[HOT] FW: Message OSM + points villages MSF

Andrew Buck andrew.r.buck at gmail.com
Sat Apr 19 02:10:46 UTC 2014

Hash: SHA1

Ok, I have finished merging in the names from the shapefile you sent
me.  The changeset for the merge is listed below.  I added
source=MSF-CH to both the changeset and the nodes themselves (maybe a
bit redundant but that was what I did).  The changeset consisted of
193 new place names to add, 176 existing nodes that are either moved
to a new position or had their tags changed or both (and 4 nodes I
deleted, 2 were correct 2 need to be undeleted as I seem to have
accidentally deleted them somehow).  Thanks to Paul Norman and Jaakkoh
for helping me with the undelet work.


In total there were 358 objects in the shapefile you sent me.  I
manually checked each one and for any that had nothing at all in OSM
already I just dumped them straight in.  For the ones that already had
a name in OSM spelled identically I just merged the nodes so that the
position is taken from the MSF node and any old tags like alt_name or
the GNS unique feature ID get copied over (this lets you cross
reference your own database against the GNS database if you want).

There were about 100 or so towns which had entries already in OSM with
slightly different spellings, in these cases I just merged the nodes
and switched the OSM spelling to alt_name and used your names for the
name=* field (except for a few cases where the accenting was better on
the OSM version so I kept that in name and added yours as an alt_name).

There were about 10 or so places where OSM already had a town name but
that name was very different from the MSF name.  In these cases I kept
both nodes and added a 'fixme' and 'note' tag to explain the
situation.  Those towns can be found in the query below.  It would be
good if someone from MSF could have a look through these and see if
they can fix any of them.  The query below will pull in all of these
nodes plus all place nodes within 500 meters of them so you can see
the towns that were already in OSM that got me confused.  If you click
the 'export' button at teh top you can load the data from this
overpass query directly into josm through the remote control by
clicking the 'josm' link on the export window that pops up.


The only question I had was regarding some of the names you provided.
 About 10 or so had a town name followed by the word 'Center' or
'Centre'.  I am wondering if this 'Center' should be deleted from the
name or left as it is.  You can see an overpass query of these nodes
at the link below if you want to see the actual names.


So aside from that no major problems.  If you want to see all the
names that have been added from MSF (both in this block and the first
ones you gave us for Guekedou) you can see them at the following query:


Thanks for all of this data, it is really interesting to get results
like this back from the field.  Not only are your town names now
merged into a much larger database allowing your teams to more
accurately find place names by searching, but they also help us find
mistakes in our own data (there were a couple I found that the topo
maps had wrong, or that we had wrongly located), and we also get
alternate spellings on a lot of places to which makes you more likely
to find a place when you search for it in OSM.

- -AndrewBuck

On 04/18/2014 12:42 PM, nicolas chavent wrote:
> *Version française infra*
> Thanks CartONG, Sylvie and JG for your appreciation of the HOT/OSM 
> community support to the response to the ebola outbreak in Guinea
> and surrounding countries by MSF CH, IFRC/ARC and through them WHO,
> Government and NGOs. I am translating this email in French; the
> actions required to make the best use of the field-sruveyed 350
> features MSF-CH localities database shared with OSM are being
> looked at by some of OSM/HOT coordinators.
> Merci CartONG, Sylvie et JG pour votre appréciation positive de la
> réponse des communautés OSM et HOT en soutien à la réponse à
> l'épidémie d'Ebola en Guinée Conakry et dans les pays limitrophes
> par MSF CH, ICFRC/ARC et à travers eux l'OMS, les structures
> gouvernementales et ONGs. Ci-dessous, une traduction libre/rapide
> de votre courriel. Certains des coordinateurs OSM/HOT examinent le
> jeu de données terrain de 350 localités collectées par MSF CH et
> partagées avec OSM pour en faire le meilleur usage possible.
> Aux équipes/ communautés OSM et HOT,
> Avant toute chose et une fois de plus, nous voudrions vous
> transmettre nos vifs remerciements à chaque individu actif dans
> cette réponse de crise. Votre travail est un don pour celles et
> ceux engagés dans la réponse sur le terrain. Votre cartographie à
> l'échelle du bâti sur de si larges théâtres d'opération est une
> aide immense et précieuse pour nos équipes mobiles, ainsi que pour
> les nombreux internationaux récemment arrivés et nécessitant un
> briefing sur la dynamique notamment spatiale de l'épidémie.
> Identifier, localiser et accéder à des lieux de peuplement reculés
> est très souvent une course contre le temps dès lors que quelques
> heures peuvent faire la différence et pour les patients, et pour
> arriver à temps de façon à prévenir de nouvelles contaminations et
> d'enrayer ainsi la propagation de l'épidémie. L'utilisation
> conjointe des savoirs de terrain mobilisables et des cartes OSM ou
> produites à partir de données OSM nous a permis de gagner en
> efficacité, d'améliorer par des gains de temps l'élasticité de
> notre réponse.
> A propos des noms de lieux de peuplement, nous avons été acapable
> de constituer à partir de données terrains collectées au GPS une
> base de données consolidée de qualité portant sur les noms des
> localités (lieux de peupelement) des sous-préfectures de
> Gueckedou-Centre, Temessadou-Dibo, and Guedembou. Nous pensons que
> cela pourrait être un apport intéressant pour le projet OSM. ce
> serait notre contribution en support (à notre tour) pour votre 
> remarquable et étonnant travail. Ces données peuvent être
> librement partagées dans la mesure où elles demeurrent attribuées à
> MSF-CH. Il se peut qu'il y ait encore des modifications mineures à
> y apporter, notamment dans les endroits les plus reculés et
> inaccessibles, mais toutes ces localités ont été relevés au GPS et
> la toponymie confirmée par certains de leurs habitants.
> En espérant que cela vous soit d'intérêt. Une fois de plus : merci
> !
> On Fri, Apr 18, 2014 at 5:57 PM, Sylvie De Laborderie < 
> s_delaborderie at cartong.org> wrote:
>> Please kindly find feedback from the MSF GIS off. from the
>> field.
>> To all OSM-Hot teams, First of all and once again, we would like
>> to send a big THANK YOU to every single one of you. Your work is
>> a gift to us all in the field. Building-scale mapping for such an
>> area is a huge help for our outreach teams, and for many expats
>> newly arrived and needing a briefing on how the outbreak is
>> evolving. Finding remote localities and access to them is often a
>> race against time as a few hours can make a serious difference,
>> not only for patients, but also to arrive in time to prevent new
>> contacts to happen and keep the outbreak from spreading further.
>> Cross-checking between local knowledge and OSM maps allows us to
>> work much faster Regarding locality names, we were able to
>> acquire through various field trip with GPS devices a confirmed
>> database of localities covering the sub-prefectures of
>> Gueckedou-Centre, Temessadou-Dibo, and Guedembou. Thought this
>> might be a nice input for you guys, to come as support to your
>> amazing work. This data can be freely shared as long as credited
>> to MSF-CH. There might be still some minor confirmations/changes
>> to make, especially in remote/inaccessible areas, but these
>> places have been taken directly by GPS on the field, and
>> confirmed with local inhabitants. Hope this helps. One more time:
>> thanks!
>> JG

Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/


More information about the HOT mailing list