RE: Optimal cluster size: Pattern language caution, subgroup decision making
From: Rob Sandelin (
Date: Sun, 14 May 2000 06:21:39 -0600 (MDT)
A pattern language was not written with a community in mind. Intentional
Communities (and yes folks, Cohousing is a type of intentional Communty) by
their intentions are very different than the kind of unintentional groupings
that are arranged by most housing. So if you evaluate cluster size info you
would be better to look at cohousing models than  a pattern language model.

Having said that, I agree that once you get more than 15  people in a
decision process you will need more attention to your techniques and process
and certainly 30 more people will require some specialized techniques in
order to make it work well. Most large groups tend to figure out that
subgroup decision processes are easier and work better and so they divide
decision processes by functional areas.

However a common mistake, often repeated, is where groups let a subgroup
(committee,task force) go off and do a bunch of work, get all the info, and
come to the large group with a proposal or plan, which the large group then
shoots down. This usually demoralizes the subgroup to the point that people
drop out of it, the issue gets abandoned.

I have even seen this done delibrately as a way to kill an issue that a few
people didn't like. They arranged to have a subgroup create the proposal,
then quietly lobbied against it while the subgroup was working, and then
when the subgroup came back they got ambushed.

One way to avoid this kind of problem is for the subgroup to be sure they
have the authority to do the work and  make the decision BEFORE THEY START.
Frequently checking in with important stake holders who are not part of the
subgroup can also keep a good balance.

Rob Sandelin
On the road, Leaving this morning for Black Rock Mountain

-----Original Message-----
From: cohousing-l [at]
[mailto:cohousing-l [at]]On Behalf Of Hans Tilstra
Sent: Saturday, May 13, 2000 10:25 PM
To: Multiple recipients of list
Subject: Re: Optimal cluster size

I recollect that McCamant and Durrett found that with smaller groups with
fewer than 10 households, any personality clashes seemed to echo loudly.

With medium sizes, which they sized as  12-18 households), there was enough
space for individual differences etc.

With cohousing groups with more than 20 households, they found that
inhabitants complained of the anonimity, unyieldy decision making etc.

If you'd incorporate A Pattern Language's conclusions on clusters, it would
seem that cohousing would ideally comprise of two clusters.

Hans Tilstra

  • (no other messages in thread)

Results generated by Tiger Technologies Web hosting using MHonArc.