Talk:Collaboration with Wikipedia
Discuss Collaboration with Wikipedia here:
Work together with Wikipedia and geonames
Please refer to the Geonames page for the latest information on this potential datasource
My name is also Kolossos at Wikipedia [1]. You can see my work: Wikipedia in GoogleEarth, Geonames in GoogleEarth and place-search.)
The API of openstreetmaps is very nice and I hope it is free, but for better orientation it would be nice to see the positions of many cities. The datas of wikipedia and geonames.org are free so you can use it, it would be possible for me to give you access to the database of wikipedia-places.
On the other site it would be nice to present the wikipedia-datas over your API, is the a chance to realise that?
we use in the wikipedia only point-datas, so we have a problem with streets, districts, rivers... . Perhaps we can solve this problem with our help. Kolossos 12:25, 6 May 2006 (UTC)
- Well sure. You seem to have a good understanding in integrating services like this. I think the biggest problem is that Wikipedia has no support for external data, and long/lat really should be external so that you can change it from a map interface. Other things are that how should you label stuff? e.g. If you look at http://placeopedia.com it's cluttered with places (it's even worse with geonames, but I think they have a sort for relevance.), how do I choose which to show? Erik Johansson 13:57, 6 May 2006 (UTC)
- One more thing is it ok to snap coordinates on a copyrighted satellite image? I think so but don't really knowErik Johansson 13:58, 6 May 2006 (UTC)
- In the wikipedia we can use for external datas templates with weblinks. So we do it with http://kvaleberg.com. Would it be possible for instance to link River Thames at OSM?
- For the revelance I use in geonames the population of a city and in Wikipedia I can use the length of an article. For paths perhaps it is possible to use the length of the path if you have no other informations like categories or speed. My idea is that a fast path like a highway it's more important than a small slow way. Or you use the number of points in the periphery of a path.
- For the copyright question I also don´t have really an answer. I think the better way is http://petition.publicgeodata.org/.
- Where is the download-page for the software and need I all satellite-image? Kolossos 14:30, 6 May 2006 (UTC)
- You can use the editting interface on front page, or read about it in Editing. Just remmember that it's almost an requirement to do you own survey, i.e. walk around with a GPS recording streetnames (and other metadata). You might want to download a standalone editor e.g. JOSM.
- Where is the download-page for the software and need I all satellite-image? Kolossos 14:30, 6 May 2006 (UTC)
- Well lets see if someone uses that for relevance.
- In my not so humble opinion the wikipedian way isn't that good, kvalberg is very tedious to use. I'm guessing the best way would be to have a template that includes a free map e.g. OSM. But I really don't want to touch wikipedias lat/long interface, placeopedia does alot better job at that. Do you think the georef templates work nicely? Erik Johansson 14:47, 6 May 2006 (UTC)
- In the german Wikipedia i think it goes good with the templates but I don't know why the english project is not so activ. Kolossos 22:12, 6 May 2006 (UTC)
This is a very old discussion from 2006. See Geonames for the latest status ("geonames.org data is probably a derived work of Google Maps and is therefore not suitable for import in OSM" ) -- Harry Wood 23:45, 22 September 2009 (UTC)
Attribution logo / message requirement for Wikipedia
Relating to Collaboration with Wikipedia#Logo requirement...
Do we require "in-image" attribution for map images appearing on wikipedia articles? Do we require a "cc-by-sa openstreetmap.org" message and/or logo on the image itself?
As with many of these things the license wording doesn't make it clear, and although we can spell out an interpretation (or even just a 'preferred' approach) stipulating size of the text etc, it will be very difficult to get everyone in the OSM community to agree to it. Some of this was discussed in relation to wikitravel.org . I shall have to assume that this discussion will not reach any definitive conclusion any time soon, but for the sake of saying "I tried"... let's discuss it here.
My view is... I don't think we should require (or try to require) any "in-image" attribution at all, because although it would be nice, it would mean that our maps might not end up being used there. This is because some people (a small but vocal minority) within the wikipedia community will be easily riled by such things. Wikipedia already makes extensive use of creative commons licensed images without giving any "in-image" attribution. This is because all images on wikipedia get their own image description page, which provides ample space for all the attribution (and promotion) we could dream of. There are already several templates to use for describing the OSM copyright, although actually I came up with my own message which I think gives a good level attribution and promotion to OpenStreetMap. Example here : http://en.wikipedia.org/wiki/Image:Holborn-viaduct-map.svg (maybe I'll adjust the templates) For me this is plenty, and if some awkward OSM people demand that we require more that this, then I'm some sure some akward wikipedia people will ensure that OSM maps never get used on wikipedia.
