[OSM-talk] Concern over untagged ways and map roulette

[Greg Troxel]

"Brian M. Sperlongano" <zelonewolf at gmail.com> writes:

(I think Frederik's points are very reasonable.)

> This was also my reaction.  An untagged way that isn't part of any relation
> isn't geodata and wasn't being used by any data consumer. The only time I
> would try to keep such objects is if it's clearly a trace of something
> complicated, for example someone traced the perimeter of a complex lake but
> forgot to add the water tags to it -- or, if it appears to be part of a
> botched edit.

Here, the way has a bunch of nodes, and is spatially close to a bunch of
other partial traces of streams.  It seems very likely to be valid
non-bulk mapping effort wiht an error.

>> If, on the other hand, the data has been created by someone who seems to
>> still be active in OSM then I will probably add a changeset comment
>> asking them to look at it.
>
> I do a lot of editing of boundaries, which is fairly complicated and
> detailed work when dealing with all the adjacent and overlapping boundaries
> at different administrative levels. It is really easy to make a mistake and
> leave behind sections of boundary member ways which invariably end up as
> untagged ways. This is always a mistake when it happens, and I always
> appreciate it when another mapper notices this and contacts me to point it
> out. So I also want to give a +1 to the idea that we should (a) not be
> afraid to delete things that provide no value but also (b) at least take a
> moment to look at the history of that object before deleting.

Agreed, with the emphasis that deleting other people's work rather than
talking to them (when they weren't a drive-by-departed contributor and
it wasn't a botched import) is not good for community.