[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian Legal summary of the X-Oz License

On Tue, Mar 02, 2004 at 04:05:22PM -0800, Ben Reser wrote:
> On Tue, Mar 02, 2004 at 04:12:45PM -0500, Branden Robinson wrote:
> > Copyright doesn't cover usage, it covers copying.
> You know what I meant Branden.  My emails aren't a license.  Please
> don't nitpick them when you know what I mean.

Well, no, I didn't know what you meant, actually; I try to make as few
assumptions as possible, especially when it comes to the opinions
people who have only recently started contributing to the debian-legal

> > Any attempt by a copyright license to prohibit protected expression is
> > probably unenforceable (and unconstitutional if prosecuted as criminal
> > copyright infringement by the U.S. government).  Since the Debian
> > Project does not have the resources to pursue court battles, we tend to
> > regard all clauses of a copyright license as enforceable, especially
> > since not all countries have equally liberal laws regarding free
> > expression, and Debian produces an OS with a worldwide audience.
> I see this statement as utterly irrelevent.

How so?  Do you not permit principles of free expression to inform your
interpretation of what is and is not legitimate in a free software

> Other countries have entirely different laws regarding trademark
> rights, copyrights, an all varieties of rights.

Sure.  It's not a complete free-for-all, though, as most countries of
relevance to Debian are signatories of the Berne Convention.

> But none of this really matters because as long as the copyright
> holder doesn't interpret their license in such a way as to prevent
> that.

Perfectly true.

> I have to wonder why you're questioning such language that has been
> long used and included in Debian all of a sudden.

Because A) that's how we find out how the copyright holder interprets
his/her/its license (see above), and B) incorporation of old language in
a new license is just as worthy of study as new language in a new

*Something*, after all, motivates people to draft new licenses.  If it
is dissatisfaction with existing licenses, it's worth attempting to
discern what the perceived deficiencies in the existing licenses are.

> To my knowledge, this language has never been interpreted to behave
> this way, nor attempted to be used to restrict anything like this.

It does not follow that a person or organization which goes to the
trouble of creating a new license is going to hew precisely to
pre-existing interpretations of license language.  Particularly not if
different copyright holders interpret the same language differently, as
occasionally happens.

> > I think any attempt to use copyright licenses to enforce common-law
> > trademark rights should be regarded with suspicioun, since common-law
> > trademark right vest regardless of any copyright license.
> In relation to how this license it depends on how you look at this
> statement.  If you look at it as taking away rights then you have a
> point.  But I don't look at the language that way.

I must confess I don't see why.

Here's where the permissions are granted:

  Permission is hereby granted, free of charge, to any person
  obtaining a copy of this software and associated documentation
  files (the "Software"), to deal in the Software without
  restriction, including without limitation the rights to use,
  copy, modify, merge, publish, distribute, sublicense, and/or
  sell copies of the Software, and to permit persons to whom the
  Software is furnished to do so, subject to the following

The remainder of the license, up to the warranty disclaimer, consists
of restrictions on the permissions granted above.

> I look as it giving you trademark rights.

I see.  Where in the above language are trademark rights granted?

> It specifices you have the rights to use the marks to comply with the
> license but not otherwise.

A mandate is not a right.  The license mandates the use of a the
trademark is one narrow case and forbids its use in, as far as I can
tell, all other communications.

As I said in my mail to <mgr@x-oz.com>:

  >        4. Except as contained in this notice, the name of X-Oz Technologies
  >           shall not be used in advertising or otherwise to promote the sale,
  >           use or other dealings in this Software without prior written
  >           authorization from X-Oz Technologies.

  We have some concerns about this clause as well.

  6) What does "or otherwise" mean?  It would seem to include all forms of
  communication other than advertising (examples include magazine reviews,
  blog postings, and so forth).

  7) What does "or other dealings" mean?  It would seem to include all
  activities that can be promoted other than sale or use (examples include
  charitable donations of copies of the software, or the "cooking" of a
  CD-ROM with a copy of the software encoded on it in a microwave oven).

