[OSM-legal-talk] [Talk-ca] Nouvelle licence de données ouvertes au Québec

Mike Linksvayer ml at gondwanaland.com
Fri Feb 21 21:58:16 UTC 2014

On Fri, Feb 21, 2014 at 1:14 PM, Richard Weait <richard at weait.com> wrote:

> On Fri, Feb 21, 2014 at 10:26 AM, Pierre Béland <pierzenh at yahoo.fr> wrote:
> > Eh good news for  OSM-Quebec community then. Let's wait for the official
> > confirmation of the exact license adopted.
> I disagree.
> Any license drafted or adopted by a Canadian government, other than a
> no-restrictions, equivalent-to-Public-Domain-license, like ODC-PDDL,
> will require a waiver or clarification from the municipality (or
> province / territory, or feds) that attribution as provided by
> OpenStreetMap (wiki page, probably listed on a sub-page) meets their
> interpretation of "attribution".  So, adoption of CC-anything-but-0 is
> bad for local OSM communities.  It would likely work out okay in the
> end for those local OpenStreetMap communities.  To my knowledge, every
> municipality approached for such a waiver has granted it.  To
> OpenStreetMap Foundation at least.
> For the Open Data community at large, and for the municipality /
> governement itself, adoption of any restricting license is a disaster.
>  For one thing, not every potential open project will be on the radar
> of a municipality in the same way that OpenStreetMap is.  Too bad for
> that potential Open Data Project.  Perhaps they'll get the waiver they
> need, perhaps they won't.
> Again, any government open data publication in Canada must be licensed
> ODC-PDDL, or else it is a not-open-enough-closed-data-failure.

I agree that all PSI ought be public domain, with ODC-PDDL or CC0 or some
other public domain instrument, since the sane default isn't the default.
But calling attribution-only terms a closed-data-failure (BTW, what does
that make ODbL? Is OSM the only entity in the world that can use non public
domain terms and not be a closed data fail?) seems over the top.

Asking for a clarification that provided attribution is OK seems over the
top too, at least for CC-BY, especially CC-BY-4.0, given "You may satisfy
the [attribution conditions] in any reasonable manner based on the medium,
means, and context in which You Share the Licensed Material." If every
attribution needs to be clarified with the licensor to determine if it is
OK, then attribution licenses truly are a fail. But that practice is
certainly not the intent of such licenses.

IMO, IANAL, etc etc.

The remainder below is most excellent.

> Another sign of bizarre, Open-blindness.  I've had government open
> data representatives say to me, the equivalent of, "So what if the
> license says something complicated. It's open, just do what you want.
> We won't go after anybody who breaks the license. We just need to be
> able to shut down anybody who embarrasses us."
> Ahem.  No.
> 0) If you plan to grant wavers and exemptions anyway, why not just use
> an unrestricted license?  Oh, did you want to only grant exemptions
> for projects / persons of whom you approve?  That doesn't sound very
> open.
> 1) If you don't plan to enforce your license terms, why select (or
> worse, why draft) a license with restrictions?  Select ODC-PDDL
> instead.
> 2) If you want developers to work with your data, do you want
> developers who care enough to read, understand and follow your terms,
> or not?  Because your license with restrictions just cut out a portion
> of those developers.  You can still keep the developers that don't
> read licenses, or don't care about the terms.  Congratulations.
> 3) What, you want to shut down a use of the data that embarrasses you?
>  No.  It doesn't work that way.  If Open Data can be shown to expose
> that your mayor is a pathologically lying, bullying, drug addict with
> possible links to organized crime, you don't get to shut down the
> analysis just because your boss finds it embarrassing.  (It's just a
> hypothetical example)
> 4) If you really do plan to grant a waiver or exemption to every
> project / user who asks for it, shouldn't you have selected an
> unrestricted Open Data License that didn't place the burden of that
> extra waiver step upon you (and each potential user) ?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20140221/e694c9e1/attachment-0001.html>

More information about the legal-talk mailing list