Another little question is, should we add "in-image" attribution to our maps image which we put on there ourselves? I think the answer is no, because it will not take very long before some wikipedian kicks up a fuss about it and demands to know whether or not this is required. I think any such debates will be counter-productive and would potentially destroy the collaboration, which would mean losing out on the promotion on all those image description pages.
-- Harry Wood 14:15, 29 April 2008 (UTC)
- I agree with you. (I have a "wikipedia point of view".) The in-image logo wouldn't be good visible in the little thumbnail in wikipedia article. The next step of the user is a click to the description page, where OSM can get so much advertisement as possible/necessary. Many wikipedia autors upload a simple screenshot and add some things for the article (i.e. A future bridge project or "Dresdner Heide" image at the top) the requirement to insert a logo without a tool would handicap this process. To use the word "copyright violation" would stop the process. I believe nobody would delete a osm-image at wikipedia because it has an in-image logo, because there is a big tolerance, but CC-license allowed it to modified the image and so to removing the logo by cropping.
- So lets concentrate to productive points. Our image Holborn-viaduct-map.svg is the first OSM-SVG that I see on Wikipedia servers, and I think it's very interesting, because the wikipedia server can render SVG. It could be perhaps a interesting way to optimize/modification the map for an article, but have a chance to updating the image by a script automatically. The file size of over 1 MB is in the moment not so good. The reason is that each letter is an own path. Perhaps this could optimize in future. --Kolossos 17:28, 29 April 2008 (UTC)
That's an old discussion. These days it's widely accepted that logos are not required, mainly because it says nothing about logos in the license text. I think it's also widely accepted that reasonable attribution does not necessarily need to appear on the map. This has been applied to mobile apps which display a map, but don't want to use up screen space while doing so (it's fine to show attribution on a credits screen and/or splash screen) and for print media if you're using a very small map on a page. It's fine to tuck the attribution away in the credits somewhere. Applying this to wikipedia, it would seem to fit that wiki sites should give plenty of attribution on the image description page (e.g.) , but no need for any in-image attribution. ...Sweeping generalisations and flaky legal interpretations abound. See Legal FAQ for more details. IANAL. -- Harry Wood 23:16, 22 September 2009 (UTC)
Usage of Wikimedia Commons
Auto importing of images from Wikimedia commons is now enabled. See Talk:Wiki/Archive01#Allow use of images from wikimedia commons for discussion details
linking to OSM objects
This may be a little far off, but I could imagine Wikipedia articles being associated with specific objects in the OSM database. Sort of the inverse of the wikipedia=*-tags. The main issue with this is that objects aren't quite guaranteed to persist in OSM (say when merging two ways) -- we'd need some kind of permalink to objects (any idea for a better name?). As an example, one might add a template call {{osm|relation|13288}} to the article wikipedia:de:Europaradwanderweg R1, which might provide a link to 13288 13288(if the resulting page were cached). Robx 08:46, 26 July 2008 (UTC)
- Please take a look to:Query-to-map and Example:Elberadweg. I hope it's going in the direction you like. I believe, pairs of key and value in combination with a bounding box are better than ID-Numbers, because they are more long time stable, human readable and can contain more than one OSM object. The biggest problem seems now the stability of OSMXAPI. Perhaps you can participate on this project. --Kolossos 11:29, 27 July 2008 (UTC)
- I integrated now also Relation into my scripts (only for paths). So you test Relation Europaradweg R1, you only need to click on Submit and than wait a moment. So also the wikipedia could link on OSM-objects. --Kolossos 16:59, 30 July 2008 (UTC)
- Not really, the "solution" is still to link to external databases (such as Wikidata) from OSM instead of the other way around. --Tordanik 17:52, 23 April 2018 (UTC)
News: Maps-Toolserver
See: http://meta.wikimedia.org/wiki/OpenStreetMap --Kolossos 21:59, 17 March 2009 (UTC)
- Update: http://journal.toolserver.org/entry/2009/04/29/faster-better-stronger/ Information about the order of the servers and changing of the toolserver-rules. --Kolossos 12:23, 6 May 2009 (UTC)
Wikimaps extension
I'm a long time Wikipedian and GIS developer. I have posted a note on Meta Wiki about an extension I'm working on, which allows embedding OpenStreetMap and other types of maps (e.g. thematic maps). I'm interested in making my extension work at a high-level of quality, robustness, and optimization so that it is usable someday on Wikipedia. I know there are other efforts and there might be opportunity for collaboration.
