On Sat, 6 Jan 2007 09:51:17 +0000, Chengke LIU <[log in to unmask]> wrote:
>Dear all,
>
>Thanks all for your good comments.
>
>When I draw axial map of Hong Kong, I find many sinuous paths located in
>sparse areas (mountain). I am confused how to draw axial lines
representing
>those sinuous paths. Please see the attached figure, the highlighted path,
>should I use many short straight axial lines or a few long lines even one
>long axial line to digitize it? I want to analyze the vehicular movement
>pattern using the axial map representation, and correctly representing
those
>sinuous paths should influence the final result, right?
Statistically speaking, it shouldn't matter too much as long as the number
of axial lines on those roads is small compared to the total number of
axial lines. But you will find yourself on shaky ground if that is not the
case.
There are few guiding lines and certainly no benchmarks as to what axial
maps should be, so there is no way of telling you have drawn an incorrect
map (or a correct one, for that matter).
It is unlikely that anyone ever considered seriously that earthlings
tracing axial maps of cities should be thinking about convex spaces, as
this procedure is impractical in systems with more than a handful of axial
lines. Whether the definition outlined in the social logic of space was
meaningful back in the 80s for anything at the urban scale (besides
looking cryptic enough to have a mystical status and endless re-
interpretations) is something that we will probably never know.
Hope that helps,
Rui
>
>I hope to hear from you again, many thanks.
>
>
|