> When viewed in the context that this is a copyright license, not a
> contract or trademark licensing agreement of some sort, I don't think
> this is inconsistent with the meaning.

You confuse me.  You "look as it giving you trademark rights", but it's
not a "trademark licensing agreement of some sort"?

What is the essential characteristic of a trademark licensing agreement
if it doesn't grant permission to exercise privileges that are typically
reserved to the trademark owner?

> I think the intent here is to permit you to use the name as necessary
> and reserve all the other rights that are usually reserved to the
> trademark owner.

Okay.  If the copyright holder shares your opinion, it shouldn't be
difficult to get a clear and unambiguous statement to this effect.

> I will admit that it is poorly worded to serve this purpose.

I agree.

> But this is a long standing clause.

So what?  It was freshly incorporated into a new license in or about
October 2003.

When *is* the right time to seek correction of poorly-worded license

I don't recall that the X-Oz license was ever submitted to the FLOSS
community for public comment prior to its adoption.

> > Furthermore, it is not acceptable for a copyright license to place
> > restrictions on software interfaces.  E.g., a Debian package should
> > be able to be named "apache", or declare that it "Provides: apache",
> > to satisfy the requirements of other packages that require an
> > Apache-compatible HTTP server.
> I don't disagree, but then I think the Apache's trademark/endorsement
> clause is far worse than the X-Oz license.

The purpose of this thread is discussion of the X-Oz license.  If you
have concerns about one or more of the versions of the Apache Software
License, please start a new thread on debian-legal so that we can
discuss it on its own merits.

> > > If you say yes to both of these questions without qualification
> > > then any other license without permission to use the names would
> > > not be free.
> > 
> > If you use language like "without qualification", we're probably not
> > going to have a very enlightening discussion.
> Is this a response or just a jab?  You obviously haven't said yes
> without qualification.

I was taking exception to your construction of a straw-man argument.  I
haven't seen *anyone* propose or even imply that the answer to both of
the (snipped) questions you raised was "yes, without qualification".

> > Most licenses don't need to provide permission.  All they need to do
> > is not step beyond what copyright law allows.
> This all comes down to how you view the license.  Like I explained
> above, I'm viewing the license the opposite way from you are.  As a
> result I don't think it's steping beyond coyright law.

I admit that I cannot understand your view of the license.  Somehow it
grants trademark "rights" without being a trademark "license"?

I think it neither grants any trademark rights, nor constitutes a
trademark license.  It imposes an obligation on the recipient to include
certain language (which in the general case does not have to include a
trademark -- the term "trademark" does not even appear in the license,
which is instructive), not any sort of discretion.

I don't think of my rent payments under a lease agreement as a "right to
pay rent" -- I think of it as an "obligation to pay rent".

> > > I don't think you have the rights these licenses take away from
> > > you even if the clause was omitted.
> > 
> > Then why saddle the license with irrelevancies?
> To make it clear that they can't be stopped from complying with the
> license under trademark law.

Huh?  There are only two possibilties here: either a copyright holder is
authorized to oblige others to use a trademark, or not.

How is trademark law going to "stop" a person from using a trademark as
commanded by the mark owner?

If a copyright holder attempts to use a copyright license to mandate the
usage of a trademark without the authorization of the trademark owner,
then surely the license is defective.

I.e., if the X-Oz license said this:

  3. The end-user documentation included with the redistribution, if
     any, must include the following acknowledgment:

       "This product includes McDonald's healthy fast food goodness,
        which gives your kids cancer!  Come visit us at

     Alternately, this acknowledgment may appear in the software itself,
     if and wherever such third-party acknowledgments normally appear.

...and if X-Oz Technologies, Inc., doesn't have a legal agreement with
the McDonald's Corporation, there's a problem.  I don't think Debian
would touch such a license with a 10-foot pole without seeing some
persuasive evidence of such an agreement.

