Subject: Re: For approval: Open Test License v1.1
From: Alex Rousskov <>
Date: Sat, 10 Jan 2004 02:47:07 -0700 (MST)

On Fri, 9 Jan 2004, Rod Dixon, J.D., LL.M. wrote:

> You are correct that a weak trademark will require vigorous
> enforcement to avoid losing the distinctive quality of the mark;


> other than that, however, I am at a loss to come up with a pertinent
> distinction - - in terms of the size of the OWNER - - in how the job
> of enforcing your copyright license differs from enforcing a
> trademark license.

The distinction is that you do not need to enforce a license _always_
for it to remain valid in exceptional situations (AFAIK). We can add
such a clause explicitly, but my understanding is that it is the
default, at least in the US. If that is true, then clause #3 costs
nothing in day-to-day expenses and offers protection from serious
violations. A trademark would cost a lot in day-to-day expenses (both
in terms of money and good karma) and would offer similar protection.
Big OWNERs can afford the latter. Small ones cannot.

> You may want to undertake a careful re-consideration of your goal
> for condition 3, and see if it can be implemented in some other way.

Well, the goal/intent behind condition #3 is years old. We have not
come up with it just to post on this mailing list. I would be happy to
implement it some other way, but I do not see any alternatives.

Again, I am surprised and do not understand why OSI certified licenses
with very similar "rename all standard tests" clauses (Open Group Test
Suite and Artistic) cannot be used as an argument that our clause is
OK. Do those licenses have something special to offset their more
rigid restrictions? Or were they certified by mistake?

> In the meantime, it sounds as if an existing approved open source
> license might meet your needs nicely.

Err.. Which one?! I cannot find any license that would prevent people
from modifying a standard benchmark while publishing results under
now-misleading standardized names. Please clarify which existing
approved open source license meets our (clause #3) needs nicely. I
would love to use an existing license!

Thank you,


> ----- Original Message -----
> From: "Alex Rousskov" <>
> To: "Rod Dixon, J.D., LL.M." <>
> Cc: <>
> Sent: Friday, January 09, 2004 7:59 PM
> Subject: Re: For approval: Open Test License v1.1
> :
> : On Fri, 9 Jan 2004, Rod Dixon, J.D., LL.M. wrote:
> :
> : > it is not apparent - - not to me, at least - - why it is included in
> : > your software license.
> :
> : Clause #3 is included to prevent serious violations and cheating when
> : publishing standardized test results. Such publications often hurt
> : OWNERs and tool image and credibility.
> :
> : > Have you considered protecting your "reputation" interest through
> : > trademark?
> :
> : Yes, I have. AFAIK, the problem with trademarks is that they need to
> : be constantly enforced to stay in force. This is difficult to do for a
> : small, user-friendly company, informal development group, or an
> : individual author.
> :
> : Imagine: every time somebody publishes a test result with your
> : trademark in it, you have to contact the publisher, demand
> : changes/corrections/removal of test results, and threaten them with a
> : law suite if you think they even sightly violated the trademark usage
> : rules. You most likely need to pay somebody to draft such complains as
> : well.
> :
> : The publisher may be an innocent guy doing some simple testing that is
> : unlikely to be visible enough to affect your reputation. Thus, you
> : would not bother him, but you have to, because if you do not do that,
> : then when a real violation happens, the violator (now a big company
> : with lawyers, etc.) will tell you to go to hell with your trademark
> : claims because they will point to 50 cases (archived on the web) where
> : you failed to enforce the trademark. All those cases were minor
> : violations by innocent, low-profile guys, but that does not matter,
> : does it?
> :
> : Is my rough interpretation of the trademark law correct? Will the
> : owner have to enforce the trademark all the time and every time? If I
> : am wrong, trademarks would be the way to go, of course.
> :
> : Thanks,
> :
> : Alex.
> :
> : P.S. If trademarks do not work, I will try to give yes/no answers
> :      to your original questions. It is difficult to do that because
> :      your questions are using terminology and use cases that does not
> :      match what we have in mind. That is why I tried to explain rather
> :      than say yes/no.
> :
> :
> : : 3. Publication of results from standardized tests contained within
> : :    this software (<TESTNAME>, <TESTNAME>) must either strictly
> : :    adhere to the execution rules for such tests or be accompanied
> : :    by explicit prior written permission of <OWNER>.
> :
> : --
> : license-discuss archive is at
license-discuss archive is at