Aude 14:29, 20 March 2009 (UTC)
linking to osm for geotagged images
wikipedia provides links to google maps and earth for geotagged images (for example, see this image. would be nice to get osm in there as well. --Richlv 15:42, 3 July 2009 (UTC)
- We can but ask -- Harry Wood 00:02, 23 September 2009 (UTC)
- Wikimedia Commons does link indirectly to OSM. For example take the image below; find camera location section; click on the coordinates and pick link next to OpenStreetMap and you get this link. Unfortunately this link only shows the location of this one image. Is there a way to show also locations of other images nearby, as in this link to google? --Jarekt 19:06, 27 October 2009 (UTC)
Übersetzung auf Deutsch
Da bald die Integration von OSM in WP ansteht, wäre eine deutsche Übersetzung des Artikels hilfreich. Gruss, --Markus 15:35, 25 September 2010 (BST)
OSM on Wikipedia
Hi there,
Wikimedia Commons has a attribution-/licensing-template for maps that come from OSM: [2]. Maybe this should be mentioned somewhere on this page? Additionally, I think it might be a good idea to mention that uploads should go directly to commons.wikimedia.org instead of the local Wikipedias (en.wikipedia.org, de.wikipedia.org etc.) Greetings, --El Grafo 13:29, 16 May 2011 (BST)
The OSM article on Wikipedia
I think that Wikipedia:OpenStreetMap relies too heavily on blog and OSM-attributable content. I've done only a little to help this myself, but it would be helpful for others who are interested to look at how the citation profile for the article could be enhanced. Thanks for considering this. --Ceyockey 15:47, 25 November 2011 (UTC)
Commons category
I'm wondering if it would be useful to have a way of declaring that an object in OpenStreetMap is the same as a Commons category. We have Tag:wikipedia to link to Wikipedia articles, but Commons contains things that aren't in Wikipedia because they fail notability. But the existence of a Commons category enables people to verify the accuracy of OpenStreetMap as people can check the data in OSM against Commons photos. I'm thinking perhaps a key like "photo:commons" or "commons" or something like that. Any thoughts? Tom Morris 15:26, 23 September 2012 (BST)
- This could be a sensible method of connecting OSM to Commons. Things that don't have a Wikipedia article (e.g. streets) often still get a Commons category. However, I wonder whether it would be possible to parse the Commons category from an existing and tagged Wikipedia link where it exists, like we do with interlanguage links? The Commons links in Wikipedia seem less consistent than the interlanguage links though.
- As for the key, I would suggest to avoid "photo". Commons also hosts other media, such as non-photographic images, as well as video and audio files.
- Another aspect to consider: Besides categories, Commons also has pages - see Wikipedia:Commons:Paris, for example. Should these be preferred where they exist, or should we limit ourselves to categories? --Tordanik 17:31, 26 September 2012 (BST)
- If you use "media:commons=URL" you can use Category, Galleries and single images. Or more simpler "commons=URL" or "url=URL". --Kolossos 19:55, 27 September 2012 (BST)
Okay, I've started using media:commons. This seems like the current best solution to the problem. I've started tagging nodes, ways and relations in London with references to Commons categories. Tom Morris 10:24, 16 October 2012 (BST)
As well as adding media=commons, please tag the object on OSM with the corresponding Wikidata ID, and then to ensure that the Commons category is included on the Wikidata item using Property:P373. Andy Mabbett (User:Pigsonthewing); Andy's talk; Andy's edits 10:37, 12 September 2017 (UTC)
- Nowadays adding wikidata key seems to be a much better idea Mateusz Konieczny (talk) 10:51, 12 September 2017 (UTC)
Listing WP articles without georef nor OSM link?
Hi,I used the WP internal map to find all WP articles that already have an geoposition, to enrich the OSM objects with an wikipedia=* backlink. Unfortunatly not all articles have already an geoposition, so I might missed a lot of pages.
