Date: Thu, 6 May 1999 17:38:59 +0100
Reply-To: John Whittington <medisci@POWERNET.COM>
Sender: "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From: John Whittington <medisci@POWERNET.COM>
Subject: Re: SAS Institute - Sanity vs. Insanity
Content-Type: text/plain; charset="us-ascii"
At 10:12 06/05/99 -0500, Michael S Hines wrote:
>Actually, according to Frederick Brooks "The Mythical Man Month" adding
>resources to a late project will more than likely make the project cost
>even more and make it later.
If one 'waits until the horse is bolted before closing the stable door',
then I can well believe that is true.
>For every Nth person you add, you have to
>add more management (some thing N is about 7) which requires more
>coordination between workers and work groups. Many of the "large group
>development projects" problems result from miscommunications between the
>groups, and individuals within the groups.
No-one is denying that increasing the size of a development group will
increase some problems, but that surely does not alter the general fact that
there will be an optimum size of group to achieve certain deadlines, and
that some group sizes would be inadequate, whilst some might be 'too large'
for that purpose? ... or are you actually saying that some large projects
are such that they are bound to take a certain absolute minimum amount of
chronological time, regardless of how many people are working on it? If one
takes the argument to extremes, that is obviously true - if it takes one
programmer an hour to write a certain piece of code, 3,600 programmers could
not achieve the same end in '1 second' - but I'm obviously talking about
more 'sensible' scenarios!
Regards,
John
----------------------------------------------------------------
Dr John Whittington, Voice: +44 (0) 1296 730225
Mediscience Services Fax: +44 (0) 1296 738893
Twyford Manor, Twyford, E-mail: medisci@powernet.com
Buckingham MK18 4EL, UK mediscience@compuserve.com
----------------------------------------------------------------
| 
