Re: [agile-usability] Re: Today's article on UseIt.com
- sepreece wrote:
> Would the team work better if we were all in one place? Maybe, but weI agree that for some circumstances, this is not an unreasonable
> wouldn't get the chance, because the people (who we value) want to
> live where they are.
tradeoff. Just as long as people are aware of what they're trading.
It's very rare for me to meet somebody who a) has worked as part of a
well-meshed, fully collocated team, b) is now on a dispersed team, and
c) thinks dispersion is just as good. In fact, I can't even remember the
last time it happened; had it come up recently, I would have asked to
observe the team.
Typically, when I ask people who meet a) and b) what productivity hit
they believe they are experiencing, it's between 20% and 50%. People on
the 20% end of things generally take aggressive steps to bridge the gap,
like launching the project with 6 weeks of collocation, having personnel
exchange programs, keeping the same time zone, and/or spending lots of
time pairing virtually.
Often people I talk to in large companies notice no difference, because
they are in effect working in dispersed teams already, or have enough
other communication barriers that the cost of dispersion is not
significantly larger. This is similar to the way that many large
companies report success with offshoring: they are already so
document-driven and working on such long feedback cycles that using
developers on another continent make things no worse.
That's not to say I don't know any happy dispersed teams. They're all
aware of the costs, but are working in sufficiently uncompetitive
contexts that they can afford the costs.
P.S. As a matter of simple math, if a collocated team of 7 adds three
people remotely and takes a 30% productivity hit, they are no better
off. This sounds ridiculous, but I just recently visited a shop that
closed down their offshore dev office because they were in exactly that
situation. Why did they do it? Well, productivity is hard to measure,
and the execs wanted it to work because it seemed cheaper. So everybody
tried to make it work for quite a long time.
> > There are other studies (I don't have the exact quote) that showthat
> > the difference between a top-notch developer and a run-of-the-millone
> > is a factor of 10 or so.http://forums.construx.com/blogs/stevemcc/archive/2008/03/27/productivit
> The back up for that is here :-
Thx James. I always assumed that this was indeed supported by actual
studies, but still had a small nagging doubt that it might be one of
those urban legends that start with "studies show that ...." ;-). This
is a good reference which eliminates that doubt.