Dear Andrea,
As Alan suggested, it is in these cases when you really need to make decision whether to use convex analysis or other types of representations like VGA. I think he explained recently on this list why it is very difficult to find near-optimum and unique solutions for convex break-up maps. Perhaps one way to do it is through identifying the largest non-overlapping cliques in a visibility graph, but I have not done this myself so I will leave it to others.
The decoding of convex spaces in the layout you have attached was done in an arbitrary way, so there is perhaps more than one way to do it, and it all depends on what you want to convey through your research question.
The manual you have is an older version of a textbook we teach on our MSc SDAC at the Bartlett, UCL. Alan Penn, Bill Hillier and I are currently working on the final version of this textbook, which will be published later this year. We hope to cover this issue there more extensively.
Best regards
Kinda
This is the case which shows why you might decide not to use a convex representation !
Alan
> On 25 Feb 2015, at 18:25, Andrea Vieira <[log in to unmask]> wrote:
>
> Dear professor Kinda
>
> I am encountering difficulties in the definition of convex spaces that between two circular walls.
>
> In pag 10 from
> Al_Sayed , K. , Turner , A. , Hillier , B. , Iida, S.,
> 2014 ( 2nd Edition) , "Space Syntax Methodology "
> there is a picture appearing to deal with this problem
> As author of the manual space syntax can you give some advide or reference on how should this be done in these situations?
> <image.png>
>
>
> --
>
>
>
> figure page 10 from Al_Sayed , K. , Turner , A. , Hillier , B. , Iida, S., 2014 ( 2nd Edition) , "Space Syntax Methodology
>
>
> <FAUP museum.jpg>
>
> best regards
> Andrea de Pera Vieira
> Andrea de Pera Vieira
> <FAUP museum.jpg>
|