[Tagging] start_date variants

[Yuri Astrakhan]

On Thu, Feb 21, 2019 at 6:51 AM Martin Koppenhoefer <dieterdreist at gmail.com>
wrote:

> I am not opposing referencing wikidata in general of course, rather I am
> doing it a lot myself, even creating wikidata items from time to time, but
> this does not mean we should _move_ information from OSM to wikidata. E.g.
> not remove the "architect" tag just because someone put an
> architect:wikidata=Q123 tag.
>

Does this essentially mean that data consumers should treat
architect:wikidata as an overriding tag?  E.g. if "architect" key is
present -- use it, unless "architect:wikidata" exist, in which case
"architect" can be safely ignored.  The only exception is data quality
tools that will check that the architect:wikidata tag matches the architect
tag (obviously this will benefit human readers as well with the same
purpose of data validation).  TBH I am not too sure how much value is in
this, or if this will create more problems than solve, but I am not
categorically opposed to it either.

I would like to somehow get rid of the "us vs them" (osm vs wikidata) --
people could come to both projects and modify/remove information. Most
people won't know/care about licensing or political differences - they
treat both as "wikis" they can contribute to. After all, people are very
happy to contribute even to Google maps despite the data not being public.
So if we are to build better open data, we should encourage as much
collaboration between projects as possible, rather than try to create extra
walls or restrictions. This doesn't mean you are doing it (you said
yourself you add stuff to Wikidata), just wanted to explain my
philosophical approach to data openness :)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20190221/321c6828/attachment.html>