[web2py] Re: Speeding up model execution for large models

[pbreit]

I wonder if conditional model loading would be best performed with either 
decorators or, dare I say it, importing (if either is even possible)?