If I may suggest, this is how I would personally oversee the project at first blush.
Each "small work package" SWP would be divided into its various sections based on a topologically corrected areas.
ie. one persons "square" SWP would border other peoples "squares" SWP in a topologically corrected manner.
Next. Work would begin assigning 'Boundaries' to individuals of their "Square" SWP.
Intersect points of various vectors to the Boundaries would be registered and shared with adjoining "squares" SWP.
for example points could be intersected out and shared topologically corrected for the shared Boundaries.
Any issues with SWP boundaries/vector matching would be corrected at the very get-go, and people working on their SWP would be free to then work inward freely.
I think it would be good idea to build some sort of "network" for sharing the above information,
intersect points could be registered and checked out at any time.
Even something as simple as google groups could work if organized correctly, it does not have to be fancy, just clear with a straight forward method of tracking and information sharing.
This is how I would begin, roughly drafted.
I'm sure there are many other ways of going about this, and I would also love to hear those ideas.
Good luck!
Mars
On 8-May-08, at 9:31 AM, Nikos Alexandris wrote:
Let's say that a big digitisation work (at national level) has to be
carried out using georeferenced GIF's as a background and it is decided
to split the workload in "small packages".
Given that many different users will produce "different" boundaries the
problem will be BIG when it comes to merge all "small work packages" in
one unique vector layer. I assume that v.clean is not enough and
"handwork" will be required to correct the "errors".
Is there a "good" strategy to do this?
And what about building an on-line tool to share the workload of a task
like this.
_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user