It depends on how you look at it and your standards.
First off we have the task manager system that came out of Haiti. It
isn't just used in HOT projects but elsewhere in OSM.
Next we have a degree of standardisation of things such as
highway=unclassified. The interesting thing here is how the African
highway tagging guidelines came into being. It was definitely driven by
the NGOs.
Buildings, well iD is so easy to use but I organised a small mapathon
with new mappers and just gave them JOSM and the buildings_tool. They
mapped about three or four times the number of buildings than a
comparative number of mappers had done with iD and they were more
accurate. Not one got called out for not being squared. Just for the
record if the buildings are rectangular and in line it's two mouse
clicks per building. Not in line then it's three.
Even the less than perfect mapping has saved a few lives. Getting
vaccines to the right place in the right quantities without any maps or
population counts is a problem. With imperfect maps it is less of a
problem. Perfect maps would be better still. Sometimes it is a matter
of balancing quality against usefulness. If a building is 10 cms out
for some purposes it is useless. For others the approximation is
acceptable.
Low quality mapping, yes but out of that came the duplicate buildings
script. I suspect that the OSM guidelines for making a comment in the
changeset are not always followed when the mapper mapped once two years
ago. If you download a chunk of the map off line than run the JOSM
validation tools a lot of crossing highways and not quite connected
highways etc can be picked out and corrected. Less effort than mapping
everything from scratch and over time the unresolved mapping issues do
get cleaned up.
Buildings are another issue. I think it has been mentioned before an
experienced mapper can map these in a quarter of the time it takes to
correct an existing building. Building validation basically doesn't
happen. Having said that standards do seem to be improving.
Local community? Do we really need face to face local mappers for
everything? The guides and tutorials are getting better. What the
local mappers need is support and encouragement. They also need time to
understand what OSM can do and how best to use it. The other side of
this is education. In order to map in OSM you need a certain amount of
infrastructure and planning. This education will get carried over into
their everyday lives. We also have the HOT training center and the
training working group which as evolved over time.
Certainly it has economic benefits and what is better there is no
corruption involved. Companies are paying mappers to map in OSM.
Companies exist to make a profit so there are economic benefits here.
Brand name recognition, it brought OSM into the headlines. This meant
OSM was recognised when I talk to my local city about Open Data. It
still took me five years to get the bus stops released under the right
license but the brand recognition after Haiti definitely helped.
Cheerio John
Ralf Bernhardt via HOT wrote on 2020-01-11 10:07 AM:
In 2010 Haiti was a showcase project for OSM. But what have we achieved
in Haiti since then? We have not succeeded in creating a self-sustaining
local community. HOT still runs poorly managed, low quality beginner
projects. There is a mass of unresolved mapping issues. After 10 years,
Haiti can certainly no longer be called a success story.
_______________________________________________
HOT mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/hot
--
Sent from Postbox <https://www.postbox-inc.com>
_______________________________________________
HOT mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/hot
