Since the dependency is from a third-party which does not follow the Apache 
release process, the definition of "released" is a grey area.

> On Mar 10, 2015, at 1:32 PM, Ted Dunning <[email protected]> wrote:
> 
> So the dependency should be on the as yet unreleased artifact?
> 
> 
> 
> On Tue, Mar 10, 2015 at 12:36 PM, Julian Hyde <[email protected]> wrote:
> 
>> Not yet. But they have told me that they intend to. And I think committing
>> that change to a public repository is a clear indication that they intend
>> to change the license.
>> 
>> As you can see from comments in
>> https://issues.apache.org/jira/browse/CALCITE-605 we still trying to
>> clarify this.
>> 
>> Julian
>> 
>> 
>>> On Mar 10, 2015, at 12:22 PM, Ted Dunning <[email protected]> wrote:
>>> 
>>> Has that change been released?
>>> 
>>> 
>>> 
>>> On Tue, Mar 10, 2015 at 12:14 PM, Julian Hyde <[email protected]>
>> wrote:
>>> 
>>>> Pentaho changed the license from GPL to Apache. See
>>>> https://github.com/pentaho/pentaho-aggdesigner/commit/8009426, made by
>>>> Pentaho's Chief Architect, Will Gorman.
>>>> 
>>>> Julian
>>>> 
>>>> 
>>>>> On Mar 10, 2015, at 9:00 AM, Vladimir Sitnikov <
>>>> [email protected]> wrote:
>>>>> 
>>>>> Hi,
>>>>> 
>>>>> Calcite has a runtime dependency on
>>>>> org.pentaho:pentaho-aggdesigner-algorithm:5.1.5-jhyde
>>>>> 
>>>>> Juilan,
>>>>> Can you please clarify what is the source of that artifact?
>>>>> Is it actually safe under Apache 2 license?
>>>>> 
>>>>> pentaho-aggdesigner is clearly GPL-2 licensed:
>>>>> 
>>>> 
>> https://github.com/pentaho/pentaho-aggdesigner/blob/master/pentaho-aggdesigner-core/gpl-2.0.txt
>>>>> 
>>>>> I am not an expert in licensing issues, however I think it is not
>>>>> valid to change project license from GPL to whatever by merely
>>>>> performing a fork
>>>>> (
>>>> 
>> https://github.com/julianhyde/pentaho-aggdesigner/commit/c9dd9a057ec5a74eefeee63b3a519aa26c7733e2
>>>> ).
>>>>> 
>>>>> 
>>>>> It looks like we should either drop pentaho-aggdesigner dependency, or
>>>>> convince pentaho to enable Apache 2 compatible usage of the artifact
>>>>> in question.
>>>>> 
>>>>> --
>>>>> Regards,
>>>>> Vladimir Sitnikov
>>>> 
>>>> 
>> 
>> 

Reply via email to