> > > Nope I didn't.  But because the name "Apache" is commonly used to
> > > refer to the software I think that qualification is necessary.  If
> > > ASF was to separate out the language for these two terms I think
> > > they could be equally restrictive on the "Apache Software
> > > Foundation" mark as the XFree86 1.1 and X-Oz licenses are.
> > 
> > ...and that may not be wise, or friendly.
> Well I think there are plenty of licenses that are free but not
> particularly nice or friendly.  But that's not really the point.

Debian has a responsibility to the health of the FLOSS community; this
is incompatible with rubberstamping licenses as DFSG-free without giving
them careful thought.

You may find the following links useful:


> > > > That is only true in one case and even then, its full name is
> > > > "Apache HTTP Server".
> > > 
> > > Yes but nobody commonly refers to it as this.
> > 
> > Does that mean nobody commonly violates the corresponding clause of
> > the license?  Or that everyone frequently does?
> The latter.  I think the Apache 1.1 license as worded is very
> difficult to comply with if you modify the software at all.

Okay.  What conclusions should we draw from this?

> > What are "other dealings"?
> I'd guess this was included to deal with inclusion in other pieces of
> software where you aren't selling it and aren't necessarily encouraging
> the use of the specific software that the license applies to but your
> derivative.

Since you're having to guess, wouldn't it be smart for us to contact the
copyright holder and find out what *they* intend it to mean?

> > > Frankly, I think this clause is trying to define an endorsement.  It
> > > probably would be better off using the BSD language.  But I don't think
> > > there is anything non-free about the language.
> > 
> > Language matters more than intent for any license that is used by anyone
> > other than the person who wrote it.

Thank you for that post-structuralist rebuttal to my point.  :)

> > Please ground your opinion on the language of the DFSG, then.
> I've already done that at length, but I'll repeat:
> 9) Doesn't contanimnate other software.  Debian already provides
> acknowledgement in every package.  This is the usual location and form
> of the acknowledgement.  I really disagree with your analysis of the
> scope of this clause.  

There's the crux.  I have argued that, depending on what is meant, it
fails DFSG 9.  You say it doesn't -- do you hold that position
regardless of what clarification as to the meaning may be forthcoming
from X-Oz Technologies, Inc.?

In other words, does even the broadest, most prohibitive conceivable
interpretation of X-Oz clause 4 pass DFSG 9 in your view?

> The only way I see it failing is if you really believe that it effects
> other software, documentation, etc...  Which I don't think this
> license does.  Especially not if you use the Alternative.

Use of the alternative (or not) doesn't really impact clause 4.

> > > > >Clause 4 is about asking you not to use their name without
> > > > >permission.
> > > > 
> > > > It's compelling, not asking. I doubt any would be upset about a
> > > > request.
> > > 
> > > You're already compelled under law anyway.  Nor do you need to use
> > > these names in order to make free use of the software.
> > 
> > Copyright doesn't cover usage, it covers copying.
> Again you're nitpicking my language and not responding to my point.

Please see the beginning of this message about the desirablility of
accuracy on a mailing list dedicated to technical issues.

I will take a guess at your meaning, then, and try to reply anyway.

Is it your contention that a license can, as part of its terms, prohibit
the licensee from ever uttering, in speech or print, the word "money",
and still be a Free Software license?

[1] Usage of the term "McDonald's" is done for satirical, humorous,
and/or pedagogical purposes only!  Any reference to cancer in this
message is entirely facetious!  Usage of trademarked terms is not a
challenge to the ownership of any such term!  Furthermore, I'm not an
oncologist, and know nothing about cancer!  Please don't sue me, Mayor

G. Branden Robinson                |    Somebody once asked me if I thought
Debian GNU/Linux                   |    sex was dirty.  I said, "It is if
branden@debian.org                 |    you're doing it right."
http://people.debian.org/~branden/ |    -- Woody Allen

Attachment: signature.asc
Description: Digital signature

Reply to: