Talk:Humanitarian OSM Team/Haiti Strategy And Proposal

From OpenStreetMap Wiki
Jump to: navigation, search

From Talbot Brooks

Thoughts on organizational structure, roles...

3. Organizing around the core functional groups has worked well for our disaster mapping team in prior events - we do this in alignment with ICS, so there's a mapping chief and deputy chief - they alternate 12 hour shifts - and then a number of roles that are best described by function (they can be left unfilled if not needed, one person may fill multiple roles, or the described responsiblities may be distributed among a larger team - this is where I wish I could draw a picture...). Here goes:

1. Administration - keep tabs on who's doing what, costs, if in the field help with logistics for where to stay, ordering materials, finance (tally up the donated labor, materials, etc.. for Haiti and drop that on officialdom - it's a factoid that's needed to give this future legs) - all that sort of stuff.

2. Data collection - a group that does nothing but listen and act like an intelligence organization - find out who has what data when and if it's any good. Vacuum worthy data up, filter it, and make it usable.

3. Mapping/Cartography - those who make the map products.

4. Technical services - those who serve up the products, help with technical issues (eg, GIS people need IT folks and a host of other specialties to support efforts like this - I might argue that the diversity of this volunteer effort is what in large part has helped with its success).

5. Distribution/advertisement - making other entities aware of what's being done, delivering products/meeting their needs, and working to integrate companion efforts is a major challenge and must- do.

6. Field services - this is a detachment of folks that may be assigned to help other organizations/missions. For example, during Katrina, the mapping team in Jackson at the State EOC eventually sent 2-person teams forward to help with specialized local mapping needs.

7. QA/QC - a second set of eyes should hit every product and service before it goes online or out the door.

From Daniel Farnan

Bravo and Hurray on everything. Sorry to have been lurking (reading the mail) out-of-the-way; my new-to-me computer is only just now getting spatial.

What would help in future is a simple wireframe for responsibilities, a chain of command with requisite need know communications. This streamlines what-data-is-needed requests and speeds response, since roles are established. Also trims communications to free time to respond.

The ICS model (emergency management, Incident Command System) recommends 3-7 people working to one supervisor, in a chain of task- dissemination, with an 'ideal number' set at 5 working responder people to one oversight person. This translates roughly into one OSM contact overseeing 5 people; including for example, a communications liaison for the affected area, who liaises proprietary data, government data, open data, new/site/ public data, new/site humanitarian org data - each with a person overseeing that data gathering, and so on. One-to-five, and again for each-of-five actors, in an ongoing pyramid of response-ability. Fits into UML nicely (would show a sketch, but not able to get UML up on the new machine yet).

Again - terrific efforts. :-) Daniel

I think designated roles are terrific, but we need flexibility both in personnel and structure ... much of OSM's strength comes from the ability to just do it, rather than go through any authority or chain of command. How to balance that with the need of a sufficient response is the real balancing act. --User:Mikel

For me, not too much involved in mapping Haiti, the difficulty was to know -What is to be done, -Who is involved, -Who do what, -What is done, finally -What are the needs, and How can I help. I think a pool model is more flexible for OSM, polls by geography (e.g. people mapping Leogane or Cité Soleil), by technical task (wrapping imagery), by topic (import of hospitals). It has already existed for Haiti, more or less formally. There are several networks of relation
I think, because volunteers are volunteers', new comers would join a pool calling people according to they skill, availability... I would have join a wiki tiddying poll.
If not designated roles, leaders would appear in pools, from experience, awareness, relationship... more or less formally.
pools can also have a dedicated subpage, a completed task index (in relation with the quality grid for geografic pools)
a kind of Village_pump page (bistro in the French wikipedia), the news channel are ways to quickly spreed needs.
I can work on templates and on an idea of wizard.
--FrViPofm 10:58, 20 February 2010 (UTC)