Request for approval: EUPL (European Union Public Licence) legal analysis

[Donovan Hawkins]

On Mon, 17 Mar 2008, Schmitz, Patrice-Emmanuel wrote:

>" The European Commission may put into force translations and/or binding 
> new versions of this Licence, so far this is required and reasonable. 
> New versions of the Licence will be published with a unique version 
> number. The new version of the Licence becomes binding for You as soon 
> as You become aware of its publication."
...
> The third paragraph in this Article provides for the application of of 
> authorised translations and new versions adapted to law as required by, 
> for instance, different jurisdictions. Note that new versions that 
> automatically apply cannot add restrictions to freedoms granted by this 
> Licence as this Licence is granted (in Article 2) for the entire 
> duration of the copyright in the Original Work.

That may be the intent, but that doesn't seem to be what it says. It says 
the new version becomes binding for me as soon as I become aware of it. 
The license duration may continue past the release of the new version, but 
it is the terms of the new version that I'm now operating under.

If that isn't how you meant it to read then I think some rewording is 
in order, but you said in a previous message:


> "required" means that the European Commission may update the licence to 
> address new legal or technological issues that would otherwise keep the 
> licence from functioning as intended

That clearly indicates that new restrictions will apply in the new 
version, the way the restrictions against Tivoization were added to GPL 
v3 to address what the FSF considered a bug in v2 (and what Torvalds 
considered a feature). Preventing what you consider to be techincal 
loopholes in the license means adding restrictions...retroactively in your 
case but not in the GPL case.


IMHO, it would pretty seriously undermine confidence in open source 
licenses if some company were to rely on being able to use the software in 
a way clearly allowed under the license, only to have it retroactively 
changed on them because that wasn't what the author of the license 
intended to allow. What is the point of having carefully drafted licenses 
approved by the OSI if the abstract concept in the minds of the authors is 
what will ultimately become legally binding?

Requiring that any new license be approved by the OSI would solve the 
OSI's problem but not the company/person who doesn't want the license 
changed out from under them into something they can't use.


Ultimately, I think the best solution to what is trying to be addressed is 
to just fix the loopholes when you find them for future versions. It's not 
perfect, but neither is your solution. You could add a clause like GPL's 
standard "v2 or greater" to mitigate the disruption caused by new license 
versions without retroactively revoking anything.

---------------------------------------------------------------------------
Donovan Hawkins, PhD                 "The study of physics will always be
Software Engineer                     safer than biology, for while the
hawkins at cephira.com                   hazards of physics drop off as 1/r^2,
http://www.cephira.com                biological ones grow exponentially."
---------------------------------------------------------------------------