[Incubator] General Blast to the OSGEO world from CollabNet

[Auke Jilderda]

Cameron,

migrating to or from a solution is always a concern with any solution,
including Atlassian Confluence (CodeHause) and CollabNet Enterprise
Edition (OSGeo).

A migration always requires effort and typically can propagate only (a
large but incomplete) part of the data set, regardless of the particular
solution you're migrating to and from.  It boils down to finding the
balance between how much of the data set you need (or want) to propagate
and how much effort you can (or want to) invest.  With OSGeo.org,
CollabNet can assist both with moving on to and off of the
infrastructure but the migration needs depend on the project and, with
that, have to be assessed and executed on a per project and on-demand
basis.  There is no generic need and solution.

Bearing in mind what data you need to migrate and how much effort that
takes, you need to balance the benefits against the down sides to decide
whether it is worthwhile for your project.  We regularly coach people
through this process and I'd be happy to assist you with this if you
like but it naturally is your decision to make.

Your point 1 is actually not a matter of CollabNet management but
OSGeo's financial sponsors continuing to support OSGeo.  And again is
not specific to CollabNet but to any infrastructure you use.  At the end
of the day, it boils down to what extent a project finds it in their
best interest to run their own infrastructure or focus their efforts on
their project.


Auke

On Friday, June 23, 2006 23:47, Cameron Shorter wrote:
> 
> Hi Andrew and Auke,
> While I agree that getting all the OSGeo projects is a good idea in
> theory there are still a few practical issues which we have at the
> Mapbuilder project.  I've outlined them at:
>
http://wiki.osgeo.org/index.php/Community_Mapbuilder_Incubation_Progress
> and copied below:
> 
>      Currently I'm reluctant to move our services to Collabnet. Here
> are the reasons:
> 
> 1. While Collabnet is currently very supportive of OSGeo which is
>    excellent to see, the reallity of the commercial world is that in
>    five years there might be a change of management at Collabnet and
>    the OSGeo support might be cut off. So before we enter in with
>    Collabnet, we need to think about an exit strategy if we loose
>    Collabnet support. As I understand it, the current exit strategy
>    would involve migrating from proprietary tools to "free for open
>    source" tools which would be expensive in effort.
> 2. Migrating to Colabnet would involve a medium amount of effort and
>    we would loose some information like bugs from the bug tracker.
> 
>      I would prefer to keep all project tools on one server rather
> than many. Currently codehaus.org is providing most of our tools.
> They provide tools for free, and the tools used are available for free
> to open source projects so we have a low cost exit stratagy if
> Codehaus folds.