[Tagging] Recreation_ground (landuse vs leisure)

Tobias Knerr osm at tobias-knerr.de
Mon May 27 13:38:22 UTC 2013

On 27.05.2013 10:01, Frederik Ramm wrote:
> Most "cleaners" become greedy with time. They start fixing a few typos
> in their local area and before too long they make overpass queries and
> fix "typos" world-wide without even thinking that the tag might indeed
> be used differently in other places.

I'm not sure about that. Your experience may describe those cleaners
that end up being an issue for DWG, but that's a biased sample.

Then again, it seems that we may have different definitions of "greedy"
in the first place. Mine excludes typo fixes, regardless of the size of
the affected area.

> Nothing against someone fixing a couple genuine typos while looking at
> the place in question, but as soon as your fixes become so many that you
> can't afford to look at what you're changing [...]
> then you would really do well to discuss your changes before. Yes,
> having to discuss makes such changes a little more cumbersome but they
> must not be too easy, or mistakes are more likely to happen.

Well, I think I see where you are coming from but I nevertheless
disagree with your position. In my opinion, these edits must not be too
cumbersome either, or useful improvements will not take place. When the
bureaucracy is more work than the research leading up to the edit and
the edit itself, I feel a line has been crossed. Current DWG policy
appears to focus solely on preventing damage from "mechanical edits" and
there is little to balance this consideration with the potential
benefits of such edits.

To elaborate, I agree that it is the responsibility of anyone performing
such edits to research the situation beforehand. However, there is a
difference between semantic and syntactic changes in my opinion:
Depending on which category your edit is in, the responsibilities are
different. For syntactic changes such as typo fixes, I do not consider
it necessary to gather on-the-ground evidence, because you are not
changing what the data says about the situation on the ground (the
semantics). Instead, your responsibility is to check the wiki
documentation, the number of uses in taginfo and so on - these are
sources for the _syntax_ of tags.

If a diligent research using these sources does not turn up a result,
then I believe it is ok to perform a change even without discussion (and
before someone posts the link again, this means that I disagree with
published DWG policy). It is not too much to ask that you document a
newly invented tag that could easily be confused with an existing one,
if you must use such a confusing tag at all that is. Insisting that we
leave rarely used, undocumented misspellings or duplicates alone on the
off-chance that they are intentional inventions would mean that we
forego the chance to correct the huge majority of such cases which
actually are errors.


More information about the Tagging mailing list