[OSM-talk] Could we just pause any wikidata edits for a month or two?
Tomas Straupis
tomasstraupis at gmail.com
Wed Oct 25 06:22:47 UTC 2017
2017-10-25 8:56 GMT+03:00 Yuri Astrakhan wrote:
> Roland, thanks for the links. Local knowledge is very important, but lets
> not make it into a sacred cow at the cost of common sense. I have not been
> to every single street in New York City. I am nearly 100% sure that all
> editors has edited objects that were near their location, but that they have
> not actually visited, or walked by without walking in, etc. Local knowledge
> is an important concept, but its physically impossible to walk into every
> building and research every tag for every building. Or are you saying Tomas
> has visited every single street/building in Lithuania?
> ...
Check the name of this topic: "stop ... wikidata ..."
This started as a discussion about automated adding of one specific
wikidata tag and about doing it without discussing with local
community. My whole response was about THAT.
Yuri later tried to change the whole theme from "osm-wikidata-sql
tool" to "new general qa tool" in the same thread. This change gives a
lot of confusion on what are we really talking about. Only when
talking about automated adding of wikidata or changing other tags
based on wikidata I am strongly opposing and doing reverts. I'm
strongly in favour of automated checking, comparing etc. - we've done
a number (~40) of Lithuania specific rules starting with addressing
information and up to topological rule checking.
So if you really want to talk about that DIFFERENT topic: new QA
tool (alongside years of existence of keepright, osm inspector,
osmose, some other ones) - you need to create a new thread and discuss
it there. I would really want to know what new functionality you are
proposing etc. Collecting errors from a number of different sources
for fixing is already somewhat complex so it would be nice to
understand what is a benefit of creating a new tool rather than adding
new rules to one of the existing QA tools.
--
Tomas
More information about the talk
mailing list