Scitools licensing guide

[pelson]

A heads up that I'm working with our legal team to simplify the licensing and CLA signing process of all SciTools libraries.

The ultimate aim is that all SciTools will be re-licensed as BSD 3-clause, rather than LGPL (we have received some highly ill-informed feedback stating that Iris is GPL and therefore couldn't be chosen for a project, rather than try to educate them, we have chosen to be more aligned with the licensing of other Python libraries (e.g. pandas)).

To limit the scope of this PR, I'm only documenting how we license SciTools projects here. I will create a similar PR for CLA and CLA signing process once this is done.

Take a look at this in rendered form at https://github.com/pelson/scitools.org.uk/blob/scitools_licensing_guide/documents/scitools_licensing.md.

[pelson]

One of my goals is to ensure that GitHub (and other licensee based tools) can identify the license of the projects automatically, hence my desire for the LICENSE file to remain unmodified.

Note how GitHub can tell the license of pandas, but not numpy, even though they are both BSD:

[pelson]

I'm going to start moving forward with this over the next few days.

I need to remember to document https://script.google.com/macros/s/AKfycbwqZ0JLlOKJEcvYUYFXo77o6nux9tSHvOJISUnk3yHi-Bk0Qanz/exec

[pelson]

Confirming that I have now had legal advice on this proposal, and we are 👍 to go. @SciTools/steering-council, please merge if you are happy (thus exercising your vote and forming a council majority).

[ajdawson]

A unicode character here can be a problem for Python 2.7. For example, a trivial hello_unicode.py:

# Copyright © <YEAR_START>-<YEAR_END> <project> contributors
print("hello")

Executing with Python 2.7 we see:

  File "hello_unicode.py", line 1
SyntaxError: Non-ASCII character '\xc2' in file hello_unicode.py on line 1, 
but no encoding declared; see http://python.org/dev/peps/pep-0263/ for details

It can be fixed by declaring the source encoding as utf-8. We have options:

1. Don't use this symbol
2. Use it and declare encoding on all files (yuck)
3. Use it and don't care about Python 2

For me the only practical choice is 1, especially since I'm not aware that the symbol has any legal significance in modern copyright law.

[pelson]

This is a fair reflection of the situation. I agree with just dropping the © altogether (not replacing with (C)).

[ajdawson]

Resolve use of unicode copyright symbol.

[pelson]

Thanks @ajdawson. Have made the suggested change.

[ajdawson]

This usage of © should also be removed.

[pelson]

Done. 👍