So my question is, if you could offer an easy way to find all pages of a location category (e.g. Schwerin), that don't have already an OSM WP backlink? IMHO this would be a nice job for a map action :) --!i! 18:01, 23 April 2013 (UTC)
Dispute on sections "Importing geodata from Wikipedia" and "Importing geodata to Wikipedia"
See my posting on the Talk mailing list for further information. --Nakaner (talk) 12:58, 5 November 2017 (UTC)
A quick note: strictly speaking, in the case of simple facts, the copyright license does not matter. For example, see https://creativecommons.org/faq/. This does not apply to texts; copying text fragments (for example, like a description=Some text from Wikipedia) would require a suitable license, and cc-by-sa is not suitable in the case of OSM. Something B (talk) 11:36, 12 November 2023 (UTC)
Data previously contributed to Wikipedia
After some changes, this article now contains this strongly worded warning about importing from Wikipedia: "do not even think doing such an import, even if it is your own data". I feel that this does not properly represent the legal situation, as Wikipedia's CC license is a non-exclusive license, meaning that the creator does not lose their ability to additionally publish their work elsewhere (e.g. on OSM) under a different license later. I'm also not aware of a general policy of mistrust about authorship. A creator's identity could be easily verified by sending a mail to the Wikipedia account if necessary.
In practical terms, importing content from Wikipedia still isn't feasible because it would usually be hopelessly mixed with other contributors' works, of course. Maybe we can just remove anything after "Wikipedia is also licensed under CC-BY-SA which is not compatible for import in OSM"? When a contributor adds their collected information to both Wikipedia and OSM, that isn't really an "import" after all. --Tordanik 18:06, 10 June 2019 (UTC)
- No, it is an import because we cannot trust it is the same "you" as one contributed to Wikipedia. Erkin Alp Güney (talk) 18:19, 10 June 2019 (UTC)
- Not even with an e-mail, because SMTP can easily be spoofed. Only exception would be an Apostilled statement jointly signed by Wikimedia and OpenStreetMap Foundation, validating these two identities are the same natural person. Erkin Alp Güney (talk) 18:22, 10 June 2019 (UTC)
- "Apostilled statement jointly signed by Wikimedia and OpenStreetMap Foundation" [citation needed]. In general there is plenty of ways to prove that account is controlled by the same person. For example one may place on their Wikipedia user page of user ABC "I edit on OSM under nick XYZ" and on OSM user page of user XYZ "I edit on Wikipedia under nick ABC". Or even more universal, add to Wikipedia user page "I license my edits under ODBL/public domain" that makes entire verification unnecessary. Imports from Wikipedia are unfeasible but for other reason. Mateusz Konieczny (talk) 20:33, 10 June 2019 (UTC)
- "When a contributor adds their collected information to both Wikipedia and OSM, that isn't really an "import" after all." - good point, I rephrased this section. Hopefully it is now better Mateusz Konieczny (talk) 20:43, 10 June 2019 (UTC)
- We cannot tell that to a copyright&plagiarism filter, though. Non-trust of identity is only automable way (C&P filters need a single origin to work correctly). Exceptions legally approved (I'm an academic and already deal with such filters) would get a formal filter exemption. Erkin Alp Güney (talk) 05:25, 11 June 2019 (UTC)
- I am not sure how it is relevant whatever something is automatically provable to be not a copyright violation (nearly all of OSM edits are NOT provable to be free from a copyright violations) Mateusz Konieczny (talk) 05:35, 11 June 2019 (UTC)
- If they are not provable to be free from copyright issues, let them be deleted. Digital Single Market Copyright Directive says we need to implement proactive measures against copyright infringement and plagiarism. Erkin Alp Güney (talk) 06:38, 11 June 2019 (UTC)
- Allowing only edits "provable to be free from copyright issues" would mean closing OSM (and Youtube, and forums, and twitter, and Wikipedia, and StackOveflow, and GitHub and all similar services etc). I am not sure how exactly this law will impact OSM (maybe OSM will qualify for one of exemptions), maybe it requires proactive measures against copyright infringement. In case that it actually requires all published materials to be automatically provable to be not a copyright violation then it is not going to be actually interpreted this way anyway. Mateusz Konieczny (talk) 07:31, 11 June 2019 (UTC)
- We are included by being public for more than three years. And do not qualify for library exemption (IA may). Erkin Alp Güney (talk) 11:25, 11 June 2019 (UTC)
- Those proactive measures cannot be applied by humans because of scale and underlying complexity of geodata. I do not think that OSMF can pay for 50k full-time intellectual property reviewers working three shifts a day, seven days a week. Upload filters are a necessity. Erkin Alp Güney (talk) 11:13, 11 June 2019 (UTC)
