 |
| IT whiz Gilb: Advocating a 'no cure
no pay' approach |
Tom Gilb first came to India in the
mid 1960s to help two big old economy houses-first the Birlas, then
the Tatas-take their first tentative steps in the then alien world
of information technology. Four decades later, it may have become
synonymous with India, but Gilb, 65, isn't exactly thrilled at the
way the global technology market has evolved. He believes that companies
often end up paying for faulty or useless code. "As long as
customers foolishly pay for body-shopping, and system construction
(writing code), without defining the real expectations (such as
time savings, or people savings) then there will always be suppliers
happy to take our money, for fully or partly failed projects."
The core of the problem, Gilb believes, is that it companies
tend to hide specific requirements from a contract in jargon when
they ink deals with their customers. The result: companies find
out there isn't really a way to determine if the project has been
successful or not. And the companies themselves are sometimes
to blame. Gilb says that requirements from customers are so poorly
drafted that managers often find as many as 80-100 major defects
per page, resulting in bug-filled projects (remember that old
tech-adage, Garbage In Garbage Out?).
For example, Gilb, who describes himself as a teacher, consultant
and writer, discovered that a tender for a telecom project funded
by the Government of India had at least 70 to 80 ambiguous statements,
resulting in bug-filled projects. Faulty projects clearly bite,
with the UK, for instance, discovering that half of the £19
billion (approx Rs 1, 59,600 crore) spent on it projects (50 per
cent was public money) in 2004 went into failed projects. Despite
these statistics, Gilb argues, "people are not motivated
to change, because the customers pay so well for failure".
Now Gilb, the author of nine books on software and process engineering
and a consultant with companies such as hp, Boeing and Ericsson,
believes that a complete makeover for the tech industry is warranted.
"My alternative is called a 'no cure no pay' method, where
suppliers are paid for actual provable, measurable and testable
results from their (new) it systems," he says. Gilb has been
part of the it industry since he was just 17-he started off with
IBM in Norway while studying distinctly non-technical subjects
such as psychology, sociology and economics-but believes that
some tough-talking may be required to make his radical approach
to software more prevalent. "We have to refuse to do business
on any other terms." One part of Gilb's recommended makeover
suggests re-examining the yearly system of contracts and breaking
these deals down into smaller (and more manageable) quarterly,
monthly or even weekly deals. Another says that code-jocks "must
learn to approach the problem, from a 'holistic' position (all
necessary disciplines must be applied, not just coding and pure
software)".
A fully-wired summer cabin (on the waterfront surrounded by
a 100 trees, a private beach and pier as well as a "private"
hill) just 30 minutes from Norway's capital, Oslo, allows Gilb
to escape the rush to do some writing and reading, he often finds
himself drawn out of his cocoon and on a plane back here to India.
Blame it on the urge to transform an industry.
|
