It's still a mystery to me how it could have been tickled in the first place.  Could you personally send me the copy of sim/portrayal/geo/GeomVectorFieldPortrayal.java that you are using?  Or your geomason.jar file?  And I do suggest you take Mark's advice and use the SVN repository version of GeoMason in general.

Sean

On Aug 14, 2013, at 10:01 AM, Norman Langner wrote:

> Hi,
> 
> no problem for me. I know that felling too well, as I am currently near the deadline for my thesis... 
> 
> That is the reason, why I didn't answer till now. Sorry for that. I found the problem. It was a mistake by me. I extended my agents, which implement Steppables to MasonGeometries but forgot to add the geometry when using them. I had -before I changed them to extend the MG-class - a Point-Method which sets and gets the location, but the mason/geomason methods work with the getGeometry from the MG - and this i forgot to add to my own set-method.
> I hope it was understandable what I've tried to say with the sentences above ;)
> 
> @Sean Luke your post helped me to find my error. Anyway thanks for your support, keep up the good work and good luck with the phd! 
> Norman
> 
> Am 14.08.2013 03:37, schrieb Mark Coletti:
>> On Mon, Aug 12, 2013 at 6:20 PM, Sean Luke <[log in to unmask]> wrote:
>> Norman, I am afraid to tell you but our two primary maintainers of GeoMASON (Mark Coletti and Keith Sullivan) are both rushing to finish out their PhDs.  So I'm not sure how soon they'd get back to you on this.  This is clearly a GeoMASON thing and not a MASON thing.
>> 
>> Sadly I must echo what Sean said: I'm deep in the throes of dissertation crunch mode.  I may have a narrow GeoMASON maintenance window when I turn in a draft sometime within the next couple months where I can address pending problems.
>>  
>> So that code should have triggered a NullPointerException *first*.  Maybe you've got different code from me and so your line 325 isn't what I thought it was?
>> 
>> 
>> Could be he's running from a jar file from an earlier version and not the latest from subversion.
>> 
>> If anything, I'd urge anyone reading this to prefer using the bleeding edge subversion problem if you encounter GeoMASON problems because a bug fix may have already been committed that repairs your problem.  (And, yes, I need to cut a release, but that's frankly going to have to wait.) 
>> 
>