Christoph, I am not talking about OSM or Wikidata or Wikipedia quality or approaches. Please don't read more into it than what I am trying to state.
If we say that we want OSM objects to link to Wikipedia (and we clearly do, judging by the number of wikipedia tags people have created), we need a good way to do it. Linking to Wikipedia with the page titles is bad. It is not stable. Wikidata tags fixes that. No other claim is being made here. On Sun, Oct 1, 2017 at 5:06 AM, Christoph Hormann <o...@imagico.de> wrote: > On Sunday 01 October 2017, Yuri Astrakhan wrote: > > > > Wikipedia created a stable ID system for these pages. Its called > > Wikidata. Please view Wikidata as first and foremost a linking system > > to Wikipedia articles. [...] > > > > Andy, you keep saying Wikidata is not verifiable data - but that's > > because you keep insisting on separating it from Wikipedia. > > Wow, i had to read this twice to really believe what i was reading here. > Seems you are still in deep denial about the fundamental differences > between OSM and Wikipedia. > > Wikipedia is based on secondary sources, it rejects original research. > Therefore you can find a lot of nonsense on Wikipedia - all kind of > urban legends and things like that, especially about remote areas, as > long as everyone believes them and no one bothers to proof them wrong > and rebut them outside of Wikipedia. So in a way Wikipedia documents > societies current beliefs about the world, not the world itself. This > does not necessarily have to go as far as an article about something > fictitious claiming to be about a real world thing, often its smaller > stuff like X being an object of type Y. The iconic 'citation needed' > of Wikipedia is not about the information being in need of actual > verification as a fact, it is about this information being verified to > be something well integrated into societies' belief system. > > OSM is fundamentally different in that because it is based on > verification by original research. This does not mean everything in > OSM holds up to this standard but we aim for this and value information > that is practically verifiable by local mappers and tagging concepts > that are targeted at verifiable mapping more than other information > that people always will keep adding to OSM to some extent despite it > being non-verifiable. > > It also means information in OSM is inherently more variable because > what people observe on the ground varies - both because what people see > depends on their experience and background and because appearance of > reality, especially of natural features, varies over time. OSM with > its original research research focus lacks the unifying and consistency > preserving effect of the filter through secondary sources you have in > Wikipedia. > > What you do when you mechanically 'fix errors' and correct discrepancies > between tags in OSM that contradict the Wikipedia/Wikidata information > is you impose the value system of Wikipedia onto OSM. > > -- > Christoph Hormann > http://www.imagico.de/ > > _______________________________________________ > talk mailing list > talk@openstreetmap.org > https://lists.openstreetmap.org/listinfo/talk >
_______________________________________________ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
