[HOT] Epworth Field Papers - The Zimbabwe Connection. Advice on JOSM categorisation.

Milo van der Linden milo at dogodigi.net
Thu Jun 11 21:39:18 UTC 2015


Hello Rupert,

The first thing to do in my opinion would be to add a proposed structure
for Zimbabwe to the wiki page for administrative boundaries at the
openstreetmap wiki:
- https://wiki.openstreetmap.org/wiki/Tag:boundary%3Dadministrative

Another possible source that would give you some idea is "free the
postcode" http://wiki.openstreetmap.org/wiki/Free_The_Postcode you might be
able to look at the blocks in a postal-code like way. When you tag the
block identification in a postal-code like way on individual addresses, you
can use standard GIS techniques like voronoy to regenerate the blocks from
the individual addresses and at the same time check for anomalies.
- http://wiki.openstreetmap.org/wiki/Free_The_Postcode
- http://my2iu.blogspot.nl/2012/02/february-2012-go-transit-zone-map.html
(explains using voronoi, to get a general idea)

So in my opinion, you have two choices;

- Generate blocks at an administrative level of your choice first and
automatically add block info to addresses

or

- Enter block information as tags to addresses and analyse and correct
afterwards.

Hope this helps.

Kind regards,

Milo



2015-06-11 23:03 GMT+02:00 Rupert Allan <mail at rupertallan.com>:

>  Hi All,
> I'm Rupert, and I work with MSF as Field Mapping Coordinator. In March and
> April Kieran O'Sullivan and I worked on a Missing Maps project in the
> Epworth, Harare, Zimbabwe, where MSF are struggling to trace patients in
> order to deliver repeated treatment and follow-up for HIV, MDRTB
> (Multi-Drug Resistant TB). This is crucial in the epidemiological struggle
> (HIV at 15%), but also Epworth is an extremely vulnerable community, and
> our new OSM community there now hold the key to Epworth's
> self-representation and its accessibility to Humanitarian logistical
> efforts, in a place where sanitation is a geological nightmare, mobility is
> perenial, and housing consistently disappears in man-made and natural
> disasters.
>
> We discovered that the population was up to five times its official count,
> so we split the 840 field papers into NE, SE, SW, and NW quarters so that
> Field Papers software could deal with it.
> Now we have a few hundred papers coming in ready to input, and many more
> on the way, although hampered by WiFi and hardware access.
> In different parts of Epworth, address formats differ from others, so it
> involves careful thought about how to deploy keys/values.
> I mention in the Wiki
> <?ui=2&ik=ccc520fed0&view=fimg&th=14de47ac7ed4e704&attid=0.0.1.1&disp=emb&attbid=ANGjdJ-FBKc-pucbTfXVUV-Siv62PeTgGYGBaPxOy8hD55ITDqzt1uAhsesOe5G01aLwPuwt-2NdOdDWPWEgG64sUx1T_nuC6YPHK-zQJW6qqDUvzBBaCMUNQfOG85k&sz=s0-l75-ft&ats=1434057793222&rm=14de47ac7ed4e704&zw&atsh=0>,
> that this has evolved a degree of protection and 'anonymity' for the
> unofficial majority there. Ethics are key, as they need to be protected
> from certain factors and protected BY other factors (i.e. their visibility
> to the world as a community). So its an important project, considering
> where they are in history right now.
>
> The main issue is to input and tag 'block' boundaries, and figure out
> which administrative level to tag them on. It would be great to discuss and
> agree a bit with experienced JOSMers. The addresses depend on numbers,
> names and qualifiers of small 6-10 house areas, often with a community
> leader as the most definitive 'name' value.
> Somebody mentioned the 'Hamlet' tag. Cells are used in some addresses, but
> not others. The field papers are informative, but the addresses are defined
> differently in different parts of Epworth. This serves to protect
> communities. It's all very interesting...
>
> I am still learning JOSM, but maybe they fit into a 'multiple-choice'
> style of categorising. Almost all have numbers, but an address could be
> identical to another, except in a different ward, so miles away, or in a
> different 'block', or might hint obliquely to status as 'unofficial'. The
> numbers, by the way, are non-sequential/randomised. But it could all be
> cascade-searched with the right OSM keys/admin level or tag.
>
> Any thoughts?
>
> Thanks, Rupert
>
>
> _______________________________________________
> HOT mailing list
> HOT at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/hot
>
>


-- 
 [image: http://www.dogodigi.net] <http://www.dogodigi.net>
*Milo van der Linden*
web: dogodigi <http://www.dogodigi.net>
tel: +31-6-16598808
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/hot/attachments/20150611/c41bcd66/attachment.html>


More information about the HOT mailing list