License for approval

[Brian Behlendorf]

Let's say I were to implement an experimental new feature in a piece of 
UOML-licensed code.  This new feature happened to cause your conformance 
tests to report an error.  It is also a feature that I intend to propose 
for a new version of the UOML specification.  As a part of building 
community support for that feature, I need to distribute my patch that 
implements it, perhaps also the whole package with my new feature added. 
Yet, a requirement to pass your conformance tests would make that 
impossible.  This would act as a significant deterrant to innovation and 
public development.

This and other scenarios are why it's generally better to tie conformance 
to a standard to a trademark rather than to copyright.

 	Brian

On Wed, 21 May 2008, alexwang at sursen.com wrote:
> The SISSL license doesn't require the software to conform with UOML 
> standard. This is the kernel term of this license. We expect that all 
> software based on our project can be interoperable each other, this goal 
> could be reached by conforming UOML standard.
>
> -Alex
>
>> Hi Ms. Shi,
>>
>> This is useful, but I don't see an answer the question "Why is the UOML
>> license necessary when the similar SISSL is already available for your
>> use?" I suspect that Russ was expecting an explanation rather than a
>> bare statement of where the text differs. Such an explanation should
>> also answer "Why will the UOML license be a success when the similar
>> SISSL has failed and is abandoned by its creator?"
>>
>>    Thanks
>>
>>    Bruce
>>
>
>
>