[Python-legal-sig] Is CLA required to send and accept edits for Python documentation?

Ben Finney ben+python at benfinney.id.au
Thu Aug 15 04:08:10 CEST 2013


Jesse Noller <jnoller at gmail.com> writes:

> On Aug 14, 2013, at 7:53 PM, Ben Finney <ben+python at benfinney.id.au> wrote:
>
> > So what is the difference that means Wikimedia Foundation do not ask
> > for additional agreement documents, while PSF do ask for additional
> > agreement documents from the contributor?
>
> Python is distributed to end users in packaged form, on operating
> systems and elsewhere. Wikipedia is distributed online and is done via
> mass collaboration.

Both of those are true for both Python and Wikipedia:

* Python is distributed online via mass collaboration, in the VCS
  repository.

  What salient legal difference is there from Wikipedia's mass
  collaboration and online distribution?

* Wikipedia is distributed to end users in packaged form (for one
  example of many, the WikiReader device). Indeed, this is a primary
  purpose of Wikipedia, to produce an encyclopedia useful for packaging
  and distribution under free-software terms to those without reliable
  internet access.

  What salient difference is there from Python's distribution to end
  users in packaged form?

I'm not asking for you to defend what Wikimedia Foundation have decided.

But I am asking for why the PSF requires additional agreement documents,
when other free-software organisations, performing what seem to be
legally-equivalent collaborations and redistributions, do not require
these additional agreements.

If there's a salient difference, I have yet to have it presented. If
there's not a salient difference, I don't see why the CLA is required.

> It would be next to impossible to prove legal provenance of the
> changes on Wikipedia. This means distributing the text in any
> commercial form is legally questionable.

Yet it is explicitly permitted by the license. The license was carefully
chosen to encourage commercial redistribution of Wikipedia, and people
do it.

> For python: we have to (as the PSF) be able to prove that the people
> committing the code have rights to that code.

Okay. That's not a reason to ask for a licensing agreement from the
contributor, though. It's reason to ask for an affirmation of the
provenance of the contribution. No additional powers required.

> For example, if I, working for Foo, submit code to core, I must have a
> CLA in place from that company stating from the company and myself
> that I have the legal rights to submit that code and it is OK for the
> PSF to redistribute that code.

Why is this a special agreement with the PSF, though? The PSF already
has full permission to do everything the Apache License allows.

> > If that permission is already in the license on the contribution,
> > why does the PSF require it again in a special agreement document?
>
> The special agreement document is the contributor agreement which
> companies, employees, and individuals sign to agree to license the
> changes under the Apache 2 license.

I would be happy to sign such a document, asserting the work is licensed
to all recipients under the Apache License version 2. It says nothing
about any special arrangement with PSF.

-- 
 \          “Ocean, n. A body of water occupying about two-thirds of a |
  `\     world made for man — who has no gills.” —Ambrose Bierce, _The |
_o__)                                        Devil's Dictionary_, 1906 |
Ben Finney



More information about the Python-legal-sig mailing list