Yourdon, in "The Death and Decline of the American Programmer", the textbook
whose success marked his descent, listed offshore development as one of
those factors in the decline of the American programmer. Some work can be
successfully sent offshore, but there are issues to consider. I should
provide the disclaimer that CGI acquired IMR Global last August. CGI already
located work in Canada, and IMR Global has been developing in India for some
time. However, my comments are my own, and should in no way be taken to
reflect any company's position.

The one word in Raghu. Rao's post the compelled me to reply was "quality",
from "if they can get the work done in shorter time with better quality". I
think that you can actually subtract that word from the equation, and the
equation still holds: What will it cost, and when can I have it? Too many
managers would not know quality if they could see it. Worse, I doubt we
could even begin to agree on what quality means in this context.

Will an offshore shop adhere to your shop's standards? Can you define your
shop's standards? What about the resulting documentation, in the form of
separate documents, and in the form of comments within the code? Will an
offshore company understand the practices of your industry? Before CGI
bought this office, we were ISI Systems, Inc., and specialized in the
peculiarities of Massachusetts insurance regulations. Some of our employees
have been associated with the state regulatory agencies, and bring a
distinctive perspective to those lines of business. Can you provide correct
and complete specifications for the work to be done by the offshore company?
Will management know the difference before it is too late?

Someone recently posted an article on the psychology of unexpected failure
or something like that, where people failed because they lacked the
knowledge to succeed, but lacking that knowledge, they did not know that
they were doomed to failure. If the decision makers cannot distinguish
between what makes for success and its absence, then success is a random
occurrence, and they may as well consult the planets and constellations as
their project plans and status reports.

OTOH, Y2K is exactly the sort of thing that can be outsourced: labor and
analysis intensive tasks that are well-defined and readily understood, where
any costs saved achieve an economy of scale. One real benefit of offshore
development is that work that is too massive to be undertaken at the
prevailing cost per line of code becomes feasible, when that cost can be
slashed. Some pervasive artifact of times long past can be cleaned out of
the code, instead of being tolerated, because it becomes affordable to do
so.

The trick is knowing the difference, and making sure that those difference
are understood and respected. These waters are fraught with perils,
especially for those most inclined toward them.

Greg Stigers
http://www.cgiusa.com

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *