25.10.2017 9:43, Frederik Ramm пишет:
"Half a good edit" is not good enough though.
When we talking about single edit "half-good" is not good. But process
of semi-automated and automated wikitags fix could be easily separated
in many independent edits even within single changeset(change of
wikitags for one object does not depend of changing wikitags for other).
And even when some of these edits is bad while most of edits is good
whole concept seems good too.
I don't research situation about (semi)automated wikidata fix deeply. I
just find out that number of arguments against it are same that used
against whole amateur cartography concept.
Well, certainly Wikipedia links should only be added by people who know
something about the feature in question, and not by a machine that
compares name tags to Wikipedia entries and takes a wild guess.
No-no... I'm not talking about generating wikitags by matching. Just
oppose: I mean we should use separate wikitags, but managing these tags
require more understanding of Wikipedia than just local knowledge. And
_*if *_we focus on local knowledge as major and only source for any tags
_*then*_ we should forbid wikitags as they could not be supported with
just local knowledge.
And I really doubt that intentionally link to redirection page instead
of true page as Tomas Strapius do here:
http://www.openstreetmap.org/node/1717783246/history is good idea in
Wikipedia-way.
--
С уважением, Златовратский Павел.
_______________________________________________
talk mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/talk
