[Patents] French State secretary for Industry against SoftPat

[mva@df.lth.se (Martin Wahlén)]

On Mon, 2 Apr 2001, Gregory Aharonian wrote:

> >I do not think that a good prior art search is the real problem, because
> >as I explained you before, if an invention is really worth of patent
> >protection, the probability of being duplicated or reinvented is small.
>
> I have long stated that for a fee of $5000 per software patent, one
> can do a search that invalidates 80% of the issued software patents.
> Now patent fees around the world are insufficient to fund patent
> examiners to that level of support - in fact, of current patent fees,
> I suspect the equivalent of about $500 is provided for searching,
> woefully inadequate.

I would say that it takes about a day to translate the "patentees" to
scientific language and then about a day to do the search in the
scientific databases (and read through and atleast partialy grasp what is
in there). And then we haven't even looked at prior arts searches with
timestamped software, etc.

I also suspect that people would be more likely not to apply for a patent
that is likely to not pass the examiniation process if the fees are higher
(the inventors would search more thouroghly).

> Now in quantity across tens of thousands of patent applications, one
> could bring the average cost for a good search down to about $1000
> per patent, which is much more reasonable to be provided, either
> through slightly higher fees or direct government subsidies.

I don't think you can bring it down that far, and I do think that we need
government subsidies for this to happen. Possibly one could do
differentiation punishing only companies that apply for lots and lots of
patents, leaving small inventors alone.

This is not to say that I think that we should have Software Patents at
all, but just continueing on Gregs thoughts.

Could we get the patent offices to ask inventors what they would be
willing to pay for a patent application to be processed? Also I would like
to see some statistics on how many applications are turned down in
different fields of industry. If anyone knows of a good paper on this,
please hand me the URL.

Martin