Remove big files from the repository

[chrisgorgo]

At the moment due to big files the repo is quite heavy - 124 megs! You might want to consider splitting example data from code and cleaning history: http://blog.jessitron.com/2013/08/finding-and-removing-large-files-in-git.html

[jamesgao]

Yeah, we're having this dialog with Yarik Halchenko earlier in #42. There's no standardized spot to put important example and test data. Do you have any suggestions for this?

[chrisgorgo]

What if you put the data on dropbox or on the university servers and add a function that will download it and populate the local filestore? You could call this function before running tests and add it to documentation so user could play with some examples. I think nipy is doing something similar.

Speaking of datastores. I think it would make sense to default to a filestore in the user writable directory ($HOME) instead of system wide one.

Those are just small kinkd to iron. Otherwise it is a fantastic viewer.

[andrewolaughlin]

On the other hand I'd like to see all the data from the freesurfer subject folder be made available too, if possible.

And while you're at it, the blender files too to see exactly where your cuts are made