[Board] Periodic revisiting of project incubation
cameron.shorter at gmail.com
Sat Aug 11 04:40:42 PDT 2012
Yes, the incubation process has indicated that projects have a robust
community behind it, and these projects are likely to continue to be
1. Situations change, and projects come and go. One example is the
mapbuilder project which went through incubation, but was overtaken by
Openlayers and eventually retired.
2. I believe our OSGeo projects are maturing more and more, and I think
the market is wanting to understand this maturity. Ie, it would be good
to both ask projects to validate their incubation, and also consider
raising the bar for expectations required for incubation.
However, having said all that, we need a business case and volunteers or
resources to provide the validation, and I don't have the answers on
On 10/08/12 13:08, Eli Adam wrote:
> On Thu, Aug 9, 2012 at 11:58 AM, Cameron Shorter
> <cameron.shorter at gmail.com> wrote:
>> On 10/08/12 02:11, Daniel Morissette wrote:
>>>> Having said that I do think that we do a pretty good job at being
>>>> impartial, I just want to make sure that we don't forget that as we
>>>> grow. We currently have 24 projects listed on osgeo.org, each with their
>>>> own way of doing things and we do not enforce any kind of feedback on
>>>> which we could establish some kind of control. I believe that we must
>>>> implement something here.
> To pass incubation many projects have to demonstrate that they have
> already been a successfully functioning project and community for many
> years. Inertia (of already being a successfully functioning project
> and community that has already dealt with many issues) can often keep
> things moving along successfully. Do you think that you need to
> establish something here since projects and communities are going
> defunct or likely to do so? Do you think it would be substantially
> worse to wait until it is a problem (i.e. some project/community is
> defunct) and then act?
>>> That has come up a few times in the past. Someone even at some point
>>> suggested turning the Incubation Committee into the Projects Committee and
>>> extend its mandate to followup wiht the projects every year or so.
>>> Unfortunately the committee is already having a hard time staying on top of
>>> incubator stuff, so it would not get any better of its mandate was extended.
>> I know I have questioned the long term effectiveness of an incubation
>> committee if it doesn't have periodic follow up.
> I think that projects are in general resistant to additional
> restrictions being placed on them or changing the rules of the game
> over time, particularly retroactively. There is also some
> equal-footing issues of earlier projects having an easier incubation
> process if the rules change too substantially. This occasionally
> comes up in regard to *required* marketing material or OSGeoLive
>> For me, I see the business value of incubation in that it demonstrates a
>> level of quality to the market. The comment about "how to we resource a
>> periodic incubation refresh or QA refresh" is very valid. I don't have an
>> easy magic answer for this, but do have some ideas on where to start:
>> 1. The OSGeoLive project has a light weight QA process built into it, and a
>> business driver which makes it attractive for projects to participate. QA
>> process is "Verify your application installs and runs on our Ubuntu based
>> DVD, and your docs are up to date". Business value is "If you do your QA
>> process, you will get access to a valuable marketing pipeline". This
>> OSGeoLive process could start to align with the incubation process to
>> address a periodic review of project incubation status.
> Being on the OSGeoLive DVD definitely makes a project look good.
> Projects and Chapters with annual reports in the OSGeo Journal also
> look good. Perhaps a 'strongly encouraged' but not 'required' annual
> report in the OSGeo Journal and OSGeoLive support could serve as a
> public display of the functioning and update of a community/project or
> Bests, Eli
>> 2. There is been discussions over the last year with the Eclipse Foundation
>> and LocationTech, who I understand have developed a business model based
>> around external big companies sponsoring Open Source projects, and including
>> a periodic quality controlled release. I think there is value in
>> investigating this relationship with LocationTech/Eclipse further, or
>> alternatively investigating this business model.
>> Cameron Shorter
Geospatial Solutions Manager
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
More information about the Board