Tom,
> >> Surely the information needs to be about the
> >> relationship between people, their activities and
> >> theenvironment.
> >
> >Where do I say anything else? The point is so well
> known to those on
> >this list that I didn't bother to say it.
>
> This is a subject that I have tried to grapple with
> several times.  I thought you held that the measures
> used in space syntax, particularly the axial line,
> indicate relationships between buildings.
>
That is not quite what I am trying to say. Axial lines - and all the
other 'syntax representation maps' - are just that: a representation of
some aspects of the morphology of open space. Axial lines represent
linearity ('stringiness' in Bill's original definition) of bits of
space. But space syntax is more than just axial lines; it is about the
mapping between social forms and spatial structures, and social forms
are about relations between people. Thus the key theories are about
effects of spatial morphology on co-presence between people. The fact
that flow rates of people correlate with measures of axial maps has
consequences for co-presence. You are more likely to bump into people on
more integrated axial lines than on less integrated one (other things
being equal), hence it is possible for spatial morphology to have social
consequences. This is all very easy to say, but Adipat's original
question was about how do you ACTUALLY USE syntax in design. This to my
mind is a more difficult question to answer since relatively little is
formally understood about how people design at all in the first place. I
rely here on my experience of trying to do that - use syntax theories
and methods as part of a design team.

This is where I get my answer about the contribution to a design
'dialogue' from. It fits quite well to my mind with Schoen's conception
of design as reflective practice, as well as post-Vygotskian
constructions of situated learning, but that is by the by. I think the
real issue is that one has to be very careful about 'causal' theories in
this kind of system, and think much more in terms of emergence. I
believe that what syntax research points towards is that social forms
emerge as a result of constraints imposed by the structure of the
environment on distributed systems of autonomous individuals. It is the
emergent nature of social forms that makes me wary of normative 'design
recipe's' a la Pattern Language. I think that what syntax research
suggests is that we have to be somewhat more intelligent as designers
and look to understand emergence and harness that understanding to our
social objectives... ie. don't look to syntax to offer an easy answer or
an automated 'design machine' - lets face it we are dealing with people
with all their vicissitudes :-)

Alan

> I understand that this seems like a nice solid,
> scientific fact to work with , but I still think it is
> misleading.  It is easier to see in VGA, where you
> pick a grid of locations where people might be, and
> discover which of them could see each other, given the
> buildings in the way.
>
> This might seem like a nit-picking distinction, but
> one is measuring facts about buildings, and the other
> is measuring the physical relationship between people,
> specifically the potential for interaction.  This has
> implications both for what you measure and for what it
> means.
>
> >From a measure of potential interaction to a
> prediction of 'virtual community' is only a step short
> of a causal theory, and is intuitively plausible. If
> space syntax analysis were to give a general picture
> of potential interaction within a set of spaces,
> indicating discrete parts that facilitate gathering or
> encounter, pointing up relationships of observation or
> interface between these parts, then it would give the
> designer a much clearer "indication of the good or bad
> performance of a building."
>
> I realise that there is a huge amount of valuable
> anthropological work done in the space syntax
> community, exemplified by Julienne Hanson's "Decoding
> Homes & Houses,"  but it seldom shows on this list
> because it seems to make little use of the software.
> In fact, I have been unable to find much normative
> analysis at all in this work, beyond the drawing of
> 'genotypes' of houses.  Again, a general theory of the
> spatial structuring of human interaction would seem a
> likely route for forming causal links between social
> distinctions and spatial types.
>
> This is not to say that I think all configurational
> methods are about human interaction, but it might be
> useful to distinguish those that may be measuring
> other factors, related to other human phenomena.
> Peponis's partitions seem to measure the relationships
> between people and surfaces of buildings, axial depth
> seems to measure something like the perceived
> complexity of a route, but if they are all  bound
> together as just 'space syntax' I don't see how their
> function in the world can be discovered.
>
> >
> >Don't talk yourself down as a practitioner
>
> Not talking myself down,  I really do hope you know
> more about space syntax than I do.  It is just a hobby
> for me!
>
> Regards,  Tom
>
>
>
>
>
>
> ___________________________________________________________ALL-NEW
Yahoo!
> Messenger - sooooo many all-new ways to express yourself
> http://uk.messenger.yahoo.com