Re: Design principles

[Jonathan S. Shapiro]

On Mon, 2007-01-15 at 10:26 +0100, Marcus Brinkmann wrote:
> At Sun, 14 Jan 2007 13:23:37 +0100,
> Pierre THIERRY <address@hidden> wrote:
> > It is probably worth noting that a point where I feel to strongly
> > disagree with Marcus is that I think the system should not undermine
> > people's ability to implement usage patterns we consider baneful to
> > society in it's whole, like proprietary software or DRM. I personnaly
> > came to that for psychological and theological reasons, but it is really
> > a matter of convinction, not of facts.
> 
> I think that view is a bit too simplicistic.  As you will find out
> later when you concretize your plans, any specific system design will
> encourage some very few use patterns, and discourage almost all
> others.  Which ones is a choice you have to make when designing the
> system.  What I don't think is that you will be able to avoid making a
> choice.

I agree with Marcus completely (which I am sure some people will find
surprising :-).

There are many architectural decisions that do not really change what
can be done in a system. They merely force people to switch from one
mechanism to another in order to achieve their goals. These
architectural decisions are not fundamental from the standpoint of
expressiveness.

The decision to support or preclude opacity appears to be a fundamental
decision. In the real world, EITHER choice (opacity or translucency)
imposes limiting effects on what is expressible in the system.

Marcus chooses a position about what he will limit, and declares as a
matter of philosophy that these limits are morally and ethically good
limits. I believe that a balance can be found in the world of opacity,
and I think that what is "good and ethical" in this space is a very hard
thing to determine. In part, this is because I do not reject from axiom
the notion of intellectual property.

Neither position can be substantiated from current fact. We know that
there are current undesirable examples of opacity (DRM). We do not have
current compelling evidence of good uses (though recent discussion here
has been interesting), but we also know that consumers do not yet have
technical means that enable them to act effectively in this battle.
Because of this, Marcus cannot say definitively that no balance can be
found, just as I cannot say definitively that a successful balance *can*
be found. Both statements are beliefs, and (unfortunately) they can only
be tested in the field.

I have a secondary reason for pursuing my position. Sometimes it is a
primary reason. I acknowledge that it is ethically difficult. I believe
that:

   1. A less perfect system that is used by a large body of people
      and improves the status quo is much better than a perfect
      system that is not used by any large number of people. This
      belief is sometimes called

         Perfect is the enemy of good.

   2. Real users will *demand* the limitations that Marcus rejects.

I should also acknowledge that Marcus's disagreement with me over this
difference is very real and terribly serious, because

         Good enough is the enemy of great.

Products that compromise have consistently beaten products that don't in
the market. The *entire* success of Linux is based on being minimally
good enough at all times. In this sense, a Coyotos OS that accepts
opacity is a threat to the HURD. Coyotos OS will allow DRM. Hurd won't.
Users like music and movies. You figure it out.


-- 
Jonathan S. Shapiro, Ph.D.
Managing Director
The EROS Group, LLC
+1 443 927 1719 x5100