Make BAD Patents costly [LWN.net]

From:		"Anand Srivastava" <Anand.Srivastava@ascom.ch>
To:		letters@lwn.net
Subject:		Make BAD Patents costly
Date:		Thu, 5 Sep 2002 14:13:45 +0200

Hi,

Today, I realized what is wrong with the Patent system. There are not
enough balances. It doesn't penalize people who create bad patent. We
can't expect Patent Office to know whether a given Patent is about a real
breakthrough. So we should modify the Patent Law in such a way that
creating bad patents would be costly. Basically make it so expensive to
create non-defendable patents that it becomes profitable to challenge bad
patents in court.

If a holder loses in court not only does he/she/it loses the ability to use
it but also has to pay for the expenditure of the court case. Since Patents
are used for making money, it must be deemed that the Patent holder was
benefitting illegally, so must be told to pay an amount which depends on
the income of the holder for the duration the patent was in effect. This
must be reduced by the no. of patents the holder has. This fine must be
increased if the patent was done in bad faith. Also the patent should be
made invalid if any point in the patent is found to be invalid. This will
force applicants to make their patents as narrow as necessary.

Ofcourse there will need to be some balances to this as well. If a holder
loses on the basis of prior art then the holder should be required to only
pay the case fees, not the fines, if the holder can prove that he/she/it
had no idea that the solution already existed. Also if the holder wins then
the loser must pay the holder the case fees. This will act as a deterrent
to suing without much reason.

Also Patent lifetime must be short for software, something like 5yrs should
be fine.

-anandsr

(Log in to post comments)

Make BAD Patents costly

Posted Sep 12, 2002 8:57 UTC (Thu) by tres (guest, #352) [Link]

Although I think the idea needs some refinement I believe the thought to be sound. Another defense that you might choose to enable is the obvious defense. I don't remember if I read it in The Constitution or the United States Code (USC), but somewhere in my research, I remember reading that a patent "must be nonobvious". If you can take 10 programmers that are unfamiliar with the problem, isolate them, and ask them all to craft a solution to the problem you should be able to tell, within reason, how many ways it can be solved. If half of them come to the conclusion that is covered by the patent then I think you could call it obvious and thus invalidate the patent.

Just a Thought,
Tres

Make BAD Patents costly

Posted Sep 13, 2002 7:28 UTC (Fri) by Kikk3r (guest, #3273) [Link]

If i remember correctly the european patent treaty (EPT) requires "inventivity, novelty and applicability in industry" or soemthing to that effect. I would be suprised if the American system was very different in that respect.

Patent law experts in the audience are free to contradict me if i'm wrong.

Make BAD Patents costly

Posted Nov 8, 2002 6:12 UTC (Fri) by F6blue (guest, #7596) [Link]

Bad patents are a serious problem and only getting worse. I started a blog on suspect patents at suspectpatents.blogspot.com. If you have patent you would like me to look at e-mail me the patent.

Searches for prior art may be very cursory. There is hardly a patent attorney who is going to turn the work and in a way there shouldn't be, because most inventions can be distinguished from the prior art somehow.

-F6 Blue