NNSquad - Network Neutrality Squad
NNSquad Home Page
NNSquad Mailing List Information
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[ NNSquad ] Re: Worldwide cost of IT failures: $6.2 Trillion per Year? (yes, with a "T")
- To: Lauren Weinstein <lauren@vortex.com>
- Subject: [ NNSquad ] Re: Worldwide cost of IT failures: $6.2 Trillion per Year? (yes, with a "T")
- From: Barry Gold <BarryDGold@ca.rr.com>
- Date: Mon, 28 Dec 2009 17:45:44 -0800
- Cc: nnsquad@nnsquad.org
Lauren Weinstein wrote:
Worldwide cost of IT failures: $6.2 Trillion per Year? (yes, with a "T")
http://bit.ly/8XMH1F (ZDNet)
There are a couple of problems with that analysis. The biggest one is
the assumption that failures in the private sector are like those in the
government sector. But there are two major differences:
1. Government projects are more likely to fail. They are done either by
the lowest bidder (which often means the one who will cut the most
corners) or by in-house staff, who usually have civil-service
protections. That is, if you fail, you don't lose your job. At most,
you may lose the opportunity for promotion to a management position,
which most IT types don't want anyway.
2. When they fail, gov't projects fail later. Most private businesses
monitor their projects (including IT projects) fairly closely. They are
likely to be done in small stages of a few months at a time, with
interaction with the user after each stage to make sure that you aren't
designing/building a boat anchor. Government jobs can go on for a year
or more before any concrete results are expected.
So, the chance of failure is IMO lower in the private sector (including
at least some NGOs) and the direct cost of failure is also lower.
The second problem is that the list of indirect costs used in the
summary actually duplicates some costs.
So my guess is that the estimate is high. But not all _that_ high.
Probably a factor of 2, overall. Possibly a factor of 4-5. So, yes,
he's probably at the right order of magnitude ($Trillions per year).