Wikidata:Project chat

From Wikidata
Jump to: navigation, search
Shortcut: WD:PC
Wikidata project chat
Place used to discuss any and all aspects of Wikidata: the project itself, policy and proposals, individual data items, technical issues, etc.
Please take a look at the frequently asked questions to see if your question has already been answered.
Requests for deletions and mergers can be made here.
IRC channel: #wikimedia-wikidata <IRC-connect>
 Start a new discussion
Edit
On this page, old discussions are archived. An overview of all archives can be found at this page's archive index. The current archive is located at September.


Administrators'
noticeboard

Translators'
noticeboard

Bureaucrats'
noticeboard

Requests
for permissions

Requests
for deletions

Properties
for deletion

Requests
for comment

Project
chat


Trademark discussion[edit]

Hi, apologies for posting this in English, but I wanted to alert your community to a discussion on Meta about potential changes to the Wikimedia Trademark Policy. Please translate this statement if you can. We hope that you will all participate in the discussion; we also welcome translations of the legal team’s statement into as many languages as possible and encourage you to voice your thoughts there. Please see the Trademark practices discussion (on Meta-Wiki) for more information. Thank you! --Mdennis (WMF) (talk)

The archivebot is missing a date in this thread, and is therefor not sending it to the archive. By this, my nonsens-response, a date is added, and the bot can act in a few days. -- Lavallen (talk) 18:54, 27 September 2013 (UTC)

Wiktionary/Wikidata and other projects[edit]

For those who are interested, there is a wikipage concerning the Wikidata/Wiktionary-proposals and a comparison of similar projects here: Wikidata:Comparison of Projects and Proposals for Wiktionary

I'd be happy for questions and comments!  – The preceding unsigned comment was added by SaskWa (talk • contribs) at 12:07, 3 September 2013‎ (UTC).

Vagrancy[edit]

Vagrancy is linked to "Gens du voyage", beyond any political debate occuring at the moment in France about that issue, I believe a more rightful link should "Vagabondage"

Thanks

This thread also needs a date somewhere. -- Lavallen (talk) 18:55, 27 September 2013 (UTC)

Article specific QR codes[edit]

Wikipedia-logo-en.png
Wikipedia mobile en.svg

135px wide Wikipedia logo with 135px wide QR code linking to the Main Page.
Wikidata-logo-en-135px.png
Wikidata-qr-color.png

135px wide Wikidata logo with 110px wide QR code linking to the Main Page.

Qr-1.pngQr-2.pngQr-3.png

Sample version 1, 2, 3 QR codes to demonstrate complexity.

My idea has several parts.

I think it would be neat if we had a built in QR code system that works similar to qrpedia.org. Perhaps even integrating it.

One problem with QR codes is as the url becomes more complicated so does the qr code itself making the qr code harder to be read. Given how some article titles are light-years long, one solution to this would be a system which redirects based on page IDs. Say something like enwp.org/1234567890 would redirect to the relevant article.

This ID number could be converted to a hexadecimal value too (since QR code will have to be alphanumeric anyways) to shorten/compress it which the redirecting service would know how to translate. QR code version 1 (with low error correction) supports 25 characters and just with 11 hexadecimal digits 17,592,186,044,415 decimal IDs can be represented which should be more than sufficient leaving 14 characters to construct the rest of the url. QR version 2 with high error recovery permits 20 chars. See https://developers.google.com/chart/infographics/docs/qr_codes for a complete list.

Such a service can be integrated particularly to article printouts for example allowing printouts to link back to the article.

I also have a wild idea of such a qr code displaying just under the wikidata logo (and perhaps same thing can be applied to other projects too) but I realize not everyone may be as interested as I am in making the wikidata logo machine readable. Of course this QR code would be with WMF colors :P.

So what are the thoughts on the idea?

