Create a prototype application using current REST API of Scidash to connect to Geppetto as a service to run simulations.

[jrmartin]

Rick Request:
One thing that I realized would really help test a wider variety of models, and require less on-going support or new code from my end, would be to use Geppetto as the backend, much as you have it as the backend for the current SciDash app for running simulations. And when I say "Geppetto" what I really mean is the whatever NeuroML parsing you guys have done (in Java?) that allows you to run lots of NeuroML models, inspect variables, etc. As you may recall one of the selling points of sending the NeuroML to "the simulation server" was that you guys already know how to do that stuff and have been doing it for other projects. But the current "Geppetto" support only exists through the GUI, so there is no way to do the same thing if I am locally testing a project and want to use the API... and yet in NetPyne UI there is a notebook interface and perhaps there is some command that "sends the simulation to the Geppetto server"?
If so, is there any way I could do the same thing? The "Geppetto server" could be another app hosted in my MetaCell cloud services account, and I could send simulations there and get results back. And I wouldn't ask for any changes to it or new features (i.e. what it could simulate), since I'm sure the number of things it can successfully simulate is far more than I can currently do with NeuronUnit's crappy Neuron backend. I would be happy to just learn it's API and use it.

Zoran's proposal, which is what this prototype needs to do:
I think I already mentioned it a couple of times before. One option I see is to use the current REST api of SciDash. The frontend application is using this api to create/update tests, models and also to schedule a simulation.
So using the rest api (json based) we can do everything what the frontend does.