[HOT] Date of a mapped feature?

Clifford Snow clifford at snowandsnow.us
Fri May 1 06:27:43 UTC 2015


On Thu, Apr 30, 2015 at 8:37 PM, Chris Braun <brauncch at gmail.com> wrote:

> Am I the only one who believes this issue is very important and should be
> dealt with?
> Maybe it's more a mid/long term issue: while we can guess that everything
> which is mapped today in Nepal is very recent because of the effort and
> wide participation triggered following the earthquake, if another
> catastrophe happens in a few years in the same region and the mapping
> effort restarts, rescue teams will have no clue to know whether what they
> see on the map dates back from the post 2015 earthquake or was mapped
> following the second catastrophe. This is a real problem, no? And of course
> this does not only apply for Nepal.
>
> At least if would be good to systematically tag the date of the imagery
> when it is known (Bing), and maybe try to find some strategy to give an
> estimate date (base on changesets for instance) for other imagery where the
> exact date is unknown.
>
> What do you (experienced users of OSM) think of this?
>

Chris,
It is not as simple as what is the date of the image since different zoom
levels can and often do have different dates. I may be at different zoom
levels when adding in a feature. Sometimes getting up close helps, while
other times stepping back helps. All while adding the same feature. That
isn't so say that it may not be helpful.

Knowing the image source date is really helpful. As I said above, different
zoom levels can be from different dates. When adding a feature, you
obviously want the most current, but if one of the images is from MapBox
you have no idea of when the image was taken. One solution is to do an on
site survey to figure out what it looks like today.

I would suggest submitting an enhancement request to the iD, Potlatch and
JOSM developers asking them to capture the data. That may involve capturing
dates for each zoom level. You would also need to ask MapBox to have their
supplier include that information on the raster tile.

It seems like a worthwhile effort. I'd be interested in what others think.

Clifford


-- 
@osm_seattle
osm_seattle.snowandsnow.us
OpenStreetMap: Maps with a human touch
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/hot/attachments/20150430/4d4fc79d/attachment.html>


More information about the HOT mailing list