[OSM-talk] Bad (wrong?) OSM publicity?

[Dave Sutter]

>From the thread it appears OSM is not the culprit of bad data, at
least in many cases. Nonetheless it certainly makes sense thinking
about data accuracy. I am not aware of a widespread organized system
of making sure OSM data is up to date.

We know the time of the last change to a feature. It might be a good
idea to have a way of stating a feature remains valid (and placing a
time stamp on it). The easiest thing to do would be add a tag like
"validated":"YYYY:MM:DD".

There could also be a web map that loads an area of data and lets a
user mark the features as valid or label them for fixing (at which
point an editor could open or a marker tag could be set). When a
feature is overdue to be validated, it can be highlighted on the map
so the user knows it should be checked.

Features of different types have different lifetimes. Businesses can
change very often, whereas something like the ocean is less likely to
change month to month (well, in a way significant for mapping).
Different feature types can given different timeouts before they are
flagged as overdue of validation.

This is not a complete solution. It would also be nice to be able to
validate the absence of something. Suggestions?

At the very least, this would give mappers something to do once their
local area runs out of necessary edits.

Dave

On Mon, Sep 24, 2012 at 10:32 PM, Martijn van Exel <m at rtijn.org> wrote:
> From https://www.nytimes.com/2012/09/25/technology/apple-maps-errors-send-japanese-to-homegrown-app.html?emc=tnt&tntemail0=y&_r=0moc.semityn.www
>
> "The biggest problem with Apple’s map, Mapion’s Mr. Yamagishi said, is
> that much of its data appears to be drawn from OpenStreetMap Japan, a
> Wikipedia-like service that contains a lot of incorrect and outdated
> information."
>
> 1) Is Apple using OSM data for Japan?
> 2) If they are, I can't believe this is a correct statement.
>
> --
> martijn van exel
> http://oegeo.wordpress.com
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk