incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daniel Kulp <dk...@apache.org>
Subject Re: maven-repository cont.
Date Fri, 30 May 2008 14:57:59 GMT

On May 30, 2008, at 9:24 AM, sebb wrote:

> On 30/05/2008, Brian E. Fox <brianf@reply.infinity.nu> wrote:
>>
>>> we've been arguing for years about ease of use verses informed  
>>> choice
>>> for users of incubator artifacts. not sure that any consensus has  
>>> been
>>> reached. the current policy just introduces friction (until someone
>>> uploads the artifact to the central repository).
>>
>>
>> So are we considering informed choice to be transitive? The informed
>> choice is a red herring in the case of regular Apache releases  
>> depending
>> on the incubator artifacts. If they add the incubator repo to their  
>> pom,
>> then most users of their artifact won't even notice it. Only those  
>> that
>> are using repo managers or other policies are getting trapped by  
>> this.
>
> All the more reason not to allow regular releases to depend on
> incubator artefacts.

IMO, that would be a MAJOR hardship for the incubator projects and  
would make the process of growing their communities even harder.    
That's probably the hardest part of graduating and we should be  
HELPING them, not hindering them.

Lets look at a couple of recently graduated projects:

Woden:  if the major webservice projects like Axis 2, Synapse, etc..  
could not take a dependency on Woden, it would have been very hard for  
the Woden project to attract new developers and grow it's base.    
Also, the projects like Axis 2 wouldn't have been able to implement  
the features they needed, etc...

CXF: ServiceMix and Camel chose to use CXF for their WebServices stuff  
as it met their needs better than the other alternatives.   If they  
couldn't have depended on CXF, they would have been forced to use  
something that didn't work as well for them, the CXF community  
wouldn't have had those users to pull from, etc...   Also, a couple of  
the committers CXF added (jgawor, jgenender) were a direct result of  
the work with the Geronimo team as they integrated CXF.

OpenEJB/OpenJPA/etc...   Geronimo really needed these to provide the  
functionality it needed.   Again, without Geronimo needing this stuff,  
it would have been harder.


Basically, part of growing a community is attracting folks from OTHER  
communities.   If the other communities cannot depend on your  
artifacts, that becomes nearly impossible.   If they cannot use your  
artifacts, they will either look elsewhere or possibly even fork the  
incubator code into their own projects where they CAN depend on it.    
That would definitely be a bad precedent to start.


IMO, allowing the other projects to use the incubator artifacts is a  
very important part of helping the incubator projects grow their  
communities and eventually succeed in graduating.


Dan



>>
>> --Brian
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
>> For additional commands, e-mail: general-help@incubator.apache.org
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>

---
Daniel Kulp
dkulp@apache.org
http://www.dankulp.com/blog





---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Mime
View raw message