Re: QD (Quad-Double) Approach to Robustness

Martin Davis wrote:
>> But Dear Friends, it would be humorous if GEOS should use real GC :-)
>> Then GEOS existence has completely no reason, because JTS is there.
> 
> Not at all. A system like PostGIS does not want to incorporate a JVM.
> Also, calling Java code from C/C++ is painful, to say the least.
> 
> Now what you could say is "Why not just use GCJ?".  Apart from the
> limited platform support, this seems like the best way to go to me. 

Yes, GCJ is what I had in mind saying "JTS is there".

Cheers
-- 
Mateusz Loskot
http://mateusz.loskot.net