[MUD-Dev] Re: MUD-Dev's DevMUD: a word of caution

Cynbe ru Taren cynbe at muq.org
Sat Oct 31 11:29:16 New Zealand Daylight Time 1998


On Sat, Oct 31, 1998 at 02:52:45PM -0100, Greg Munt wrote:

> If the project is to move forward into something more than the discussion 
> of ideas, (I actually have reservations about whether that is actually a 
> good idea - these are described below), it needs to have a concrete 
> foundation. A foundation which minimally consists of a full requirements 
> analysis, and full documentation of EVERY facet of the software, and of 
> every decision made and why it was made.

Um, "minimally"?

Not to play the heretic, but does anyone want to list their favorite
10 software success stories and count how many of them followed this
methodology?  :)  Emacs?  Gcc?  Linux?  MOO?  (But probably X, since
it was a multi-corporation formally funded project, and likely DIKU
as a formal academic project?  I haven't seen an early history of
either.)



I -do- think that it is essentially impossible for a committee to do a
coherent requirements or design doc.

What it -can- do is supply a cogent critique reflecting a lot of
broad-based experience, driving development of a revised doc.

So if there is to be any progress beyond endlessly tossing ideas
around, one or more very small groups ("It is committees of one / that
get things done") are going to have to hammer out fairly realistic
proposals offline and then toss them at the list.

Mebbe people willing to do this should speak up.  If the count is
zero, I expect everyone can save their breath.



I'd cite the IETF at its best as an example of the above process:
members develop the drafts offline, then the group tears into it
in a review.   Iterate to fixpoint.   No?

Sometimes the results don't converge, and parallel development
efforts emerge.  This is also ok.



 Cynbe




More information about the MUD-Dev mailing list