Re: [Fsfe-uk] Fwd: UK Leading Patents?

[Xavier Drudis Ferran]

Alex Hudson wrote:
> 
> On Sat, 2002-03-09 at 18:02, Xavier Drudis Ferran wrote:
> > Excuse me for not following this thread as well as I should. I'm not at
> > home, and I only have a 486 with a 33K modem a broken hardware clock
> 
> I used to know that feeling :-)
>
The hardware clock seems not so bad after all,  I don't understand 
what is making old messages appear as new. Anyway,  that's offtopic. 
Just remember I may have missed some post. Sorry about it. 

> > Please read the directive in full. And then try to find an example of
> > software which would not be patentable under that directive.
> 
> The draft EPO directive? I'm not really concerned with that; we're
> talking about the UKPTO. As I've said before, I know the EPO has
> problems.
>
The CEC directive proposal ("COM(2002)92 final", IIRC). I've almost
given
up trying 
to criticize the UKPTO "domestic affairs". Not because I  think they're
fair, but because 
I think I don't understand UK enough to give an opinion, so I let that 
to UK citizens,  after all it's none of my bussiness. But I can pray 
you 
to try to convince UK government (UKPTO or whoever) not to accept 
the CEC directive, since that would have an impact on me too. 

> 
> Okay, let's take your case of the missing FET. We assume that the FET
> isn't invented, and that the current emulators don't know about FETs.
> And let's say someone patented the FET. Okay, a circuit with a FET
> cannot be simulated. But, what happens if I rewrite the simulator to
> include FET simulation? Is that a patent infringement? If so, that means

If FET is patented and then you write a simulator for a
FET,
you are not infringing. You are publicizing, disseminating a patented
invention,
and that is the whole purpose of the patent system. You should be given 
a medal by UKPTO and another by EPO. You are contributing to the
technical 
knowledge the patent system want to foster and put in the open instead
of 
keeping it as industrial secret. 
It is exactly the same as when you write a book on a patented invention. 
Of course the fact that the book is written after the patent is not an 
infrigement, even if that allows you to reproduce the output function of
a FET with pen and paper. 

> that you allow software patents - the patent holder would be able to
> write the simulator without any problems. So, is it not an infringement?
> If so, that allows you to by-pass the discovery of FETs by emulating
> circuits which make use of it (sounds an odd thing to do; but people
> already do this - for other reason - see Roland VGuitar, etc.)
>

Emulating a FET is not like manufacturing and selling FETs. The patent 
only prohibits commercialization of the patented invention. You can 
commercialize knowledge about the invention, equipment useful with 
the invention, etc. You can also build the invention itself as long as 
you don't commercialize it (for education or research, e.g.).

If you program/buy/download a FET simulator and run it on a BJT
computer, you are using 
BJT transistors (in a funny complex way) to emulate the output of a FET. 
That is a legal circumvention of the patent. Patents do not forbid 
circumventing with other inventions that achieve the same goal.  
I've  even heard people say that because patents encourage
circumventing, 
they encourage research and progress. I  don't buy that much, because 
after all I don't want _that_ research and development, I don't want
people 
to reinvent wheels all day long, I want them to use wheels to invent
the segway (that strange electrical scooter).

I don't know about guitars, but I pressume is the same case, a
circumvention 
of a patented technique with unpatentable  logics and possibly
patentable
output sound devices. If you invented a sound card with a circuit that 
build a sound wave from digital  input,something in that sound card
might 
teach us something on sound waves and therefore be patentable. If you
get a
patent then I'll be possibly paying you toyalties when I buy a
soundcard,
but then I should be able to use it for whatever I want with my 
software, not pay royalties again to the guitar patentee.

I'm supposing there is a guitar with some device that produces some 
efects and that device is patentable, andthere is a program that uses 
a sound card to produce soundeffect undistinguishable  from the guitar 
device effects, at least to the naked ear. If that is not what you
meant,
sorry.

> > When you (was it you?) says that software that can circumvent a device
> > should be patentable  (please correct me if that  is not what was meant),
> > then all software becomes patentable. All software can be implemented
> > in hardware or mechanics, or whatever.
> 
> We kind of agree, then. I didn't say that software should be patentable,
> though; I highlighted it as a problem with the argument. I suggested
> that if you do away completely with "technical effect" (or whatever
> variation), you essentially destroy a raft of other non-related patents,
> electronics, etc.
> 
Errr... I dont' know. I think "technical effect" is meaningless, so by 
removing it you would change nothing, and everything would be
patentable, 
just like when you use "technical effect". I believe any human discovery 
may be construed to have some technical effect, specially if you
consider
changes in the state of a general computer as technical effects. If I
download
"Pride and Prejudice", that changes magnetic fields in my hard disc, and 
surely electromagnetism is technical, so 
Jane Austin (or her heirs or buyers of her patents) should be paid
royalties
by anyone who sells novels on love between strong characters,if you only
look 
at technical effects. 

So we don't agree, do we?. :(

What you need to keep electronics, mechanics whatever patentable but let
software,
logic, bussiness rules, literature,etc. not patentable is criteria that 
studies where is the new contribution of the patent, not its effects. 
And the contribution should be technical, so "technical contribution" is
not
bad aas such ;op. The problem is when you don't define "technical" as in 
DispositionnenProgram (sorry, one day I'll learn  to spell it), but you
leave it open or, horror, explicitly put software (or anyother logical
achievement)
in the technical fields.  Then all contributions become technical, and 
everything becomes patentable. 

not 
> Personally, I don't believe software should be patentable, at all, under
> any circumstances. However, this is nothing to do with patents only
> applying to 'technical' inventions - harnessing nature, whatever. The
> costs of software research are vastly smaller than other areas (such as
> electronics), and thus doesn't warrant patent protection as far as I'm
> concerned. I also find the argument about what is "technical" and what
> is not to be confusing, contradictory and ultimately unsupportable.
>
I don't know any other criteria that separates what I want patentable 
and what I don't. And I don't want to leave that open when there is at 
least a criteria we can use. Giving an explicit list of what is
patentable 
and what not  was tried in the EPC and has been abused. Besides, it is 
much more difficult to work out such a list and avoid contradictions. 
I find the technicity criteria easier. 

