[HOT] Typhoon Haiyan Mapping Progress

[Andrew Buck]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I agree that wider coverage will be needed and I had hoped that by now
we would have a better indication of where to map as well.  My reason
for staying with Tacloban for so long was largely due to lack of
knowing where else to shift focus to (although I did allude to this a
bit by suggesting the other villages on the coast northeast of
Tacloban), more importantly due to a second fact...

When we map an area, it is only really useful for us to map areas that
the aid organizations we work with will be responding to.  For the aid
organizations that don't know about, or don't know how to use, our
map; then no matter how good the coverage is, it doesn't help them.
This is the main reason I chose to focus on Tacloban.  It is badly hit
(as were many other places as you rightly point out) but it is also a
provincial capital, and it is the largest town in the immediate area.
 Because of this I figured that most of the international response
would likely be directed there, and since it is mostly the
international orgs that we tend to work with I figured the map data
would be most useful there.

Now, that being said I want to make it clear that the explanation
above is not necessarily an argument for continuing to focus entirely
on Tacloban, just merely an explanation of why I hadn't directed
people elsewhere yet.  I agree that we will need to spread out our
efforts at some point, and that point may be approaching, the question
is where to focus next.  As I mentioned previously, I think the
villages along the coast to the northeast will be hard hit (and due to
their proximity to Tacloban will likely receive international aid).
There are also villages along the coast to the south of Tacloban that
will have been hit hard as well since the eye passed directly over
them.  The eye track will likely have done the most damage, or the
area to the north of the eye track since the storm rotates
counterclockwise as it moves westward.

If anyone has better suggestions of where to spread out to I am
certainly open to them.  Like I said I am not saying we need to stay
at Tacloban (and the surrounding area) just explaining why I was
continuing focus there.  I know the storm affected a lot to the west
as well but I figured this would be trickier to map for two reasons.
1) it is a larger area with not such and obvious target for
international aid, and 2) the wind speeds were lower to the west due
to the storm being disrupted by the islands.  As for the idea of
mapping the area affected by the earthquake to the south, my
understanding (and this could be wrong) was that most of what we could
do remotely has already been done when the earthquake hit.

So that is all of my reasoning at the current time for our current
focus.  I hope to begin hearing more concrete info from aid orgs today
so I might redirect people when I hear from them, but for now my
advice would be to try to continue with Tacloban (especially the low
lying areas) and simultaneously spread out into the surrounding
villages until/unless we get something concrete from an aid org.

- -AndrewBuck




On 11/09/2013 05:25 AM, Jean-Guilhem Cailton wrote:
> Hi,
> 
> According to Al Jazeera, the death toll could be very high, sadly.
> And several millions of people have been affected.
> 
> I'd like to remind that an often-mentioned weakness of OSM is the
> uneven quality of the coverage, and that it is not because you have
> a hammer that everything is a nail.
> 
> So, while Tacloban was indeed hit very badly, and a detailed
> building map there is undoubtedly useful, it might also be useful
> if some of the mappers who wish to contribute took a broader view,
> to map, for example, some of the roads and villages that are
> visible on (sometimes recent) high resolution Bing imagery 
> (http://osmph.github.io/Imagery_Coverage_Map/), but sometimes
> still unmapped in OSM. (Not to mention the rivers).
> 
> GNS (http://wiki.openstreetmap.org/wiki/GNS) can also be a good
> source for names, even if it sometimes includes old versions of
> duplicated nodes with inaccurate location. High resolution imagery
> can be useful to tell which is right in these cases.
> 
> Best wishes,
> 
> Jean-Guilhem
> 

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=Mv9K
-----END PGP SIGNATURE-----