[Incubator] how fond are we of open development

[Cameron Shorter]

One of OSGeo's roles is to highlight the best of breed software for 
future Geospatial users, and save these users the need to go and try 
every single project available.

So the criteria that we apply to incubation is closely aligned with the 
criteria that is used by someone assessing a project:

* Does the project do what I need?
* Is the project stable?
* Is there a support community?
* Is there support? (read quality commercial or enterprise level support)
* Does the project integrate with my other applications (read is it 
standards compliant)?
* What are the future prospects for the project?
* Who else uses the project? Does it have a lot of money behind it?

And when evaluating an Open Source project, add:
* Is there a strong community behind the project? Will the project die 
if one of the sponsoring organisations goes under, or decides not to 
support it any longer?

Deegree seems to be failing on this last criteria, which is a big cross 
against them. However, I think they are in a difficult position. Trying 
to build a community is hard, and will take years for a project the size 
of Deegree, as the barrier to entry (in learning hours) is high, and 
potential contributors will be competing directly against Lat/Lon for 
sponsorship dollars.

So to pass incubation, I'd be looking for Deegree to be extra strong in 
other areas. In particular LatLon have lots of customers using Deegree 
who will keep the project working for a good time to come.

Frank Warmerdam wrote:
> Jody Garnett wrote:
>>>  However I am more interested in seeing a community response to the
>>>  inccubation process rather than it being the responsibility of 
>>> lat/lon.
>>>  One of the things I am supposed to figure out is how diverse the
>>>  community is that supports the deegree project.
>>
>> And here is the question from Markus Lupp:
>>
>> Does this diversity have any relevance for the incubation process? As 
>> you will have noticed, the deegree 2 code is currently maintained by 
> > developers working  at lat/lon -- and most parts have been written by
> > them as well. However, we're hoping to change this with deegree 3, so
> > more active external developers become involved...
>>
>> Can I ask the email list here to help me respond. I seem to recall us 
>> checking out GeoNetwork and wanting to make sure that it was 
>> supported by a number of organizations and not just an internal FAO 
>> project.
>
> Jody,
>
> From the Principles document, I think a key is the phrases are 
> "encouraging
> participation from all contributors" and "Contributors are the scarce
> resource and successful projects court and encourage them."
>
> Implicit in the above is the assumption that new contributors can join
> a project, contribute, and have a role in the management of the project.
> To my mind, the two reasons we do this are:
>
> 1) Allowing contributors to have a role in management of a project helps
> them feel comfortable that they can influence the direction of the 
> project
> and that it won't be beholden to just one organizations or individuals
> wishes for the project.
>
> 2) Ensuring a diversity of contribution to a project gives some assurance
> that it will survive the changes in organizational or individual
> priorities that might otherwise threaten the continued operation of
> a project with a narrower base.
>
> In the past, we have required projects like MapGuide and FDO that were
> dominated by one organization to give folks outside the project 
> meaningful
> influence on the project, and to demonstrate some degree of outside
> contribution.  In the GDAL project we required a loosening from a
> dictatorial governance model to a broader PSC approach so things were
> not too tied to one individual.
>
> With the deegree project, I as an incubation committee member will be
> looking for significant signs of outside contribution (from outside 
> lat-lon),
> and that outsiders can have meaningful input on governance and technical
> directions.
>
> Others can speak up on their feelings.  To some degree graduating
> incubation is based on a subjective judgement by committee members on
> the basis of lots of criteria.
>
> There are definitely grey areas in regard to how open, and diverse
> is expected.  The MapGuide project was allowed to graduate even though
> I saw little evidence that there were developers from outside Autodesk
> who had mastered the core code base adequately to do substantial
> improvements.
>
> Best regards,


-- 
Cameron Shorter
Geospatial Systems Architect
Tel: +61 (0)2 8570 5050
Mob: +61 (0)419 142 254

Think Globally, Fix Locally
Geospatial Solutions enhanced with Open Standards and Open Source
http://www.lisasoft.com