[OSM-talk] Who interprets semicolon in tag values?

[Jochen Topf]

On Tue, Sep 24, 2013 at 09:50:41AM +0200, Martin Koppenhoefer wrote:
> > Am 23/set/2013 um 22:44 schrieb Paul Norman <penorman at mac.com>:
> > 
> > This of course doesn't avoid contradictory tags like tiger:separated=yes;no 
> > but it at least avoids tiger:separated=yes;no;no;no;yes;no;yes;...
> 
> 
> at least they remain as indication where a combine shouldn't have been done, so you can find these easier for reverting/fixing. Insofar this is preferable behavior compared to forcing the mapper to choose one, which would result in a consistent-LOOKING situation afterwards.

I disagree. If the mapper who does the merge doesn't know which tag is
right, he should not do the merge. He is mapping in the blind. A broken
tag like "oneway=yes;no" isn't usable and it isn't fixable without local
knowledge (which the mapper who broken things might have had, but it
is unlikely that somebody seeing the broken tag has). So these cases
linger around for too long a time.

Jochen
-- 
Jochen Topf  jochen at remote.org  http://www.jochentopf.com/  +49-721-388298