On Sun, Oct 1, 2017 at 1:29 PM, Tomas Straupis <tomasstrau...@gmail.com> wrote:
> 2017-10-01 20:04 GMT+03:00 Yuri Astrakhan: > >> 2. Its not a WORK to automatically update one osm tag according to > another > >> osm tag (anybody can do it online/locally/etc). It adds NO value. > > > > It adds HUGE value, as was repeatably shown. Thanks to Wikidata IDs, the > > community was able to see and fix tens of thousands of errors in > Wikipedia > > tags. <...> > > It is mostly because you pushed the effort, not beaucse of > "advantage of wikidata". The same fixing has already been done for > YEARS before your effors based on wikipedia tags only. Tomas, you claimed that "It adds NO value." This is demonstrably wrong. You are right that the same fixing was done for years. But until wikidata tag, there was no easy way to FIND them. About a year ago when I first started this project, I created lists of thousands of such errors, that were very rapidly fixed once they were identified. This was not possible before. > And fixing > wikipedia tags is in no way inferior to your method. Maybe even > better, because it involves less „geekiness“ - they are more > understandable to larger portion of OSM community. > > My method is for finding broken wikipedia tags. What method are you talking about? Can you describe what method you use to identify errors? >> 3. It is totally unacceptable to introduce idea that wikipedia tag could > >> be removed at some time, because some other new automatically filled > tag has > >> been introduced. > > > > First, it is always acceptable to introduce and discuss new ideas. Any > > ideas. Always. <...> > > Yes. But when you're told by numerous people numerous times that > current mechanism works, and there is nothing BETTER in your advice > (other than your theoretical rambilngs), you cannot advice to destroy > existing working mechanism. > Here is the DATA for my "theoretical ramblings". Can you show any data to back your theoretical ramblings? https://commons.wikimedia.org/w/index.php?title=Data: Sandbox/Yurik/OSM_objects_pointing_to_disambigs.tab&action=history Now there is a simpler http://tinyurl.com/ybv7q7n6 query - it used to have about 1300, now down to ~750. And these are JUST the disambig errors. There are many other types as I listed in the Wikipedia improvement project on osm wiki. Lastly, what am I proposing to destroy?!? I am ADDING a tag and ADDING a new search mechanism, because there is current no reliable mechanism to fix these things. > > We are discussing the way to improve them, > > because they are currently broken. Badly. > > And they are perfectly being fixed without involving wikidata tags > there, where people WANT to do that and do WORK to fix them. > > Do you have any data to back that up? When I first looked at them, Wikipedia links were often incorrect (see links above). Now they are fixed thanks to all the work done by the communities. Yes, all that manual work that people did. But in order to WORK, you need to FIND issues first. > > Also, please elaborate which community has asked me to stay away??? > > Lithuania. We are in active action on not only fixing wikipedia > tags, but also adding missing tags to OSM, adding missing coordinates > to wikipedia, aligning coordinates between OSM and wikipedia etc. For > YEARS! > This is wonderful that you are fixing all these issues, could you tell me how you find them? Also, funny enough, I used to live in Vilnius a long time ago, near Gineitiškės. Should I be allowed to edit there? (I hope this doesn't lead to another huge but unrelated discussion :) ) > > When we create a POI detail page, we want to add a link (url without > redirects) to a wikipedia page. To do that it is straightforward to > use a value in wikipedia tag. > Great, thanks. As you can see, nothing in what I do breaks that. Instead, it actually helps your POI links to be more accurate.
_______________________________________________ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk
