NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/comp/software/5142

Internet Message Format | 1992-12-23 | 3.1 KB

Path: sparky!uunet!olivea!spool.mu.edu!umn.edu!csus.edu!netcom.com!mcgregor
From: [e-mail] (Scott Mcgregor)
Newsgroups: comp.software-eng
Subject: On CM and programmer productivity
Message-ID: <[e-mail]>
Date: 23 Dec 92 20:19:53 GMT
References: <[e-mail]> <[e-mail]>
Organization: Netcom - Online Communication Services (408 241-9760 guest)
Lines: 53

I note that in the "When do we inspect" articles that conversation has
migrated to the issue of CM systems and the overhead they introduce.

I'd like to note that in my opinion, this overhead is a result of poor
designs of the existing CM solutions (both tools and processes), at
least from the human factors point of view. (To be fair, with respect to
other issues such as data storage, retrieval, and management many CM
products are truly works of art and represent superior engineering).

Perhaps this is appropriate in a community that values the internals
over the external cognitive and organizational psychology factors, and
which would gladly accept a more limited model  over a better
commercial alternative because the limited model was free.  I know
this is a difficult sell from personal experience--our Merge Ahead
product (new version announced today) beats all others in head to head
tests on both timeliness and accuracy. It is really no surprise--we
are the only developers who spent more time on the cognitive aspects
of the user interface than on the internals.  But the number of people
who are happy driving an engine rather than a car is surprisingly
large. This may be true with regards to CM solutions as well.

Now, after having been involved in designing CM solutions for over 6
years, I'm quite convinced that we can do far better, and I think I know
how to do so. I note several other people have reached the same
conclusion and several of us are working independently to make such a
paradigmatic shift. Application of Prescient Agents (anticipatory systems that
learn your habits without programming) seems quite appropriate to simplifying
this CM area, and could eliminate a great deal of the perceived
overhead. But using such systems will be a fundamental paradigm shift
for users and purchasers.

However, without that shift there will be no way to cross the chasm
from today's CM early adopters to the mass market of people who could
benefit from CM but can't stomach the current systems.

I'm sorry I can't say a lot more about this, as this is a matter of
proprietary  advantage to clients, partners and investors.  Perhaps,
when we have a shrinkwrapped product readers will be able to see if my
predictions are correct in a set of head to head tests.  But I wanted
to challenge other people to question the way CM is done today.
Perhaps once they do so they will contribute to great improvements as well.


--

Scott L. McGregor        [e-mail]
President            tel: [phone]
Prescient Software, Inc.    fax: [phone]
3494 Yuba Avenue
San Jose, CA 95117-2967

Prescient Software sells Merge Ahead, the tool for Merging Text or Code and
offers consulting  & training in project management and design for usability.