Xref: utzoo comp.software-eng:4693 comp.ai:8463
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!sdd.hp.com!ucsd!pacbell.com!sirothe
From: sirothe@PacBell.COM (Sheldon Rothenberg)
Newsgroups: comp.software-eng,comp.ai
Subject: Re: Estimating Knowledge-Based Software
Message-ID: <1991Jan30.180344.17841@PacBell.COM>
Date: 30 Jan 91 18:03:44 GMT
References: <1991Jan28.234259.7213@Neon.Stanford.EDU>
Sender: news@PacBell.COM (Pacific Bell Netnews)
Organization: Pacific * Bell
Lines: 24

In article <1991Jan28.234259.7213@Neon.Stanford.EDU> lockheed@Hudson.Stanford.EDU (Lockheed) writes:
>I'm looking for information about methods to support estimates and
>costs of knowledge-based software development efforts, i.e., something
>similar to COCOMO but for knowledge-based software.

Joe: 
We're looking at the same questions. It looks like it gets easier as you
use more standard techniques, shells, "ordinary" applications, whatever
THEY are. A lot of course comes down to the level of risk involved
in the project, which has such components as familiarity of user with
the technology, or with computer systems in general, and skill levels
of the development group, and organizational solidarity (i.e. leadership
with common sense of desired outcome). I have just started a lit search
on risk and so far have gotten, but not yet read:
Estimating Project Risk by W.H. Roetzheim in Information Executive mag
A Methodology fro Evaluating Potential KBS Applications by SC Laufman
DM DeVaney & MA Whiting (IEEE Expert 12/90)

 will be getting more stuff, hopefully with more AI targetting soon and
can post some to those interested..re conclusions, impressions

Shelley

A