Re: Software Patent Concerns => New Black Markets?

[patentguy@huter.us (Patent Guy)]

"Warren W. Gay VE3WWG" <ve3wwg@cogeco.ca> wrote in message > I don't fully buy this idea, though I have not read that
> piece of law. How would you classify a computer's memory?
> RAM? Core memory?  I can only imagine this as hardware.

Yes some computer hardware inventions do not lend themselves as easily
to software implementations as others.  However, most can still have a
software implementation.  For example, a hardware emulation model. 
You may model you new RAM via VHDL, Verilog, Spice, cell libraries
etc, and test the correctness of the design without ever building
hardware.  You may even build a cell library that includes this RAM so
that it may be used in other designs.  A claim to a software
implementation may allow the patent holder to enforce his rights to
his RAM invention against hardware emulators, hardware library design
houses, software design and test tool manufacturers etc.

> > Take an MPEG codec for example.  There are software codecs and hardware codecs.

> This is surely an interesting example. But I would suggest that
> it requires hardware to "use" MPEG codecs, even for a fully
> software implementation (no software runs without hardware!)

Your correct that software achieves its utility when executed by
hardware.  But the general idea that ultimately resulted in the US
courts allowing software patents is that software basically configures
general-purpose hardware to perform in as special-purpose hardware. 
Why should the special-purpose hardware (e.g. hardware codec) be
protectable and the software/general-purpose hardware not be
protectable?

A US patent if drafted correctly can also cover essentially bits
stored in/on a medium (e.g. CD with stored executable program).  This
is a bit more of a stretch.  But I would argue that an executable
program has but one true purpose and that is to be executed and
therefore to configure general-purpose hardware (PC) to operate as
special-purpose hardware (e.g. MPEG codec).
 
> If there is a patent on XYZZY compression technique
> does the patent law prevent me from writing
> and using it in the privacy of my home?
> 
> Just curious. Where does the patent law stand on this?
Technically, yes the patent holder could prevent such activities.  The
reality is what is the likelihood that the patent holder 1) will know,
and 2) will care about such a deminimus?  It's vey unlikely that it is
in the patent holder's best interest to sue you under such a senario.
> 
> If I then share that same XYZZY code, am I the software creator in
> violation?
A patent generally gives a patent holder the right to prevent others
from making, using, selling, offering to sell, or importing the
invention.  As software creator you have probably infringed the making
right.
> 
> Or the user of the code?
Those people that receive the code are likely to infringee at some
future date the use right.

> >So in my opinion is the key
> >question is how to we retain the benefits of non-commercial and
> >hobbiest activities without abolishing software patents?  I personally
> >think the best way is to lobby Congress for some type of
> >non-commercial/hobbiest defense to patent infringement.
> 
> But how do you do that? I mean, will it be legal to write
> and use MPEG codecs for hobby purposes in GPLed code? What
> if businesses use internally, without selling/distributing
> it?  What if they distribute Linux distros with it in it?
It really depends on how such legislation is drafted.  For example, I
could see the FSF lobbying that all making, using, selling, offering
to sell, importing of GPL and LGPL software was non-infringing.

The OSI might want to expand this to include other open source
licenses.

My guess is that the legislation couldn't name a certain license (GPL,
BSD, etc) but would need to define characteristics of the software
(e.g. open sourced, freely copyable, freely distributable, freely
alterable, etc) that would fall under the exception.

> 
> What about DVD decoders on Linux? There is a real _need_
> for such, and the motion picture industry does not want
> to provide any such thing for Linux. What choice
> do the Linux users have? The only legal choice is not
> to view DVDs under Linux. But many feel this is going
> too far.
The problem here from my understanding is the DMCA (copyright) and not
patents.

> 
> I doubt that any hobbiest is ever going to receive the same kind of
> treatment that a life-saving physician will get. It is easy to sell this
> idea to law makers for saving lives, but much tougher in the name of the
> FSF/GNU. First of all, they have a difficult time understanding how this
> is in their best interest.
I agree that a life-saving physcian is more sympathetic.  However, I
also believe that a patent exception of OSS software is much more
tenable than abolish of software patents.  You may find segments of
the computer industry and business sector that support the exemption. 
I suspect that such industry support does not exist for abolish. 
Also, I think that an exemption truly achieves the FSF goals.  It
would keep open software open and free from patent issues.  It may
even spur more open software in order to avoid patent issues.
> 
> I have trouble agreeing that simple ideas need to be protected
> in the first place.
Let's say you came-up with a great idea for a toy.  With proper
marketing, distribution, etc you you could make millions on the toy. 
However, the toy is simple to make and once Milton Bradley or one of
the other toy makers notices your success they will be able to
undercut you and grab the whole market.  Without a patent, you are
basically screwed because there isn't much you can do to stop the
other manufacturers from appropriating your idea.  Do you still think
simple ideas should not be patentable.

> 
> "They decreed that the stored-program idea rightfully
> belonged in the public domain."
> 
> And that is how it was decided, and we enjoy its  benefits today.
If the patent had been filed in 1944, it would have expired roughly
1964.  I doubt that it's existence would have had much effect on the
computing industry from 1944 to 1964.

 
> >Also, most simple inventions are easily avoided.  For example, use a
> >2-click ordering system instead of a 1-click ordering system.
> 
> This only works until some enterprising individual recognizes
> this and patent protects the 2-click idea.
> have to throw away your mouse to do online shopping because
> you can't n-click either, without paying royalties to the sap
> that is sucking the system dry.
The problem here is that 2-click has been done.  As a result, if the
US patent office is doing their job right (which is a big assumption
since they are doing a crappy job in all areas not just software),
then such a patent will not issue.  Also by definition, a patent can
not cover a product that existed more than a year prior to the
patent's filing date.