Re: Article: The shift away from user directed projects

[Harlan Grove <hrlngrvNOhrSPAM@aol.com.invalid>]

You're spamming a lot of newsgroups, but I suppose that's
unavoidable given you want a wide range of developer
responses.

There's an unavoidable tension between horizontal (generic,
IT-directed development) and vertical (industry-
specific/user-directed development) knowledge and
technology. Accounting systems are a good example.

Manufacturers, distributors, industrial and consumer
service providers, banks, insurance and other financial
service providers have significantly different needs when
it comes to accounting systems. Manufacturers and
distributors have much greater concern managing
depreciation and inventories than the other types of
business. Banks, insurance companies and to a lesser extent
other finanical service providers have to have systems that
support risk-based capital analysis, a concept pretty much
foreign to the other types of business. It's not possible
to design a one-size-fits-all accounting solution, at least
not for large corporations.

Once you have to design specific solutions for certain
industrial sectors this nirvana of IT-directed, nonspecific
development dies off. It may work reasonably well for the
(horizontal) software foundation - examples: ODBC, OOP,
RDBMSs, ftp and electronic data transmission in general,
IEEE floating point math, image formats like GIF, TIFF and
JPEG, etc. It doesn't work for satisfying (vertical)
business needs like accounting, decision analysis and
support, EIS, budgetting and forecasting.

As for developers on their own anticipating business and
private user needs, there are a lot of failed products that
have come out over the last 15 years - Compaq's integrated
phone and PC, integrated productivity applications, almost
all 'paperless office' schemes. And arguably the most
widely used programming language, perl, has been designed
by its users (I'm not saying this was a good thing, but it
was apparently a popular thing).

The best model for software development is evolutionary
change, and that's best served by as many design and
development channels as possible. Both user-directed and
ivory tower projects contribute to software progress. Maybe
it's all boils down to answering the question: how much
more could have been achieved in the last 15 years if
software development had remained in IT departments and the
entire shareware and open source phenomena hadn't occurred?
I think my response would differ from yours.


* Sent from AltaVista http://www.altavista.com Where you can also find related Web Pages, Images, Audios, Videos, News, and Shopping.  Smart is Beautiful