Hi
Just to weigh in here with a point on 'robustness' and 'reproducibility', I was quite interested in this, so that once, when I was at Georiga Tech, and after I had explained the rudiments of axial lines to novice (no prior exposure to space syntax) MArch students, I performed a test. As an exercise, at the end of the class, I gave them all an underlying printed map (part of Atlanta) and an piece of over-head transparency clear-plastic, and they spent 10 minutes performing an axial break-up of the map, by hand (using over-head pens and rulers). We then combined them together and project the results onto the wall. I was duly surprised that with so little training, the results were so homogenous, with very little variation at the line-level and almost none at the meta-graph level. Naturally, with training, such variation would be reduced even further. This is just to support Alan's comment below about "being well defined and non arbitrary".
Unfortunately, I never published this, but there is an interesting study here for someone to reproduce! It would be especially interesting to test it against different, individual's interpretations of 'segments' particularly in more complex, non-rectilinear, urban environments. It would also be interesting to test learning effects - how much more robust do people become with training and how much is needed, and how this varies between different representations? With a large enough cohort (I did my test with about 20 students), we could have some hard data on this phenomena. A good master's thesis topic for someone perhaps?
Regards
Ruth
-----Original Message-----
From: [log in to unmask] [mailto:[log in to unmask]] On Behalf Of Penn, Alan
Sent: 27 October 2014 11:30
To: [log in to unmask]
Subject: Re: How to have a 'segment map' in Depth map
Hi,
just one point on your last line. As an approach to representation of the morphology of built space I have yet to find any approach that comes near that of space syntax in terms of being well defined and non arbitrary. I am thinking here of the axial line and derived segmental representations in which global angular and metric distance minimising routes in a network through two dimensions open space are guaranteed. If the analysis you with to run on the network representation is about angular or metric distance minimum routes, then this representation of the two dimensional open space morphology would seem to be prerequisite. The fact that the representation is defined well enough for automation using an algorithm should be enough to put a criticism of arbitrariness to bed, and so I find it odd that people still sometimes choose this as a criticism of the space syntax field. Part of the problem is in the nature of publication. Where an ill informed paper is published but subsequently responded to and all its criticisms addressed, it still remains in the public domain. People who come on the first without also seeing the response can be left with the impression that there are valid criticisms. Where both criticisms and responses are quite convoluted I can understand that people might find it hard to decide in their own minds on the validity of the arguments, but that is in the nature of science.
I always find it a bit difficult to know when to intervene in this discussion as I would not want to fall into ‘protesting too much’, and all that happens is people come back and discuss and it make the appearance that there is still a debate going on with some validity. Think of the creationist critique of the theory of evolution - this is in a similar camp.
Alan
> On 27 Oct 2014, at 08:47, SUBSCRIBE SPACESYNTAX Anonymous <[log in to unmask]> wrote:
>
> Thanks Professors Alen, Bin, Juan for your replies on this post. Sorry if I'm asking for bit of technical information on this post, but I would be appreciate if other people also could come on to the board, and mention the process through which they run an 'angular analysis'. For the purpose of this post, let's assume we decide to go with 'angular analysis;. So which steps you take (compared with those I mentioned below). Specially, what type of 'axial lines' map you use (traditional ones based on visibility drawing by HAND, Auto, or the one proposed based on natural roads'. I believe, this step would be important in kind of standardization and avoid arbitrary process in space syntax research (which many pick on this as an obvious weaknesses of our SS research).
>
> Thanks
|