perfsonar development work

["Jeff W. Boote" <>]

maxim wrote:
> but thats the point. there is no need to be specific.
> it can be timeStamp. 

And if in the future we come up with another select parameter that is also time 
related? Perhaps one that is used to select only sub-sessions of a specific 
duration within the given time range? Or perhaps one to select only data of a 
specific precision?

I'm not saying these particular examples are likely - only that it is pretty 
easy to think of cases where you would want more than one parameter in a 
parameters list that should be of type nmtime.

I simply prefer to have more meaningful parameter names. I fully agree that we 
should handle the same conceptual things in a common fashion. I do not agree 
that we should create generic instance names for them.

jeff