From: Bill Kelly <billk@...>
Date: 2004-12-03T11:48:54+09:00
Subject: Re: SOT gmail invites

From: "Gavin Sinclair" <gsinclair@soyabean.com.au>
>
> On Friday, December 3, 2004, 9:25:50 AM, Bill wrote:
> 
> > Software patents are like, if you play guitar, Eddie Van Halen
> > patenting Tapping (two-handed hammer-on technique).  The damned
> > *IDEA* is the easy part.  It's the implementation that's hard.
> 
> In building software, though, turning ideas into profit is an
> expensive and risky exercise.  Much more so than busting a few guitar
> moves.

Turning ideas into profit focuses on where the real
work is involved - the implementation.  Which is protected
under Copyright law, and which has always seemed pretty
reasonable to me.

> Once you've seen the benefit of an idea you'd never have thought of
> yourself, the implementation is often trivial.  Just like many guitar
> techniques :)

Hmm...  To me, software patents and guitar techniques often
fall into a category I'd say is pretty opposite from that:
trivial to explain, very difficult to implement.

Anyone can explain Tapping (two-handed hammer-ons) or 
Rasgueado (flamenco strum) or Slap/Pop (bass) techniques
conceptually in about 20 seconds.  Now hand the instrument
to the student and ... Maybe after about a year of diligent
practice, you'll be listening to a pretty good implementation.

With software, at least where I hang out, the situation has
seemed pretty similar.  It's gotten to where there should be
a FAQ for amateur 3D graphics / game programmers doing voxels.
Someone will post their application/demo to a community site,
saying, "Hey, I was thinking about 3D pixels (voxels) and I
wanted to make a game where I could fly around through a 
cave.  Here's how I did it, in my spare time, in between
studying for finals.  What do you think?"  And the standard
answer is, "Neat! It may interest you to know that what you
discovered there looks a lot like the Marching Cubes algorithm,
which someone has decided to patent." 

Over and over, it's NOT about "an idea you'd never have 
thought of yourself."  Not even remotely.  It's about bleeding
obvious* ideas that come up when you get into the nitty gritty
of figuring out how to implement something.  ((*) By bleeding
obvious I mean when you're deep, deep into the problem, and
suddenly it's like, Ah-ha!!  I.e., bleeding obvious to anyone
who has arrived at that same deep context, driven out of 
necessity to solve the problem.)

http://www.base.com/software-patents/disputes.html
... The IDEA of host-independent network byte ordering ? ? ? ?

>  Entity A proves the commercial viability of an idea at
> great expense; entities B-Z exploit that idea at no expense.

Why is the software developed by B-Z less expensive than that
of A to implement?  Take a look at Quake (the 3D first-person
shooter playable over the internet.)  id Software not only
didn't patent their techniques, they released Quake open source
under the GPL when Quake II came out.  And you know what?
Even having the *source code*, let alone just the *idea*, I
can't turn around and snap my fingers and come up with my own
implementation of what their game does.  Ideas are easy,
implementation is hard.  Even staring at someone else' source
code.


Regards,

Bill