Re: [jos-general] new to jos

[John M. <jm...@ma...>]

Hi;

Gilbert Carl Herschberger II wrote:
> At 12:01 PM 8/25/01 -0700, Iain Shigeoka <iai...@ya...> wrote:
> >When I joined, the entire purpose of JOS was
> >to create a jos within a relatively short period of time.
> 
> It has never happened; it isn't going to happen. Nobody can create an
> operating system from scratch in a "short period of time". I think it is a
> little naive. Let's take a fresh look at this realistically.

No it is not naive.  The best way to do this is not the classical
"waterfall" model (i.e., use clairvoyance to design everything the Right
Way up front as the obsolete MIL-STD-2167 military software development
model dictated).  It is the "rapid prototyping" model (get something
running ASAP, learn from the prototype both about problem space and
solution space, and refine indefinitely in cycles).  Depending upon
clairvoyance is naive, and smacks of hubris.

> The speed in which the operating system is developed is hardly important.
> What is important is the high quality of its design, the high quality of
> its interfaces, the high quality of its implementation. Difficult issues
> must be solved. While the inspiration for an operating system is easy, its
> construction is difficult.

I agree that high quality is important, and that difficult issues must
be solved.  I disagree that they can be solved clairvoyantly.  I have
never seen the solution to an "interesting" (read "complex" or
"challenging") problem fail to evolve significantly based upon
experience.  If you can solve the problem a priori with respect to
trying, it's not an interesting problem.

-jm

-- 
==== John Morrison
==== MAK Technologies Inc.
==== 185 Alewife Brook Parkway, Cambridge, MA 02138
==== http://www.mak.com/
==== vox:617-876-8085 x115
==== fax:617-876-9208
==== jm...@ma...