[SciPy-dev] Starting a datasets package, again

[David Cournapeau]

Robert Kern wrote:
>
> If you want a declaration from me, I would say that the surrounding text and
> code in scipydata packages should always be under the BSD license. This should
> be noted using the "License :: OSI Approved :: BSD License" classifier in the
> setup script and in a *comment* in the code following the copyright notice.
> However, the copyright notice and license should be accompanied by a note that
> the data does not fall under this license or copyright and the metadata to look
> at to find the status of the data. I'm not good at legal boilerplate, but
> something like the following would be fine, I think:
>
> # The code and descriptive text is copyrighted and offered under the terms of
> # the BSD License from the authors; see below. However, the actual dataset may
> # have a different origin and intellectual property status. See the SOURCE and
> # COPYRIGHT variables for this information.
> #
> # Copyright (c) 2007 Enthought, Inc.
> #
> # Redistribution and use in source and binary forms, with or without
> # modification, are permitted provided that the following conditions are met:
> # ..., etc.
>
> David, thank you for pursuing this with the care that you have, and thank you
> for bearing with my long-winded pontificating while you do all of the actual
> work.  :-)
>
Ok, I checked in the new old faithful data (temporary in pyem), as 
exactly copied from Azzalini's reference. Does that look Ok to you (from 
a copyright point of view) ? If yes, I will use this as a template for 
all other datasets I will convert:

http://projects.scipy.org/scipy/scipy/browser/trunk/Lib/sandbox/pyem/data/oldfaithful/data.py

David