Congratulations Nick,
I think this finishes the discussion about open source in space
syntax. The father of space syntax software started a (real, meaning
free software) open source project :)
I hope the project succeed and you turn to be the father of the open
source space syntax software as well :)
Best Regards,
Lucas Figueiredo
On 08/07/07, S. N.C. Dalton <[log in to unmask]> wrote:
> Hi , given all the talk recently about open source space syntax, I've
> decided to create a truly open source project under the GPL 3
> licence. This is more focused on the more traditional areas of axial
> analysis.
>
> The initial version can now be found on
>
> http://sourceforge.net/projects/spacesyntaxmap/
>
> Along with links to documentation, CVS repository access and a binary
> compiled version. I was inspired by the CFD mechanism were GUI's are
> reserved for tools of known authenticity and command lines tools were
> reserved for experimental development. This way it if you get a copy
> of webmap you know it is a unadulterated version. If you want to do
> something new and original you can use the spacesyntaxmap code base
> to develop a command line tool which you can give/sell to others*. If
> a user is smart enough to use a command line tool they should be
> smart enough to test it to make sure it is giving correct results
> rather than just making the assumption. naturaly I cannot prevent
> someone developing an entirely separate GUI based on the source
> provided but this is another matter and unlikely to cause confusion.
>
> This code/binary is written in Java version 5.0 and will run on a
> number of platforms including Mac,PC,PC-Windows,SunOS, Linux and so
> on. The source uses templates which I hope will make the code more
> modifiable and better engineered against digital disaster.
>
> The current code is largely focused on axial lines, but the
> fundamental graph class can be used to compute a number of
> integration values for example for a social network or for Isovists.
> Things are fairly basic as they stand - you can open a .lines
> or .axmn file process it and get a printout of the standard measures.
>
> Most importantly the class Ring.java has all the relitivisation code
> so if someone wants to know how to relativise they can look at this
> class for a gold standard.
>
> I'm not sure how much people really want a good Space syntax library
> and how much people just enjoy reinventing the same code again and
> again. If I get one or two developers signing up on source forge I
> promise to significantly enhance the code base beyond the basics I've
> creating here (adding segments and angularity to the code base, addin
> in DXF and .AXL format readers, EPS and .SVG output and so on, ).
>
> Source forge has tools to track bugs, add new features and track the
> code base automatically ( via subversion or CVS), it has help fora
> and discussions areas and documentation zones so if you need more
> best to try there than the mail base.
>
>
> have fun and if you build anything cool please tell me.
>
> sheep
>
> *with source see GPL3.
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> ps In Computational fluid dynamics (CFD) the creation of the modle is
> highly operator dependant - using to small cells makes the
> computation to long (or go bad) using to big cells makes the results
> inacurated. No two CFD modellers will make a CFD model exactly alike
> (I'm told). Given the inablity to define a CFD 'cell' unambiguously
> does this make CFD outputs useless or unscientific? Unnerving to
> think that the automatic generation of minimal line axial maps puts
> space syntax on a sounder footing that CFD :-)
>
--
Lucas Figueiredo
http://www.flickr.com/photos/lucasfigueiredo/
Mindwalk
http://www.mindwalk.com.br
|