sounds like a tough call w/o seeing the progress, but id say you should strengthen your case for the fractal as a system that seems appropriate. for example, the idea of merging 3 cities sounds like you need something to serve as a "stitch" that binds these in terms of infrastructure.

another approach might be to argue for the adaption (mutation) of a fractal system to operate more contextually.

i personally dont like to word fractal, so the real measure of success is if you can find another way to describe your design agenda w/o relying on the word "fractal" at all.

as for precedents, alison + peter smithson berlin haupstadt project is a good start. (Also other Team 10-ers). enjoy!

could be an applicable overlay system - reasons
#1 villages (versus towns or cities) are not necessarily "planned" developments. they evolve into being. often in relationship to terrain, property control, etc. they are non-linear clusters where surveying may not have entered into play
#2 the location in relationship to one another would express some triangulation - versus linear alignment (unless its a straight cut valley terrain or similar)

your approach is more organic to clustering - a more natural linking
in contrast to a conventional planned urban alignment from the ground up. typical linking is very linear - which does not reflect the original practicality of small settlement / village structure. I think of the negatives of adding highways & interstated to link small established communities (villages & towns). They invariably bleed the commerce that made them come into being away by them being bypasses as opposed to connective threads.

You might want to look at what the future long term impact would be as they expand in scale taking into consideration the typical negatives of urban infrastructure in a century or so.

curious - is the project Asian or ?

I think it would be interesting to utilize the overlay as a starting point then actively engage it to see how the fractals would impact the new and then in return the existing- especially through time. It is similar to a computer program with a flawed line of code. It will work and continue to do so, but as time goes on and the flawed code propogates infesting the system it will take on characteristics of a virus altering pieces of the code of other parts of the system and of course the system will react accordingly to expunge the virus- thus a Red Queen race through the generations.


Peter Eisenman has an article on his "Scaling" theory applied to a case study of... I think a castle for Romeo and Juliet. Very good and may be pertinent.

thank u all for ur responses. all were very helpful and im currently looking into the berlin project and eisenman's study .
cobra i see ur point and it would be interesting to have a settlement-free point as a referential one from which i can grow this system and then test the adaptations that would occur simultaneously on the three. though i'm still not sure which is more a valid approach doing this or following a more natural scenario and then extracting the fractal order as an inherent one and carrying on with the same rythm. so in both cases a contextual mutation/adaptation is definitely gonna happen!

the project is Asian, more accurately a Middle Eastern one – 3 villages in Kerak governorate, Jordan.
i was actually interested in adopting a fractal order as an underlying one which is inherently existent in any natural system. connecting the three together does produce a triangular relationship, so in a way this would be allowing them to reach their natural organized complexity by following a natural flow. ideally this flow would eventually slow down and reach an interface that connects to the 2 other villages, so by converging inwards they would naturally merge into one.

i’m still looking into the impact of this on each village separately, tackling the existing infrastructure is very important but i think it would eventually serve to my benefit, as a geometrically coherent network of paths. generally the three are already an independent unit; each is complementary to the other two and undertakes a specialized role, mainly production, education and culture/tourism. that’s why i’m considering a hexagonal structure in terms of hierarchy and maybe later on as a physical application of the fractal as well. again.. thinking out loud!

In a hierarchy, if the three existing were to merge the new would be critical in facilitating this event and it would be paramount for interaction to occur freely throughout. I believe the project lies in the small differences you mentioned between the three existing and the new, the rest are only controls and can be quickly adapted- but if the differences are not infused the project could end as another Los Angeles albeit at a smaller scale.

On the otherhand, you may want to look at utilizing a heterarchitical system where each node can be connected to any of the other nodes essentially removing the word node from the project. In this typology the infrastructure provides the facility.

Also, it may not be necessary to focus on the nodes converging inwards. Rem Koolhaas has also written on this specific subject when considering... I belive the Lille project. I know there is an essay in Theorizing a New Agenda for Architecture : An Anthology of Architectural Theory 1965-1995, amazing book. Essentially, he studies the relationship between time and distance arguing time is distance which may discourage convergence and encourage emergence.

I am thinking aloud as well

My first impressions are that, 1) you'll need to find some repetitive element in the cities to base your larger geometry off of. This shouldn't be *terribly* hard considering the popularity of the grid system, but it would be great if you could take it beyond that, possibly finding more than one element to use in this way. And, 2) you need to look at some very complex and compound fractal art in order to see the vast variety of ways you can go with this. Everyone thinks they understand fractals because they've seen a basic Mandelbrot (sp?), but there are so many variations on the Mandelbrot, and also so many other equations to start with, that I think it's important to delve more into this to wrap your head around it instead of taking 'fractal' at its most literal and basic meaning, as so many people do. Possibly get ahold of some shareware of UltraFractal or a similar program and play with it yourself.

I think this is definitely a doable concept, you've just got to figure out how far you'll take it, and how much you want to impose vs. how much you want to discover. I'll leave you with a link to some award-winning fractal art, albeit on a very badly done website (sorry, grandpa! but it really is). The Fractal Art of Red Williams

You might want to look at tiling and patterning, specifically, Roger Penrose's work on repetitive patterning on a flat plane which is quite a compex science, and related to fractals.

I investigated randomly generated tile patterns [tied to some classifying system], and then blended and overlayed tiles to generate blurred figure-ground diagrams for use in generating integrated urban landscapes and building forms. You might be able to extract some patterns from the existing towns, or accentuate 'successful urban spaces into a 3d pattern to be applied to infrastructure.

I think that some extensive invention and application is necessary for this kind of project, rather than establishing a framework to be filled in that is justified as being a natural event, and therefore 'good'.

Here are some links to webpages/software that generate Penrose Tiles:

Link 1
Link 2
Link 3

danger of penrose tiles is that you could end up w/ something like this thing by ashton raggatt mcdougall, melbourne:



aaaaaaaaarggghhhhh!

Hey,

I walked past that building everyday for 2.5 years - RMIT Architecture School is right next door. It rocks, and I will have fisticuffs with anyone who says it's not.

addendum: who says it does not.

I don´t know how much you know about fractals in relation with architecture, but there are many examples about this, recently I read an article in wich the autors investigated the relation between fractal and the city (specialy, that kind of cities that suposely haven´t been planned) and the results are amaizing, so I think your idea is not an imposition (in bad sense) I think is a great idea.

Continue hibz.

interesting idea, but if you treat urban planning merely as a graphic design exercise you might as well make a city that looks like winnie the pooh as a fractal.

that fractals appear in existing cities/towns is interesting but in the end not as important as one might think. it certainly is not the answer to making utopia.

spiro kostoff did a good overview of planning history that makes, almost as an aside, the very good point that a law of the indies (via vitruvius) type city, such as new york can be as dynamic and as lively as a medieval organic city, like london. more interesting he showed several cases where the city plan as diagram was not only irrelevant to the inhabited life, but where the inhabitants did not live in an overstrict society the diagrams tended to be erased as people moved homes and streets to more efficient/useful/comfortable configurations (have a look at circleville for an example).

not that you shouldn't have a go at the fractal game but you should be aware of the 2000+ years of similar attempts at diagraming the city and the very mixed results that went with them...in the end it is not the plan of the city that matters but the details of inhabitation that really does it, and that is a cultural issue more than anything.