MC

How are you thinking of writing the user interface with the code? Are you thinking that a typical use is to run one model and store its outputs? That is, similar to how Brett's code is run? I think that's probably the right 'simple' usage, but I would also like us to think about running this code within an MC chain (perhaps MCMC, although it might be too slow). That is, I think the interface should allow one to easily run many instances, without having to generate 'parameter' files etc.. I think this should be easy to accomplish, just by allowing the default parameters or those from a configuration file to be adjusted on the fly.

Another thing to keep in mind from the beginning is that I think we should keep all ingredients to the chem-ev model very flexible. I.e., gas accretion, outflow, even yields as a function of (mass,Z,...) should in principle be adjustable (one thing I'm interested in exploring with this code is whether we can figure out the yields from a sufficiently large set of abundance measurements).

We should probably also immediately start writing analysis code, so we can easily visualize the outputs. An object like pynbody's SimSnap would probably be good.

For documentation I guess sphinx is still the right choice? I'm fairly happy with it, except that I think it doesn't work well with ipython notebooks and notebooks are a useful way to document the code...