[Yt-dev] Cookbook

[david collins]

I like the idea of providing a dataset and the output of the cookbook scripts.

> and then have that spit something out into stdout, which it'll grab
> off the web.  (I'm toying with different ways of doing this.)

I'm not sure I follow. You mean you'd somehow correlate what the
cookbook script is supposed to do with a url storing the correct
result?  That seems like it would add machinery that would make the
recipes less clear.  I do think that having the expected output on the
webpage with the script itself (stored both online and perhaps in the
docs directory) is good, but I'd almost rather have that a manual
process.

>
> Anyway, can we get some discussion of the right things to include in
> such a cookbook?  I'm thinking the following:
>
> simple slices
> simple projections
> simple phase plot of a sphere
> simple radial profile of a sphere
> cutting plane
> overplotting contours on a slice
> overplotting grid lines on a slice
> quiver plot

Things I've found essential that are a little more advanced, but I
think of fundamental importance and worth being in the cookbook

setting and using field_parameters
using add_quantity and the quantities dictionary.
Writing new callbacks
adding new fields, scalar combination
adding new fields, using derivatives
phase plots and profiles on non-spheres
accessing data from projection/profiles/slices/etc. to examine details
(for instance, summing the mass above a certain density:  easily done
with the output of the profile)
changing colorbar, xyz limits

d.