Rui, Anzir,
This just shifts the problem from 'what is a street' to 'what is a city' -
Defining a density threshold might sound attractive as a proposition but it
is not as simple as that. Amongst the problems are: 1. 'density of what?' -
is it density of residential population, habitable rooms, area of
floorplate, floor area of buildings? 2. Density is something per unit area -
what is the area over which you are measuring? You are suggesting a density
threshold as a means of defining a boundary (the built up area of the city)
but cities are essentially inhomogeneous over the kind of factors in 1. i.e.
at some spatial resolution Regents Park would be considered devoid of
residential population or buildings and so not part of the city. At a finer
scale of resolution so would my back garden or the street in front of my
house. If I eliminate all of that 'open space' then I am only measuring the
building area, but then a single house in the middle of the Sahara would be
just as much a city as anything else. To make any sense of this kind of
thing you need to start with a boundary and then calculate density within
that. Back to the Italian problem of the administrative boundary including
functional regions which encompass rural hinterlands....
It is all a bit like the old problem of defining an elephant - 'its large
and its grey and I know one when I see it'. There is no doubt in my mind
that Milan and Rome are cities regardless of how they choose to define their
administrative boundary. There is also no doubt that density of the built up
area is part of that judgement. But there is also no doubt that density by
itself is not enough and can become circular if you try to use it to define
the boundary over which it itself is measured. It may sound simple, but its
not..
Alan
>
> Rui,
> On 3 Jun 2007, at 11:18, Rui Carvalho wrote:
>
> > On Sun, 3 Jun 2007 12:45:11 +0300, Yodan Rofe <[log in to unmask]>
> > wrote:
> >
> >> Oh really?
> >
> > Well, I didn't say you would recover cities "as we know them". I
> > said that
> > a city could be defined by population density and you would need to
> > decide
> > on a threshold.
> >
> > Localized population peaks (like the villages you mention) are
> > obviously
> > not a city. Of course, you wouldn't recover the boundaries of those
> > italian cities you mention, but that seems only natural to me.
>
> Only cover a city to the edge of its built up area, and do the
> density count within that. Similarly for the villages...
>
> I think you mean to define a "built up area" rather than a city per
> se. Whether the built up area is small or large is irrelevant if
> you're going to define a threshold for "built up area". You may also
> need to have different thresholds in different cultural milieux,
> where different patterns of "urbanism" have occurred.
>
> --
> Anzir Boodoo MRes MILT Aff. IRO
> transcience, 72 Staplehurst, BRACKNELL RG12 8DD
|