[cap-talk] Capabilities and Freedom vs. Safety

[Jonathan S. Shapiro]

On Fri, 2007-07-27 at 11:44 +1000, James A. Donald wrote:
> I am profoundly irritated when some says that I have not
> explained X, and urges me to do so, when I have
> explained X rather too many times, but evidently not to
> his satisfaction.

James:

Speaking as a teacher, my observation is that you haven't explained X
until the listener understands you. In any process of explanation, the
primary burden of explanation must lie on the explainer. This is
necessarily true because the explainer knows what it is that they are
trying to explain, while the explainee does not.

This can be very frustrating.

> A more appropriate statement would be the pious "I am
> unable to follow your reasoning" - or the blunter "Your
> argument makes no sense to me", preferably followed by
> some illustrative explanation of why the argument seems
> to fail in relevance or internal coherence...

Several people have attempted variants of this response to you many many
many times in the present conversation. You have often responded to them
with frustration and intolerance. Many of your "arguments" have appeared
(at least to me) to be devoid of logical structure -- by which I mean
that you string together a sequence of sentences whose chain of
reasoning lacks a continuous thread. In the future, I will try to
identify those sequences by saying so bluntly. I ask in return that you
fix them by filling in the gaps rather than by merely reiterating them.

In some cases you have explicitly ignored or denied some clear
statements of fact and mathematics made by other participants. When you
do this, people simply lose interest in your view entirely. It may be in
some cases that the differences lie in differing assumptions. If so,
figurative shouting will not reveal those differences. In either case,
ignoring mathematical fact is not productive. If the mathematical fact
is not relevant, it should be possible to clearly state why without
resorting to ad hominem statements.

In other cases you appear to ignore designs alternative to your own
which demonstrate that your assumptions are incorrect. One recent
example is your assertion that useful programs require a lot of
authority. Several people offered other designs for functionally
equivalent programs in which this authority was properly factored, which
demonstrates that the requirement for aggregation is a consequence of
bad design rather than an intrinsic characteristic of those
applications. I have not yet seen you so much as acknowledge those
counterexamples. Unless I have missed something (which perhaps I have),
you continue to insist that aggregation of authority is intrinsically
necessary. This view does not reconcile with the counterexamples you
have been given. As a concrete counterexample, you might look in detail
at the DarpaBrowser document.


shap
-- 
Jonathan S. Shapiro, Ph.D.
Managing Director
The EROS Group, LLC