Thank you to
everybody who encouraged me to continue with these newsletters: it's nice to
know that you don't feel that this is spam. Procrastination almost won a total
victory - it's almost a year since the previous newsletter.
If you do not
want to receive this newsletter, please email me to have your name removed.
In this Newsletter:
-
IT
Doesn't Matter. Is "IT
Strategy" passé?
Feedback
from the previous newsletter. The
new economy, open source.
Book
Review - Feature Driven Development
I've added a discussion page to
my web site: hopefully you'll feel inspired to contribute your thoughts.
CEO's routinely
talk about IT's strategic value, and how they can use it to gain a strategic
advantage. In the most recent HBR(May
2003), Nicholas Carr takes aim at this view.
He argues that, while opportunities to gain enduring strategic advantage
with IT existed when the technology was new, now that IT is ubiquitous this era
is over.
"IT is the latest in a series of broadly adopted technologies -
think of the railroad or the electric generator - that have reshaped industry
over the past two centuries. For a
brief time, as they were being build into the infrastructure of commerce, these
technologies created powerful opportunities for forward-looking companies. But, as their availability increased and
their costs decreased, they become commodity inputs. From a strategic standpoint they became invisible; they no longer
mattered. That's exactly what's
happening to IT, and the implications are profound".
Most commentators
have compared the boom and bust of IT investment with previous rollouts (Rail,
electricity, etc), but little has been said about the way the technologies
influence competition at the level of individual firms. Carr says that it is important to
distinguish between proprietary technologies (example, drug patents), which can
confer lasting competitive advantage to a single company, and infrastructure
technologies where maximum value requires that they be shared. There is only a brief window for gaining
advantage from an infrastructural technology: when its commercial potential is
broadly appreciated huge amounts of cash are invested in it, and its buildout
is very rapid. By the end of the
buildout phase, opportunities for individual advantage are largely gone. The rush to invest leads to more
competition, greater capacity, and falling prices: at the same time the
technology standardises, and best practices are widely understood and
emulated.
Carr acknowledges
the way in which the superior IT insight of companies like American Airlines
(Sabre), Federal Express, Reuters, eBay have allowed them to stake out
commanding positions, and others like Wal-Mart and Dell have been able to turn
temporary technology advantages into enduring positioning advantages, but he
argues that opportunities such as this are dwindling, believing that IT
buildout is nearing completion. Signs
of this are: -
In this
environment, Carr's rules for IT management: -
IT and the Internet
are simply part of doing business.
Businesses cannot ignore IT, but they don’t have to be in the
vanguard. As always, IT projects should
be backed by a sound business case.
Where does this
leave "The Strategic Use of IT"?
Is this view obsolete? Not if the message is that a good understanding
of the capabilities of IT, and how it can affect your business strategy, is
essential. But "Yes" if the
message was that innovative IT could be a substitute for a sound business
model, and that it was essential to seize the first-mover advantage.
More on this
subject: -
Harvey Lockie comments that we can learn much from
previous "New Eras" (Canals, Railways, Radio, Motor Vehicles). 90% of
those who invested in the new technology lost their shirts, but there were
profound (and often unexpected) benefits to the economy as a whole. I've put Harvey's comments on the
discussion page.
Several readers considered that IBM's support for
Linux was driven by a desire to hurt Microsoft as much or more than because
it's a revenue opportunity. I
disagree.
If IBM really wanted to attack Microsoft, they could
have made OS/2 an open-source product in about 1997? If the open source advocates are right about the power of the
open source development model, then this could have provided a competitive GUI
with a developer pool that could have threatened Windows' hold on the
desktop. A desktop environment backed
by both IBM and the open source community could have been very attractive.
One of the
central challenges of computer science is to get a computer to do what needs to
be done, without telling it how to do it. In the May 2002 newsletter
I commented on a new (to me) approach to problem solving, genetic programming,
which addresses this challenge by providing a method for automatically creating
a working computer program from a high-level problem statement of the problem.
Genetic programming achieves this goal of automatic programming (also
sometimes called program synthesis or program induction) by
genetically breeding a population of computer programs using the principles of
Darwinian natural selection and biologically inspired operations.
The approach is
mentioned again in a recent HBR (May 2003) article, "Don't
Trust Your Gut", where "Interactive Evolution" (where human
judgment is used to guide the artificial evolution) is cited as a method of new
car design.
I thought that
I'd better learn a little more about it. A search for "Genetic
Programming" on Google found Genetic-Programming.org,
which seems to be a good starting point for anybody wanting more information.
Feature-Driven
Development (FDD) is a methodology that combines the benefits of small-scale
agile methodologies like eXtreme Programming, with scalability and enough
formality to satisfy the requirements of enterprise programming. According to the authors: -
"FDD does enough work up front to provide a resilient conceptual
framework - the domain object model (structure) and features list
(requirements). The highly iterative,
self-organizing controlled chaos of the Design by Feature and Build by Feature
iterations provides an agile operational framework that can quickly adapt to
change.
"This balanced approach advocated by FDD avoids the analysis
paralysis often found in teams following traditional processes with long
analysis phases. However, FDD also avoids the large amounts of unnecessary
reworking of code that are almost guaranteed when a team dives straight into
coding without any reasonable shared understanding of the problem to be
solved."
As in eXtreme
programming (XP), there is a very strong focus in FDD on rapid iteration and
incremental delivery, with both methodologies sharing the belief that work
units should be kept very small and focussed on delivering function to the
user. Unlike XP, FDD recognizes the role of modelling in defining the solution
architecture (the methodology is derived from Coad's "Modelling in
Colour") and FDD supports formal roles and responsibilities such as
technical architect and project leader.
It's also more management-friendly, recognizing the traditional roles
(software architect, project leader, etc), and so FDD will be more acceptable
in larger organizations than XP as it fits much more comfortably with normal
management structures.
I agree with Steve
McConnell that the answer is not in any particular methodology, but in
having a good toolkit of methodologies, and knowing which to apply. FDD is a
good addition to the toolkit, and will be of particular interest to those
wanting to get the benefits of agile methods in larger projects, or when the
company culture would be antagonistic to less structured methods.
In Office 2000
Microsoft have allowed us to get rid of the **** paperclip, but they have
introduced a new improvement that's almost as annoying - now the menus only
show the recently used commands unless you click again to expand them.
There is an
option to have full menus, as normal - but don't expect to find it in the
obvious place, under /Tools/Options (View tab). Instead, it's under the Customize menu. Select the Options tab, and uncheck "Menus show recently
used commands first".
