For approval: MXM Public license

[Chuck Swiger]

Hi, Nigel--

On Apr 9, 2009, at 8:36 AM, Tzeng, Nigel H. wrote:
>> From: Matthew Flaschen [mailto:matthew.flaschen at gatech.edu]
>> Tzeng, Nigel H. wrote:
>>>> There is no way to produce a FOSS license that explicitly denies  
>>>> patent
>>>> rights.
>>>
>>> Mmmm...not providing patent rights is the same thing in as much as  
>>> the
>>> result is the same.
>
>> Right, and every OSI-approved license provides patent rights, not
>> withstanding the efforts by some to muddy the waters.
>
> If this were absolutely true (the waters were not muddy) then there
> would not have been a second (or third?) generation of OSI-approved
> licenses with explicit patent grants.
>
> If implicit patent grants were crystal clear and provided all the
> protection required for open source why then almost all modern open
> source licenses have explicit grants?

Modern open source licenses are now being written and/or reviewed much  
more carefully by lawyers and by organizations such as the FSF and OSI  
who have more experience with licensing, then the folks who wrote  
something like the BSD or MIT licenses.  I happen to think that  
something like "Permission is hereby granted, free of charge, to any  
person [ ... ] to use, copy, modify, merge, publish, distribute,  
sublicense, and/or sell copies of the Software, ..." is completely  
clear that the copyright owner has conveyed all necessary rights to  
use the software, including those related to any patents which might  
apply (to the extent that the copyright owner is able to convey such  
rights, anyway).

However, just because I have the above opinion, does not mean that a  
diligent lawyer who wishes to make things clear, would not want to use  
more explicit terms of the legal art-- such as the Academic Free  
License clause 2, "Grant of Patent License" or similar....

>>> The question is whether OSD #1 requires patent grants
>> No provision of the OSD /explicitly/ requires any specific grant of
>> intellectual property.  There is no mention of copyright, patent, or
>> trademark in the OSD.  So saying, the OSD doesn't require patent  
>> grants
>> makes no sense. It doesn't require copyright license grants  
>> explicitly
>> either.  In practice, /both/ copyright and patent rights must be  
>> granted
>> in order to ensure the actions listed in the OSD are possible.
>
> Perhaps.  However, there appears to be licenses in the past without
> explicit grants of patents and explicitly do not grant trademarks.
>
> Also, there has always been some "muddied" waters with respect to  
> how far
> some explicit patent grants in some older OSI approved licenses go  
> with
> respect to forks.
>
> The caveat "The patent license shall not apply to any other  
> combinations
> which include the Contribution" in some licenses (mostly retired?)
> has caused some discussion in the past I think.

The main examples Google can find are the Eclipse Public License, and  
the Common Public License.  (And Plan9.)

>>> Eh...ultimately an ISO reference implementation will get used with  
>>> or
>>> without OSI approval as "open source".
>
>> If the submitter doesn't care about OSI approval they're free to  
>> leave.
>>  But OSI shouldn't weaken its standards on the grounds of
>> "inevitability" (though it's far from inevitable that a given ISO
>> standard becomes popular).
>
> I'm not suggesting that.  I'm simply saying that the OSI has been
> acknowledged as the arbiter of open source by ISO which should be  
> giving
> some folks (at least those not so self-righteous in their indignation
> that someone would dare submit a license not up to spec) warm fuzzies.
>
> I would hope that the OSI would take the opportunity to say "yes,  
> sorry,
> this isn't open source but perhaps we should think about a category of
> "reference software licenses".  This might be helpful to the  
> community.
>
> It might not.

Well, I think that the FSF http://www.fsf.org/licensing/licenses/ page  
is helpful, although it is written from a perspective that I'm not in  
a position to always agree with, at least as far as my day job is  
concerned.  :-)

> However, Creative Commons has a wide range of licenses.  Some would  
> meet
> the OSD,  some, like the noncommercial share-a-likes, would not.   
> While
> they don't step on OSI's toes by doing software licenses, if the OSI
> simply shows ISO the hand they might go to CC and ask for a wrapper on
> their site like for GPL v2.0, LGPL and BSD to clearly explain to  
> developers
> their rights and responsibilities under the MXM license.  Then  
> display a
> CC logo.
>
> That's probably almost as good for them as OSI branding.  Yep, they  
> sure
> are free to leave.  Do you really want them to?

Branding is mainly valuable for the sake of product identification and  
labeling, so that people know what to expect when they see something  
like the OSI logo, but the value of branding is compromised if it is  
used for things which don't actually meet the standards which users  
expect.

I don't have a preference whether the MXM or another submitted license  
under consideration receives OSI approval; my concern is whether the  
license meets the OSD in terms of the letter and spirit or intent, and  
whether the license is well-crafted in terms of being understandable  
to someone with some background in computers rather than law.

In particular, I don't have any objection to the ISO folks deciding to  
make their license free for non-commercial purposes, but require a  
separate patent license for commercial use.  However, if that is what  
they want to do, that circumstance is not "open source" per the OSI  
and as is commonly understood by the larger community, and therefore  
should not receive OSI approval.

Regards,
-- 
-Chuck