<html>
<body>
This is a re-post from yesterday as it was too big and was rejected by
the list moderator (image?). I've deleted the image. My
apologies to those who did receive it already. Cheers . . . .
Spring Harrison . . . .<br><br>
Hello Mappers,<br><br>
Thanks for your comments and observations in reply to mine of yesterday.
Unfortunately, no time to deal with all of them separately right now. A
forum format would be much more efficient for exploring discussions; this
e-mail approach is quite fragmented. I'm sorry if I have missed anyone
in this reply, the cast of characters seems to vary.<br><br>
Have been exploring the verification process for task 1026-236, helipad
identification. Some observations follow:
<dl>
<dd><font face="Symbol">·<x-tab>
</x-tab></font>one potential
helipad identified, no actual helipads
<dd><font face="Symbol">·<x-tab>
</x-tab></font>it looks like
there are likely quite a few more but hard to be sure using this imagery,
although it is not bad quality
<dd><font face="Symbol">·<x-tab>
</x-tab></font>although listed
in the layers pane as Bing Imagery, the caption at the bottom of the
photo says Digital Globe, this is confusing
<dd><font face="Symbol">·<x-tab>
</x-tab></font>three forested
polygons have been drawn but have very crude outlines and don't actually
represent the forested areas very well at all; many other apparently
similar forested areas are not mapped. These polygons lie mostly outside
the boundaries of tile 236. During a natural disaster response, is there
some purpose to roughly mapping random blocks of forest land?
<dd><font face="Symbol">·<x-tab>
</x-tab></font>On a second
monitor, I viewed the same area (236) in Google Earth and immediately got
a vastly better feel for the terrain and was able to quickly identify
several good helipads with good certainty as to quality. My background
includes helicopter piloting as well as GIS.
<dd><font face="Symbol">·<x-tab>
</x-tab></font>At least for
locating helipads, I would highly favour using Google Earth, the
perspective view and better image quality vastly increases productivity.
<dd><font face="Symbol">·<x-tab>
</x-tab></font>Markups could be
done directly in Google Earth, saved as KML files and forwarded to OSM.
<s>Image attached below.</s>
<dd><font face="Symbol">·<x-tab>
</x-tab></font>The only problem
with this method is that the Task Area tile grid would need to be
provided for navigation. I doubt if that would be difficult as a KML
file.
<dd><font face="Symbol">·<x-tab>
</x-tab></font>In some cases,
the age of the Google Earth imagery may be a slight drawback but since
production is the chief imperative here, that shouldn't be a big issue.
<dd><font face="Symbol">·<x-tab>
</x-tab></font>Perhaps a
customized Google Earth application using current disaster imagery could
be fired up for the duration of this exercise? On-the-fly innovation is
needed in emergencies.
<dd><font face="Symbol">·<x-tab>
</x-tab></font>As a test, you
could send me a collection of helipads for inspection in Google Earth, it
is very difficult to verify them in JOSM. In Google Earth, the
reconnaissance process is quite fast and effective.
<dd><font face="Symbol">·<x-tab>
</x-tab></font>I noticed that
there are countless Key Terms, many of them having obscure meanings at
best. This would surely confuse most new users and lead to inaccurate
tagging. The forest polygons are labeled as natural = wood in one case
and land-use = forest in another although they appear to be much the same
type of forest. No wonder data verification cannot be accomplished on
input; it looks like every user invents their own terminology! E.g.,
leisure = common denotes a helipad? JOSM is certainly not for the faint
of heart.
<dd><font face="Symbol">·<x-tab>
</x-tab></font>Most
database/GIS projects use a data dictionary approach with a more limited
but meaningful list of potential attributes. The existing data structure
would make effective querying almost impossible, far too many overlapping
options now. It would be interesting to see how these tags actually get
used when there is so much near-duplication and ambiguity.
</dl>I hope these observations are helpful. I think the use of Google
Earth would improve the helipad selection process by orders of
magnitude.<br><br>
<x-tab> </x-tab><x-tab>
</x-tab>Thanks, Cheers .
. . . . . . . Spring Harrison<br>
</body>
</html>