[Geowanking] Dissertation Advice

[Ian Turton]

On Wed, Mar 11, 2009 at 5:23 PM, Eric Wolf <ebwolf at gmail.com> wrote:
> FYI: http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=4782703
>
> Sort of antithesis to WPS.

Sounds interesting from the abstract (I need to be at work to read the
full text). But doesn't the inability of the browser to manage more
than about 100 features at once kind of kill this off. And that's
before you try to make the browser do the any hard work like
generalization.

>
> What I'm playing with is creating a distributed architecture for web
> mapping that, on the back end, handles petabytes of data but allows
> for insanely high traffic bursts. WMS and WPS don't quite hack it. So,
> you push the basemap rendering back further away from the user and any
> analysis and computation closer towards the user.
>

Well caching and tiling would seem to solve the basemap problem fairly
easily. How you handle the analysis remains an interesting and open
problem. If you want to make it a bit more exciting throw in
asynchronous and/or different place collaboration.

Ian


-- 

Ian Turton
http://www.geotools.org
http://pennspace.blogspot.com/