[OSM-talk] Could we just pause any wikidata edits for a month or two?

Yuri Astrakhan yuriastrakhan at gmail.com
Tue Oct 24 17:07:55 UTC 2017


On Tue, Oct 24, 2017 at 11:19 AM, Tomas Straupis <tomasstraupis at gmail.com>
wrote:

> 2017-10-24 15:56 GMT+03:00 Ryszard Mikke wrote:
> > Why, in this case is it better to have Wikipedia links in OSM point to
> > disambiguation page instead of link Hillfort 1 in OSM to Hillfort 1 in
> > Wikipedia, link Hillfort 2 accordingly and fix Wikipedia doubts in
> > Wikipedia?
>
>   So that the case is not forgotten and fixed properly (i.e. ALL tags
> fixed) by people who know how to do it, not by those who are doing
> guesswork and just silencing the "qa" script.

  But in general all automated guess-edits are reverted for the time
> being because it was clearly stated they are unhelpful and so
> unwanted.
>

Tomas, I do agree that there should not be an automatic script setting tags
based on a heuristic. But what you are saying is very strange if I
understood you correctly.  What I read here is that the only people allowed
to fix things are those that know ALL tags and their meaning.  This goes
counter to the common sense (nobody knows all 65000+ tags), and counter to
the existing warnings, such as JOSM's validator "when in doubt, ignore
them".  You can never have a person who knows everything about both - the
place and OSM tags.

There are two axis of editing:  local knowledge and OSM knowledge. They are
orthogonal - I could be a tagging expert, but not know the area, or a
novice editor with the expert local knowledge.  Additionally, "local
knowledge" very rapidly decays as you move away from where you live -
another street, neighborhood, city, state, country, continent.  If I see a
problem, I can reasonably research the topic, gain knowledge, and fix the
problems in my area of expertise. Of course someone who lives in the
incorrectly tagged building, and happens to be an expert OSM editor would
be ideal, but sorry, no such luck.

In most cases, the editors who decide to help will make data better. It
might not be perfect, but it is better than before.  When you say you will
revert things despite making data worse, just because you disagree with HOW
the problem was found, and not on the basis of decreasing data quality, you
go against the very idea of a common sense.

There is only one reasonable approach to editing - data should be in a
better shape after you than before.  More accurate. More complete.  Please
don't make assumptions that the data has gotten worse just because you
disagree that there should be a qa script - after all, you are using them
yourself, and no one is reverting all your work based on that.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20171024/e66b9cb9/attachment.html>


More information about the talk mailing list