On 6/24/06, Doug Cutting <[EMAIL PROTECTED]> wrote:
[I don't appear to be on the incubator PMC mailing list, so I don't know
what's been said about this there besides the message Erik forwarded
(below).]
The Lucy project will be all-new code. I explained the importance of
this to David and Marvin before we proposed the project. We are not
attempting to bypass the incubator. I am familiar with the incubator
and understand and support its purpose.
Marvin and I were both made very aware that any existing code would
have to pass through the incubator. The only thing we will be taking
from our original projects will be our experience.
David and Marvin are new committers, added to start this project.
Marvin and David were both active members of the Apache Lucene community
before this, as well as implementors of respected non-Apache ports of
Lucene to Perl and Ruby.
Doug
Erik Hatcher wrote:
> FYI, the Lucy project is being discussed in the Incubator PMC.
>
> Doug, Dave, and/or Marvin should probably jump in on this conversation.
>
> Erik
>
>
>
> Begin forwarded message:
>
>> From: Justin Erenkrantz <[EMAIL PROTECTED]>
>> Date: June 22, 2006 8:58:48 AM EDT
>> To: [EMAIL PROTECTED]
>> Subject: Re: new projects bypassing incubator
>> Reply-To: [EMAIL PROTECTED]
>>
>> On Thu, Jun 22, 2006 at 08:11:49AM -0400, Garrett Rooney wrote:
>>> On 6/22/06, Noel J. Bergman <[EMAIL PROTECTED]> wrote:
>>>> Justin,
>>>>
>>>> re: http://lucene.apache.org/lucy
>>>>
>>>> Is this a new project starting within Lucene, or an import? Or
>>>> something
>>>> grey in the middle? Perhaps Doug Cutting can give us more details.
>>>>
>>>> Incubation is not required, as I understand it, for projects
>>>> originating
>>>> within the ASF. For example, JAMES has a new internal project called
>>>> Postage, which is our mail server test suite, written entirely by our
>>>> Committers.
>>>>
>>>> In all likelihood, we should do more about promoting new projects that
>>>> originate within the ASF, but are we saying that they should go through
>>>> Incubation?
>>>
>>> I don't see why incubation would be required for that kind of thing.
>>> It's new code, so there shouldn't be any IP issues, and it's an
>>> existing community, so there shouldn't be any "this is how we do
>>> things" issues.
>>>
>>> Assuming that both of those assumptions are true, then I don't see
>>> what the incubator's involvement would be.
>>
>> The phrase that sends up a flag for me in the Lucy website was:
>>
>> "KinoSearch's back end will serve as a template"
>>
>> That sounds like a code import to me. But, if I'm wrong, then I agree
>> and I
>> don't see why the Incubator needs to be involved...
>>
>> *shrug* -- justin
I can see how that phrase would be misleading. However, the template
we are referring to is the algorithm (namely, the KinoSearch merge
model), not the code. You can read about the algorithm here;
http://wiki.apache.org/jakarta-lucene/KinoSearchMergeModel
Many parts of Lucene will also be acting as a templates for us in the
same way, but all of the code will be new.
I hope that clears things up.
Cheers,
Dave
