Dear colleagues:
I have been lurking on the list for a while, so I thought it was
time to ask a few questions, and perhaps raise some discussion.
First of all, can anyone recommend any detailed published studies
(ones that I can easily get my hands on) that have used Space Syntax
to examine how the plan of an entire city changes over time?
I have found one very interesting study (Kubat 1999), but the
lack of statistical detail and discussion of the evidence has left me
a little confused. (I do not mean to attack the author, I recognize
that page constraints leave little room for details). Kubat claims
that: "Over the period covered by these plans, the peninsula area
has become less integrated, low values indicating shallow or
integrated systems and high values indicating deep and segregated
systems" (1999: 36). Why do low integration values mean higher
integration? Doesn't the normalisation of depth measures
(Hillier 1996: 52) result in higher integration values meaning
higher integration?
A procedural question: When comparing two computed axial
maps -- of different cities, or of the same city over time -- does
one usually adjust the legend scale so that similar colours/shades
mean the same thing on both maps? (I don't think I have ever
seen legends on integration maps in published studies!) If we don't
adjust the scale, aren't we comparing apples and oranges?
Furthermore, given the nature of the normalisation procedure,
aren't we comparing apples and oranges anyway? There seems
to be a paradox. As a city grows from Year A to Year B, we will
see an increase in traffic in the system as a whole and on the most
integrated street(s). Since it has been shown that high integration
is associated with high traffic, we might also expect that the mean
integration value of the system and the integration value of the
most integrated street will also increase from Year A to Year B;
however, from what I have seen, this doesn't seem to be the case.
How come? If we want to examine how a given street in a city
changes between Year A and Year B, should we use values
before normalisation instead? If yes (or no), why? I am sure there is a
simple explanation, but it eludes me.
Any references or information would be greatly appreciated!
Cheers
Jason
Reference:
Kubat, A.S. 1999. The morphological history of Istanbul. _Urban
Morphology_ 3(1):28-41.
Hillier, B. 1996. Space is the Machine.
_____________________________________________________
Jason Gilliland, Dept of Geography, McGill University
805 Sherbrooke St W, Montreal, Quebec, Canada H3A 2K6
Phone: (514)398-4111, Fax: (514)398-7437
|