- Can you link me to some opinion by an expert that confirms that "automatically provable" is required (maybe de facto required) and that OSM is not exempt? From quick look at Wikipedia I see "The proposal makes explicit that this does not include private cloud storage services, non-profit encyclopedias (such as Wikipedia), nor non-profit educational or scientific repositories.". Depending on exact language OSM may fit "non-profit encyclopedias" and/or "non-profit educational or scientific repositories". Also, "effective and proportionate measures" may be broadly interpreted so I would prefer opinion from someone who is an expert before I will become scared. Mateusz Konieczny (talk) 13:45, 11 June 2019 (UTC)
- Article 17, Section 4: " If no authorisation is granted, online content-sharing service providers shall be liable for unauthorised acts of communication to the public, including making available to the public, of copyright-protected works and other subject matter, unless the service providers demonstrate that they have:
(b) made, in accordance with high industry standards of professional diligence, best efforts to ensure the unavailability of specific works and other subject matter for which the rightholders have provided the service providers with the relevant and necessary information;" Does this not mean you have to delete infringing/plagiarising work immediately?"Erkin Alp Güney (talk) 14:37, 11 June 2019 (UTC) - Article 17, Section 6:"Where the average number of monthly unique visitors of such service providers exceeds 5 million, calculated on the basis of the previous calendar year, they shall also demonstrate that they have made best efforts to prevent further uploads of the notified works and other subject matter for which the rightholders have provided relevant and necessary information."Erkin Alp Güney (talk) 14:37, 11 June 2019 (UTC)
- In order to implement section 6, you have two practical choices: one is to have a brigade of reviewers working three shifts a day, seven days a week, reviewing every changeset, which would break online editing of OSM (no one wants a Wikimapia-like editor hierarchy nor a settlement delay of 6 hours in OSM); the other is to have a software similarity detector automatically filtering copyright infringement and plagiarism (again, a settlement delay of 10 minutes and atomic changesets are a must for such a filter to work correctly). There is no third option. Erkin Alp Güney (talk) 19:22, 11 June 2019 (UTC)
- Army-of-reviewers idea immediately breaks up as peak contribution happens in weekends (no one would like to work a 42-hour review shift). Also forget spot-on mapping parties due to long settlement periods. Erkin Alp Güney (talk) 19:18, 11 June 2019 (UTC)
- I asked about link to "opinion by an expert", I can also randomly quote parts, interpret it and claim that I am sure (I would probably start with exceptions for "non-profit encyclopedias" and/or "non-profit educational or scientific repositories"). Mateusz Konieczny (talk) 08:21, 12 June 2019 (UTC)
- Article 17, Section 4: " If no authorisation is granted, online content-sharing service providers shall be liable for unauthorised acts of communication to the public, including making available to the public, of copyright-protected works and other subject matter, unless the service providers demonstrate that they have:
- Can you link me to some opinion by an expert that confirms that "automatically provable" is required (maybe de facto required) and that OSM is not exempt? From quick look at Wikipedia I see "The proposal makes explicit that this does not include private cloud storage services, non-profit encyclopedias (such as Wikipedia), nor non-profit educational or scientific repositories.". Depending on exact language OSM may fit "non-profit encyclopedias" and/or "non-profit educational or scientific repositories". Also, "effective and proportionate measures" may be broadly interpreted so I would prefer opinion from someone who is an expert before I will become scared. Mateusz Konieczny (talk) 13:45, 11 June 2019 (UTC)
- Allowing only edits "provable to be free from copyright issues" would mean closing OSM (and Youtube, and forums, and twitter, and Wikipedia, and StackOveflow, and GitHub and all similar services etc). I am not sure how exactly this law will impact OSM (maybe OSM will qualify for one of exemptions), maybe it requires proactive measures against copyright infringement. In case that it actually requires all published materials to be automatically provable to be not a copyright violation then it is not going to be actually interpreted this way anyway. Mateusz Konieczny (talk) 07:31, 11 June 2019 (UTC)
- If they are not provable to be free from copyright issues, let them be deleted. Digital Single Market Copyright Directive says we need to implement proactive measures against copyright infringement and plagiarism. Erkin Alp Güney (talk) 06:38, 11 June 2019 (UTC)
- I am not sure how it is relevant whatever something is automatically provable to be not a copyright violation (nearly all of OSM edits are NOT provable to be free from a copyright violations) Mateusz Konieczny (talk) 05:35, 11 June 2019 (UTC)
- We cannot tell that to a copyright&plagiarism filter, though. Non-trust of identity is only automable way (C&P filters need a single origin to work correctly). Exceptions legally approved (I'm an academic and already deal with such filters) would get a formal filter exemption. Erkin Alp Güney (talk) 05:25, 11 June 2019 (UTC)
To identify yourself as an OSM editor on Wikipedia, place username on your Wikipedia user page.
Does this still work? --Ivanbranco (talk) 09:14, 21 November 2022 (UTC)