Subject: Re: For approval: Open Test License v1.1
From: Alex Rousskov <rousskov@measurement-factory.com>
Date: Thu, 8 Jan 2004 22:28:14 -0700 (MST)


On Thu, 8 Jan 2004, Rod Dixon, J.D., LL.M. wrote:

> The issue I see for the Open Test License v1.1 is with regard to the
> provision containing the third condition. This may not be strictly
> an OSD problem, but, it is not clear to me how condition three is
> triggered.

The only way to violate condition #3 is to publish one of the
<TESTNAME> results in violation of the <TESTNAME> rules _and_ without
OWNER's permission. Said rules will usually include a requirement to
use a specific unmodified version of the software for any published
<TESTNAME> result.

> As written, it seems to require a POTENTIAL licensee to obtain
> written permission from the owner of the copyright to the original
> program if the potential licensee intends to say "my falcon (a
> derivative work) past Owner's test by 5 parsecs" if parsecs are not
> a standardized way to measure whatever Owner's tools measure.

If "parsecs test" is not a standardized test (i.e., not one listed in
the license as <TESTNAME>), then no permission is needed.

For example, if SPEC uses Open Test License for its tests, then I can
modify SPEC's Web99 test to measure the temperature of the CPU during
the test, call the new test HotShots2004 and publish a result without
contacting SPEC. One the other hand, I cannot modify Web99 and then
publish a "Web99" result without SPEC permission.

> Is this one way condition 3 might work? Is the intent of condition 3
> to retain for the original copyright holder some degree of control
> over what licensees say about test results?

The intent is to prevent users from using standardizes test names to
name non-standard tests and their results. Such naming would mislead
those who read/use results and may damage OWNER's reputation.

Non-standard tests include any violation of standard rules, including
modification of the software if standard rules require a specific
original version to be used.

It is perfectly fine to use non-standard names for non-standard tests.
No permission is needed for that and no standard rules can prohibit
that.

> If the answers are yes, could you provide a specific example of what
> you are trying to prevent when you say: "We just do not want users
> to fiddle with what is already standardized and frozen." It may be
> easier to provide a suggestion on how to fix your clause.

Does the above example and discussion clarify? Can the wording of
clause #3 be improved without adding 10 terminology paragraphs to the
license?

Thank you,

Alex.

> : 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 http://crynwr.com/cgi-bin/ezmlm-cgi?3