Please explain how this twisting works in practice. I just don't
see the point.
Many programmers are forbidden by their employers upon
advice from counsel from reading patents for this very reason.
This is
not an argument about the programmer being productive. This is an argument that
the costs/benefits analysis doesn't work out for the programmer because even if
he is more productive he is still not productive enough to overcome the costs
side of the analysis. The liability aspect cannot be dismissed because it is
part of the over all analysis.
This is an argument that the USPTO should
do their homework and check whether patent law works in practice according to
theory. This is not an argument that case law must be overturned, although the
analysis may eventually lead to this conclusion.
Please remember, the
issue this response raises is whether patent law actually promote innovation
when it comes to software. It is not whether case law is correct. We are not
arguing in front of a court. We are giving the USPTO feed-back on how things
play out in real life as opposed to how they should play out in theory.
Using the word "twisting" might be a mistake from me. Let me
try to restate the same thoughts with other words.
The thing I am after
is that patent lawyers/judges start with the assumption "reading patents improve
productivity unless proven otherwise". We can argue about how true this is in
the real world, but they would not be patent lawyers if they did not have this
belief.
Arguments about costs/benefits is tricky to state since the
lawyer with an assumption about patents on software being a good thing will view
the costs/benefits argument to mean "many companies prevent their employee from
benefiting from reading patents because they know the increased productivity
will be insignificant to the treble damage caused by the companies deliberate
stealing of inventions from the patent holder." The cost/benefits argument is
not conclusive to a lawyer unless you can make them realize the actual value in
software patents is zero. The software programmer knows this, but our argument
can not use this as an assumption because it is not shared by the one we are
trying to convince.
The problem with lawyers reasoning is thus the
assumption about there being value in patents themselves. That is why I think
the treble damages argument is counter productive to our cause.
The true
problem with software patents is that programming is math/language, but if we
speak about about the practical reasons why patents without full disclosure is
useless then the true answer is that the problem is not that there is no value
in the formal language without disclosure of sourcecode, but that the computer
science field have massive and better information that include both the formal
language and enough disclosure. The difference between programming and other
fields is that since programming is math and nothing else it is possible for the
computer scientist to cover it all, where you with other fields always end in
pesky details about some aspect not being abstract but a real thing that has
aspects not covered by the math/model.
The argument about FLOSS is
likewise inconclusive. It is true that FLOSS offer superior disclosure, but the
lawyer will go with the fiction that software patents is needed because not
every software is FLOSS or can be expected to be FLOSS. If the argument is
changed more to that the field of computer science already contain every value
there is in formal language specifications and disclosure of working programming
code is the only way to actually add useful information to one skilled in the
art then the arguments get much more useful.
I think an argument to make
in final part is that skilled in the art when you consider computer programming
mean that you know all the math needed so that formal language specification is
trivial in each and every case. The only difficulty is if the domain expert
insist on demands that are contradictive.
Being skilled in the art does
not mean that you have the skills to implement every problem you can state with
formal language, the field of computer programming is too wide for anyone to
master all areas. This is why FLOSS that insist complete disclosure even of
things that is directly applied from the computer science books is such a
superior development model. A group of programmers from different backgrounds
write better code together than any single company can ever succeed with.
Maybe it might be good to add an argument about disassembling of source
code meaning there is no trade secrets in computer programming. This is cince if
you can run the code you are by definition able to disassembling it to learn how
it works. Patents in general is in theory a balance between being giving a
monopoly and revealing your trade secret. If there is no trade secret there is
nothing to balance the benefit. This especially hold true if you don't have to
do code disclosure to prove that you really know how implement your formal
language specification. [ Reply to This | Parent | # ]
|