by encasing the situation after this decision in law - that is what 
the fight is all about.

-- 

Neil Williams
=============
http://www.dcglug.org.uk/
http://www.nosoftwarepatents.com/
http://sourceforge.net/projects/isbnsearch/
http://www.neil.williamsleesmill.me.uk/
http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3

Attachment: pgp00026.pgp
Description: PGP signature