nozmore

Just to tease the comments apart, while the last reply applies to this PR I think it was more in response to the "use dictionaries for everything" comment. For this...

The entire last comment was in response to your earlier comment. I am thinking of this from my perspective as a member of a Product Security Team where I would...

haha well I forgot about the book. Either way that discussion is not really related to this PR, only in response to a comment I made about this being even...

Yes, I meant to mention that in the description, the redundancy is another reason I had this as draft. We haven't used methods for assignment in tm.py before so I...

> Source could also be an author, so that you have someone you can contact? > It would potentially be useful to know the source of a finding as automatic...

> Findings should have common fields such as summary, description (long or short), etc, and should probably also contain references or some field to support a contact url or email...

Ok. I think I incorporated all of the comments. Only thing I didn't was @colesmj comment about do the findings need to be stored in the model. No its code...

> > When I take a look at json support I would like to separate the input (model, manual findings) from the output (all findings) > Why is this separation...

Maybe but not currently. Command line arguments or shell/ui interface could be an incoming Dataflow, this does not make it a Server. I actually thought of this as a 3rd...

I updated the option list above with with the one I left out and @nineinchnick referenced.