Alan & Noah,

I have just one comment on the effect of weather on social behavior in
outdoor environment. As I understand from the theory of space syntax and the
concept of "natural movement" that spacesyntax model only counts for the
predicted "percentages" of dynamic activity, which is movement. So if any
effect would happen due to weather it would be on the global scale with
equale impact on all segements, resulting in the same percentages but
ofcourse with diffirent "values". So I agree with Alan in the importance of
recording the weather while observing the site in order to make any
adjustments due to diffirent weather situations.
In other words, I believe that if Space Syntax was able to predicte the
percentages of movement in a good weather it would be the same in bad
weather (ofcourse apart from any affects due to special weather treatments
in some paths like archades, trees, ect..)

Regards,

Mohamed Salheen

>Noah,
>
>I'm not sure that Ive seen any responses to your questions so I thought I'd
>have a go.
>
> >The first question is a logistical one and has to do with the
>capabilities
> >of Axman, Pesh, etc. Does this software enable me to perform the kinds
>of
> >predictions I discuss below? Will I be able to conduct isovist analysis
> >using this software? Are there any barriers to this prediction which I
> >may be missing?
>
>The answer is that Axman will do a lot of what you need, pesh will be
>possibly of less relevance, but my guess is that you really need isovist
>analysis if as I understand it your focus is on interactions within the
>main body of open square type spaces. Isovist integration will
>differentiate between different internal locations in a large convex space.
>There is a download site at www.vr.ucl.ac.uk for the isovist program for a
>Silicon Graphics (finding a computer should be possible somewhere in
>Brown).
>
>Whether the programs will 'predict' is a research question and needs to be
>evaluated in your own context. I'd be fairly sure that movement flows will
>be predicatble from axial integration, given an appropriate model boundary,
>but the focus on detailed aspects of behaviour and interaction within open
>public space is not well understood, so you will be doing new stuff.
>
> >
> >Second, how does one control for the effects of weather? Has any work
> >been done on the effects weather has upon behavior in outdoor spaces?
>Any
> >recommendations?
>
>What we know about weather is that it does not seem to affect movement
>rates much (except for in downpours) but it does affect static behaviour. A
>fair amount of work was done on this on open spaces in the City of London
>for the Mansion House Square public inquiry a few years back by Bill. It
>would be interesting to know if this is reproduced in RI - another research
>question really. The thing to do is to record weather during observations
>and to control for it in your statistical analysis.
>
> >
> >Finally, do you think the time I have proposed to complete this project
>is
> >reasonable?
>
>The project overall looks very interesting, but ambitious (how much time do
>you have?). Qualitative research into behaviour is very time consuming and
>I am assuming that you have local support for this side of the methodology
>and interpretation of your data - you will need it.
>
>I suspect that your hypothesis - that 'integration = impersonal' is wrong -
>but it depends on a precise definition of what you mean by 'impersonal'. I
>would expect greater levels of interaction of all sorts in more integrated
>spaces. In order to look at this means that you need to put a good deal of
>effort into coding of behaviours to translate observaions into different
>classes that may vary on the personal-impersonal axis before you look for
>any configurational correlates. Finally, a word of warning on campuses -
>these have quite different compositions of users (students) and may well
>behave in a qualitatively different way to general public spaces.
>
>Best of luck and keep us informed on progress.
>
>Alan
>
> >Dear list,
> >
> >My name is Noah Raford and I am planning to conduct a space syntax
> >analysis of public space in Providence, RI, USA for my honors thesis. I
> >have already contacted several of you with specific research questions
>and
> >I am very interested in the comments of the general list. Included below
> >is a copy of my research proposal. I must admit that there are no
> >professors at my school with space syntax experience, so this may be a
> >somewhat different approach towards applied research than the UCL takes.
> >Please take this into consideration when you read it and I would greatly
> >appreciate any comments or suggestions you may have.
> >
> >In specific I have three questions which I feel that my proposal does not
> >address:
> >
> >The first question is a logistical one and has to do with the
>capabilities
> >of Axman, Pesh, etc. Does this software enable me to perform the kinds
>of
> >predictions I discuss below? Will I be able to conduct isovist analysis
> >using this software? Are there any barriers to this prediction which I
> >may be missing?
> >
> >Second, how does one control for the effects of weather? Has any work
> >been done on the effects weather has upon behavior in outdoor spaces?
>Any
> >recommendations?
> >
> >Finally, do you think the time I have proposed to complete this project
>is
> >reasonable?
> >
> >Thank you in advance for your time and I look forward to your thoughts.
> >
> >Yours,
> >Noah Raford
> >Brown University
> >Providence, RI, USA
> >
> >***********
> >
> >Experimental Design:
> >A Sociospatial Analysis of Providence Public Spaces
> >Noah Raford, November 29, 1999
> >
> >
> >Hypothesis:
> >
> > I hypothesize that there is a specific, quantifiable link between
> >the physical configuration of public spaces in Providence and the social
