So, to tie up the discussion. As in everything OSM, we should be
transparent and crowd-sourced. A community rebuild group is beginning
work on use the WTFE algorithm as a basis for the final rebuild. Its
deliberations are public [1]. WTFE is indicating that roughly 2.6% of
the database is affected if we rebuilt today.
From the feedback received, it is important to check that is actually
true or not before doing anything permanent. I am therefore strongly
encouraging the group to not only develop the necessary re-build scripts
but also test them in a manner that we can see, ideally in the form of a
test map, on a *copy* of part or all of the OSM database. The OSM
Foundation board is firmly committed to supporting and providing the
resources the group may require.
Meanwhile, the LWG will focus on further reducing the 2.6% by
contacting undecided users, encouraging constructive remapping and
resolving country issues. Our next step is to launch a global personal
contact-and-remap campaign, (well, almost global, there remain some
special cases). This has been piloted in a few countries and been very
successful. Given the straight-line trends [2] I am confident that we
will get a lot closer to a residual 1% of data that we will have to
remove no matter what.
Mike
LWG
[1] http://lists.openstreetmap.org/pipermail/rebuild/
[2] http://tools.geofabrik.de/osmi/munin.html
_______________________________________________
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
