[OSGeo-Board] Sprinting

[Jo Walsh]

dear Arnulf, all, 

On Thu, Apr 20, 2006 at 08:36:23PM +0200, Arnulf Christl wrote:
> I'd like to sprint in circles around hobu and pester him to get 
> telascience up faster. Seriously - I would really love to get together 
> with some like minded (and much more competent) hackers and set up an 
> initial stack hands on. 

This process should open up a bit once hobu gets the LDAP server up
and running (and Norm really wants to wait for this, as it'll simplify
user account management so much for the telascience crowd.)

In the meantime, we can still fantasise about what we want there, and
the most useful thing to do is spec out what needs you have for the system. 
Arnulf, your email that became http://wiki.osgeo.org/index.php/Software_Stack 
was a good start, this probably wants to turn into a working document
detailing versions, dependencies etc. 

There are a whole load of interlocking use cases here: demos of OSGeo apps, 
public data services using the apps, backend build/test facilities. A
focus on what shows different projects working well together, also on
providing things that individual projects don't find it worth
sustaining themselves but would be useful to many (build/test specific systems).

> I'd come or go almost anywhere to do this for 
> two, three days ((or a weekend, but don't tell)). This would be 
> something for edu, geodata, viscom, webcom and obviously the projects 
> involved. And additionally it would add some playfulness that we should 
> have to lure more members into becoming active. If you can actually use 
> the stuff thats there and see how things develop because you say 
> something... Well I'm just talking Open Source, I know. Its kind of a 
> habbit - hard to drobbit.

Nod, this is what I was trying to get at the other week, when talking
about "pilot projects" that have good use cases for the 'stack'. 
And I want to be able (in future) to talk to people outside the Foundation, 
who have a lot of data and are interested in moving to open source already
(if not quite yet) about what use cases they have that could become
interesting and unusual demos; real-world applications with real-world 
value and beaming people being initiated into open source geospatial and 
feeling appreciative of it, would also be a good motivator ;)

I know Frank was worried about resource commitment - OSGeo is not in a
position to take "projects" on at this point - and I wouldn't suggest
a formal project with milestones, just a domain to experiment with and
to provide a goal state for what can be achieved. 
 
The Edu group are looking for specific regions - probably in the US, just
due to data availability issues - where they can collect a lot of
high-density data for use in demo packages. This could definitely be
something to focus on, to provide an agenda for a sprint. 


jo