 What do you  find contradictory exactly?. Confusing?. Unsupportable?.

> Patenting, like copyrighting, should not be a right, but a gift. We
> allow copyrighting to benefit society as a whole; patents should work in
> the same way. Patenting software does little if anything to benefit
> society, and (I think all of us would argue) actually does damage.
> 
I agree. A patent is a deal between the state and the inventor, not a
natural 
right of the inventor. We must make sure the deal is profitable for
society.
And that is all about economic considerations (and also basic  rights
losts 
due to the patent). 

But I don't think you can tell the patent office to  make an economic
assessment 
of each particular case and see whether it is worth it, without any
guidelines. 
I don't know about the  UK, but I think a patent office should be civil
servants
not elected politicians, and they should apply cear criteria, not decide
on 
economic policy.
You must give the patent office  criteria to decide what is patentabe
and what not.
We agree that software shoud not be patentable, even if only because it
is a bad deal.
So we better find a good criteria to decide what to patent. I think the
harnessing nature
idea works well to make that choice. We coud also leave it at simply the
exclusion
of software, as in the EPC, as long as we can make sure it is not
ignored by word plays:
"sure, software is not patentabe,but look, I'm only patenting software
to run on a computer, 
not software as such".
"sure software is not patentabe, but I'm patenting ways to solve an
equation through a 
computer, and that is patentable2,and blah, blah
Or more genuiney: what do we do with VHDL ? For me it is software, but
you may not call it 
so. It is logic, after all. So  we better exclude logic and that will
save us trouble

> > Does that mean that any invention containing software is not patentable.
> > Not. If the invention contains a program but also gives us new insights
> > in how to use nature to our benefit, the invention should be patentable.
> 
> You're essentially saying what I asked earlier: would it be fair to
> state your position as allowing software as a patented claim, but not as
> a patentable invention?
>
Can you please expain?. I thought that basically claims were the part of
a patent
that describes what it is prohibited to sell without a license.  I don't
want 
software to be prohibited to sell without license. I want that in case
you 
patent,say, an engine in which a sensor feeds pressure data to an
embedded computer
that outputs control signals for fuel valves,because it wasn't
previously known 
that the pressure sensed there would be an useful input to calculate the
optimum 
input of fuel, then I want  to be prohibited to sell the engine, but not
the 
embedded computer software (which in principle is pretty useless by
itself, but 
could be useful in a simulator or something, or I could publish a
program that calculates 
the same function for a totaly different purpose). 

So if software caims mean software can be prohibited to distribute
without a patent license, 
I don't want software claims either. But I'mnot so sure of my knowledge
of patent anatomy,
so please correct any false assumption. And of course an invention that 
is completely software should not be considered an invention when
granting patents.

> > HArtmut answered already, I think.
> > Programs  are not patentable. Inventions containg programs might be
> > patentable, but you would not be patenting the program as such, you'd
> > patent it as part of a bigger system. And that system should be patentable
> > without the program.
> 
> This is the argument I find contradictory. Do you have any examples of a
> part software patent which you would consider acceptable? If an
> invention was 99% software, would you think it patentable? And if the
> system was patentable without the program, how could that program be
> included in the patent claim?
>
I think I just answered. I don't care whether the  invention is 99%
software 
or 1% software, I wouldn't even know how to measure that. What I care is
that 
to be an invention it should teach us something new about the world, not
about 
how we reason about the world, so the inventive part shoud be on
physics, mechanics, 
chemistry,etc. not software, business, math, logics... If that inventive 
part on physics is just a millionth of the invention (whatever that
means), 
so be it, we'd still be selling freedom for knowledge we didn't have and
we may have 
a deal, if the rest of criteria hold. 

I think I think a program shoould not be included in a patent claim, it
could 
be included in the description of how to attain those claims (the part
that 
enables a person skilled in the art to carry out the invention). But I
wouldn't
approach it in such a"syntactical" way, but in a semantical one.

> > Maybe there is a cultural wall between us. I can't understand case law
> > systems.
> 
> Remember, the case law doesn't define the law. The law is defined - it
> just leaves parts of the implementation to the courts. The current UK

I understand that courts always have to interpret the law,  because
there 
is no other ways to enforce  it. But I don't understand why a court
decision 
should be binding for the next court decision, because that puts
legislative
power in the hands of people that have not been voted. But I think that 
is not something I want to challenge. The UK uses case law and it is
working, 
so  I just have to admit than there is more than one way to set up a
society.

> patent law states that you cannot patent art, speech, games, software as
> such - "technical effect" applies to all of these, not just software,
> and is intended (as I read it) to say that the you should be able to
> build inventable items using computers as you would a lathe or any other
> tool.
> 
 ???. I'm not objecting to patenting engines dessigned with CAD, just
saying 
that you should put in that patent some knowledge you gained from
experimentation
and prototyping, i.e. expensive use of laboratory, not only thoughts
expressed 
in the design  through CAD. Use software as tool, but don't patent
software. 
I think software with a technical effect should not be patentable, you
should
patent whatever machine uses that innovative technical effect, and let
the software 
alone. 

Thoughts, logic, math, software is better handed by copyright than
patents. 
And they may always cause a technical effect, because we always use our
thoughts
to our advantage in the world, but that is no reason to patent them,
because 
the economics and  ethics are not the same.