Noah,

One of the reasons I suspect we've stuck with CA type growth (or GAs
controlling building location, or ecomorphic systems) is that we model
the process as it's sometimes perceived to be -- primarily led by the
placement of built form rather than the construction of linear
structures (and this includes the early models in the Social Logic of
Space).  The growth of urban form along roads, rail tracks, rivers or at
the intersections of any of these might well add a skeletal input to
many models.  However, there is also the question of what you might hope
to gain by doing so.  The development of a town around a particular
feature (or set of features) may be of historic interest, but what is
gained by repeating it?  This brings us back to what the ecomorphic
models are attempting to do: rather than recreate any specific town,
they form an analysis of the process by which settlements may emerge.

There's a similar thread you might like to look at from 2004:

"self organisation & morphogenesis and the structure of human sett lements"

http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind04&L=SPACESYNTAX&P=1089

(which contains an enlightening post from Alan Beattie)

Alasdair

Noah Raford wrote:
> Thanks Mike, that are very good citations and I've got most of them.
> 
> What is interesting is that although most of the work on scale-free network growth has been node - link based, most generative approaches to growing urban systems have been CA based.  There must be a way to represent similar processes in vector format?  I'm imagining an iterative all line map using something like GA to select the "best" line configurations at given growth steps.  Alain Chiaradia mentioned this idea in a private email to me.
> 
> Paul Cotes also sent along the following interesting reply which I am also forwarding to the list below.  Thanks everyone for your thoughts.
> 
> Best,
> Noah
> 
> 
> ********************
> 
> 
>>From Paul:
> 
> "prodded by your email and the interest on your mailing list, I have stuck a
> couple of things on our wiki
> 
> 
> if you go to the ceca wiki published version
> http://wiki.uelceca.net/avamsccomputingdesign/published/HomePage
> scroll down to Paul Coates
> click on and then look for alpha syntax bits (new) on my page. There you will
> find a live netlogo app of some stuff I did on an implementation of
> generative space syntax with  three state CA which people can play with.
> .The paper describes the model and also early stuff I did for Bill Hillier
> many years ago.
> 
> -----------------------------------
> 
> also I examined some dipl eng students in Berlin in 2005 who had done some
> good work using space syntax measures as fitness functions for a GA that
> worked really well in generating spatial layouts in architecture including
> corridors and such. their thesis was called  'Design Code' by Jorg Kramer and
> Jan-Oliver Kunze under Prof Finn Geipel, Labor fur integrative Architectur TU
> Berlin (sorry I cant be bothered with the German umlauts)
> 
> I have the pdf but its 14 mb. maybe they have got a website by now worth
> googling for
> ----------------------------------------
> at CECA we have been doing some generative urban modelling with our partners
> AEDAS architects, and have just got a grant to trial these techniques for two
> London boroughs, but this is not strictly alpha syntax, just anything that
> works.
> 
> Paul Coates
> Centre for Evolutionary Computing in Architecture
> University of East London
> UK
> 
> 
> 
> 
> 
> -----Original Message-----
> From:	Michael Batty [mailto:[log in to unmask]]
> Sent:	Wed 4/11/2007 12:15 PM
> To:	[log in to unmask]
> Cc:	
> Subject:	Re: Generative syntax applications
> 
> Lot of work done on network growth many years ago in geography in the 
> 1950s and 1960s on diffusion analysis a la Hagerstrand. Scale free 
> networks are essentially growth models and all the stuff in that 
> area, thousands of papers in this area - see
> <http://www.amazon.com/Structure-Dynamics-Networks-Princeton-Complexity/dp/0691113572/ref=pd_bxgy_b_text_b//gp/reader/0691113572/ref=sib_dp_pt/102-7970370-7181763#reader-link>
> The Structure and Dynamics of Networks: (Princeton Studies in C
> 
> 
> 
> The Structure and Dynamics of Networks: (Princeton Studies in 
> Complexity) (Paperback)
> by 
> <http://www.amazon.com/Structure-Dynamics-Networks-Princeton-Complexity/dp/0691113572/ref=pd_bxgy_b_text_b//exec/obidos/search-handle-url/102-7970370-7181763?%5Fencoding=UTF8&search-type=ss&index=books&field-author=Mark%20Newman>Mark 
> Newman (Author), 
> <http://www.amazon.com/Structure-Dynamics-Networks-Princeton-Complexity/dp/0691113572/ref=pd_bxgy_b_text_b//exec/obidos/search-handle-url/102-7970370-7181763?%5Fencoding=UTF8&search-type=ss&index=books&field-author=Albert-Laszlo%20Barabasi>Albert-Laszlo 
> Barabasi (Author), 
> <http://www.amazon.com/Structure-Dynamics-Networks-Princeton-Complexity/dp/0691113572/ref=pd_bxgy_b_text_b//exec/obidos/search-handle-url/102-7970370-7181763?%5Fencoding=UTF8&search-type=ss&index=books&field-author=Duncan%20J.%20Watts>Duncan 