> >uses and behaviors which occur in them. I speculate that spaces which
>are
> >more integrated in the urban network will receive more public usage and
> >will accommodate activities which are more likely to be characterized as
> >impersonal, public behavior. Conversely, spaces which are the least
> >integrated will receive less public usage and the activities which occur
> >there will be of a more intimate, personal nature. I also speculate that
> >within each space, this same relationship will apply when focusing on the
> >specific location of different types of behaviors.
> >
> >Method:
> >
> > Phase One: Prediction
> >
> > To evaluate this hypothesis, I will create a space syntax model of
> >Downtown Providence and its surrounding neighborhoods. This model will
> >generate predictive values for usage based on each space's physical
> >integration and configuration. I will then choose four sites for
> >analysis, one with a high integration value, one with a low integration
> >value, and then two in-between. Currently, I plan on using Kennedy Plaza
> >as the high integration value site, Cathedral Square as the
> >low-integration value site, Brown's campus as the third, in-between site,
> >and the fourth will be chosen based on results from the model. This
>model
> >will estimate relative volumes and types of usage based upon the
>placement
> >of the site in its urban context. These use magnitude values will form
> >the first part of my analysis.
> > The second part will focus on each specific site and will attempt
> >to predict the location and types of different activities within the
>site.
> >I will create a detailed map for the site, with all it's fixed features,
> >and then conduct an axial line analysis, convex space analysis, and
> >isovist analysis to create integration values for specific areas of the
> >site. These values will be used to create a predictive activity map
>based
> >upon the physical shape of the space.
> >
> > Phase Two: Observation and Data Gathering
> >
> > Once the predictive analysis has been made, I will begin field
> >verification and data gathering. The first step of this phase will be to
> >create a detailed plan of each site. Then, using video cameras, I will
> >record four 18 hour segments of each space. Two segments will be
>recorded
> >on a weekday, two will be recorded on a weekend. It will be supplemented
> >with two in-person observations per site, one on during the week and one
> >during a weekend. Together, this raw data will capture both the number
>of
> >people in the space as well as how they use the space. It will thus
>cover
> >the objective, observable behavior component.
> > I will then conduct 30 longitudinal observations of individuals
> >within the space, tracking their behavior for 20 minutes. To gather data
> >on the more subjective, individual components of space usage and
> >perception, I will then conduct short a survey with each of these same
> >individuals. The variables gathered will be: proximity of residence,
> >proximity of work, race, age, sex, self-reported personality type,
> >familiarity with the space, their purpose for being there, their
> >impression of how 'social' the space is, and how often they see people
> >they recognize in the space. By combining these individual variables
>with
> >the gross scale movement and activity patterns recorded earlier, I will
>be
> >more able to posulate a link between these scales and variables.
> >
> > Phase Three: Analysis
> >
> > The first stage of analysis will be to construct an activity map of
> >each space. Using the plan constructed in Phase II, I will divide the
> >space into a one meter grid and graph the average location of different
> >classes of activities. This will then be statistically compared to the
> >predicted values created in Phase I to test my estimates of activity
> >intensity. Then the type of activities will be classified on a scale of
> >the most private to the most public, graphed on the map, and compared to
> >the predicted map to test if there is a relationship to the type of
> >activity. Variables measured will be the number of interpersonal
> >interactions, the length of interactions, group size, and the type of
> >interaction (reading, buying hot dogs, ice skating, people watching,
> >etc.). These variables will be compared between spaces to test if the
> >urban integration values can make accurate estimates of the volume of
> >activity. Finally, each of the variables measured in the survey will be
> >statistically compared to the activity map to determine if these
>variables
> >are significantly associated.
> >
> >Timeline:
> >
> > Once the software is received, the modeling will take approximately
> >one week. Research into the possibility of time lapse photography will
>be
> >conducted and officials will be contacted for the placement of the video
> >cameras before the semester ends. Once second semester begins on Jan.
>23,
> >I will be able to gather the video footage and conduct the surveys. This
> >should take approximately three weeks. Analysis should take no longer
> >than two weeks, and the remaining weeks will be allocated to drafting the
> >final report.
>
>
>________________________________________________________
>Alan Penn, Reader in Architectural and Urban Computing
>Director, VR Centre for the Built Environment
>The Bartlett School of Architecture and Planning
>1-19 Torrington Place (Room 335)
>University College London, Gower Street, London WC1E 6BT
>tel. (+44) 020 7504 5919 fax. (+44) 020 7916 1887
>mobile. (+44) 0411 696875
>email. [log in to unmask]
>www. http://www.vr.ucl.ac.uk/
________________________________________________________




Mohamed Salheen
Tel: 0044 131 221 6280
Fax: 0044 131 221 6157
Edinburgh College of Art,
79 Grassmarket Campus,
Building 6

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%