> J. Watts (Author) "Networks are everywhere..." 
> (<http://www.amazon.com/Structure-Dynamics-Networks-Princeton-Complexity/dp/0691113572/ref=pd_bxgy_b_text_b//gp/reader/0691113572/ref=sib_fs_top/102-7970370-7181763?ie=UTF8&p=S00D&checkSum=IIVNnHLoz0SsgVQkluiVTgTGtACyhqeWvakvlJk25LI%3D#reader-link>more) 
> 
> Key Phrases: 
> <http://www.amazon.com/Structure-Dynamics-Networks-Princeton-Complexity/dp/0691113572/ref=pd_bxgy_b_text_b//phrase/New-York/ref=cap_top_0/102-7970370-7181763>New 
> York, 
> <http://www.amazon.com/Structure-Dynamics-Networks-Princeton-Complexity/dp/0691113572/ref=pd_bxgy_b_text_b//phrase/United-States/ref=cap_top_1/102-7970370-7181763>United 
> States, 
> <http://www.amazon.com/Structure-Dynamics-Networks-Princeton-Complexity/dp/0691113572/ref=pd_bxgy_b_text_b//phrase/Department-of-Physics/ref=cap_top_2/102-7970370-7181763>Department 
> of Physics (more...)
> 
> 
> and I did some stuff recently on this -
> 
> Batty, M. (2006) Hierarchy in Cities and City Systems, in D. Pumain 
> (Editor) Hierarchy in the Natural and Social Sciences, Springer, 
> Dordrecht, Netherlands, 143-168.
> 
> Batty, M. (2005) Network Geography: Relations, Interactions, Scaling 
> and Spatial Processes in GIS, in P. F. Fisher, and D. Unwin (Editors) 
> Re-presenting GIS, John Wiley, Chichester, UK, 149-169.
> 
> and also in CA modeling
> 
> Xie, Y., and Batty, M. (1996) Possible Urban Automata, Environment 
> and Planning B, 24, 175-192; also in E. Besussi, and A. Cecchini 
> (Editors) Artificial Worlds and Urban Studies, DAEST, Conveni. n.1, 
> Venezia, Italia, 1996, 177-205.
> 
> Mike
> 
> 
> 
>> Mike / CASA readers, I'm aware of your CA urban growth models in 
>> this regard.    Has anyone explicitly looked at models for network 
>> growth algorithms?  Alan, I'm very much looking forward to learning 
>> more about your and Chiron's work.
>>
>> Best,
>> Noah
>>
>>
>>
>>
>>
>> -----Original Message-----
>> From:   Alan Penn
>> Sent:   Wed 4/11/2007 4:42 AM
>> To:     [log in to unmask]
>> Cc:
>> Subject:        Re: Generative syntax applications
>>
>> Noah,
>>
>> Most of the work I am aware of in this area has aimed at defining rules or
>> processes in which these type of 'norms' emerge, rather than just coding
>> them directly as rules into the generation process. This is because analytic
>> interest is in investigating the question 'how might what we see in real
>> cities have happened?' The first example is in the Social Logic of Space
>> with the beady ring villages. The main name to search for in the literature
>> is Paul Coates. He and his students at the University of East London have
>> been working in this area for years. Bill Erickson and Tony Lloyd-Jones gave
>> a paper in the first space syntax symposium:
>> http://www.spacesyntax.net/symposia/SSS1/SpSx%201st%20Symposium%2097%20-2003
>> %20pdf/1st%20Symposium%20Vol%20I%20pdf/3%20-%20Comparative%20cities/11-Erick
>> son%20300.pdf
>> It would be worth looking up Fatiha Salah-Salah's (1987) Cities in the
>> Sahara: spatial structure and generative processes, PhD Thesis, University
>> of London, in which she reproduced local building rules from North African
>> Islamic settlements. There have been a series of relevant pieces of work at
>> CASA - look in particular at David O'Sullivan, and of course Mike Batty's
>> work on urban generative processes. Chiron Mottram and I are working on
>> combining agents with forward long distance vision and a kind of retail
>> economy in the generation of urban landscapes. Agents actively seek
>> different kinds of goods, shops survive or fail according to how many agents
>> find them, shops interrupt vision and so define the spatial morphology. He
>> will be presenting a paper on this at the Istanbul Symposium this year.
>>
>> Alan
>>
>>> Hi everyone,
>>>
>>> I've been getting some great questions recently from students at MIT
>>> asking if any space syntax "rule sets" have been applied in a generative
>>> design setting, i.e., urban growth modeling, street network evolution,
>>> design option generation, etc.
>>>
>>> Examples of such rules might include "longest lines tend to meet at
>>> shallow angles", "minimize block size in central areas", etc. as per some
>>> of Bill's early work.
>>>
>>> Does anyone know if anyone has pursued this line of thought in a
>>> computational or design setting?  Any examples on the web perhaps?
>>>
>>> Thanks and best wishes to all.
>>>
>>> Noah
> 
> 
> 
> ----------
> Michael Batty | CASA | University College London | 1-19 Torrington 
> Place London WC1E 6BT UK | Tel 44 207 679 1782 | Mobile 44 7768 423 656 |
> email: <mailto:[log in to unmask]>[log in to unmask] | web: 
> <http://www.casa.ucl.ac.uk/>www.casa.ucl.ac.uk

-- 
Alasdair Turner
Course Director MSc Adaptive Architecture and Computation
Academic Director EngD VEIV Programme

http://www.vr.ucl.ac.uk/people/alasdair