-- とある白い猫 ちぃ？ 16:14, 20 September 2013 (UTC)

QRpedia ownership is being transferred to Wikimedia UK the United Kingdom country branch. See uk.wikimedia.org/wiki/Water_cooler#QRpedia_update. As I understand it QRPedia works (at the moment) by detecting your language settings and then checking enwp to find a link to an article in that language. Upgrading this to use wikidata items instead seems like a good idea and I hope WMUK have it in their plans for QRpedia. Filceolaire (talk) 00:56, 21 September 2013 (UTC)
Mind that currently QRPedia generates QRCodes for the actual wikipedia URL which can be a problem. Simpler QR codes are easier for machines to read. -- とある白い猫 ちぃ？ 02:45, 21 September 2013 (UTC)
Also we can go beyond the base 16 hexadecimal system to a base 36 (0-9+A-Z) system. Above 11 hexadecimal representation can be expressed with 9 base 36 digits: 68HQLCQV3. And 9 base 36 digits can represent a maximum decimal value of 2,176,782,335 which should be more than sufficient for any wiki. Mind that current highest page ID for en.wikipedia is mere 40,592,460 or O61CC in base 36 so a mere 6 characters can even be sufficient depending on the implementation. -- とある白い猫 ちぃ？ 11:32, 21 September 2013 (UTC)
Proposed the redirection component at Bugzilla:54459. -- とある白い猫 ちぃ？ 02:43, 23 September 2013 (UTC)
Mediawikiwiki:Requests for comment/URL shortener service for Wikimedia filed for technical input. -- とある白い猫 ちぃ？ 21:20, 25 September 2013 (UTC)

Adam and Eve German Wikipedia links[edit]

On the German Wikipedia, "Adam" redirects to "Adam und Eva". This is preventing other wikis that have separate "Adam" and "Adam and Eve" articles, such as English Wikipedia, from linking to them. This is the error message I got while trying to add a link for German on "Adam":

Site link Adam und Eva is already used by item Adam and Eve (Q58701). Perhaps the items should be merged and one of them deleted? Feel free to ask at Project chat if you are unsure.

This suggestion seems kind of silly. English Wikipedia has chosen to split up the topics, while German Wikipedia has combined them. Why should both wikis have to have a one to one structure in order for the articles to have relevant inter-language links?

The bottom line is that if I'm on English Wikipedia's "Adam" or "Adam and Eve", and I want to read about Adam in German, "Adam und Eva" is where I want to go, so both English articles should have this link.

Attys (talk) 15:17, 22 September 2013 (UTC)

OK, we got this question several times a week these days :) This is the Bonny and Clyde problem. In an ideal world I would just link to Wikidata:Help:Bonny and Clyde, but it's not done yet. So short answer : community has agreed to a solution to this problem : allow the link to redirection pages in interwiki links. It's not been done currently by the dev team for different reasons, but there is a workaround : replace the redirection by a non redirection articles for a few minutes, a soft redirect for example, set the interwiki, then restore the redirection. TomT0m (talk) 15:33, 22 September 2013 (UTC)
I think this is a probem in the german wikipedia and not a problem of wikidata - there was a tendence in de.wikipedia to level down the amount of pages to one by stuffing everything in the article "Wikipedia" ;-) short: they broke it, they should fix it. --FischX (talk) 14:37, 24 September 2013 (UTC)

Notability change for Commons[edit]

Since Commons is being launched tomorrow, we should probably add Commons to our notability criteria: i.e. "It contains at least one valid sitelink to a Wikipedia, Wikivoyage, or Commons page." Are there any objections? --Rschen7754 23:34, 22 September 2013 (UTC)

Symbol support vote.svg Support Definitely. --DangSunM (talk) 23:37, 22 September 2013 (UTC)
Symbol support vote.svg Support Of course. It would kind of be pointless to launch Commons and not add it to the notability policy. TCN7JM 23:44, 22 September 2013 (UTC)
Symbol conditional support.svg Conditional support I agree that the new criteria should include Commons, but it should have a more general wording so that we don't have to make votes like this everytime Wikidata is deployed in sister projects. Pikolas (talk) 00:47, 23 September 2013 (UTC)
Symbol support vote.svg Support with reservations, see below--Ymblanter (talk) 07:20, 23 September 2013 (UTC)

Exclusions[edit]

We should excluded categories in commons:Category:User_categories and commons:Category:User_categories_(flat_list) Because, these are user categories. So I propose exclude this.--DangSunM (talk) 23:42, 22 September 2013 (UTC)

  • We do not otherwise exclude these categories, so unless you want to start an RFC to exclude all of them.... --Izno (talk) 00:46, 23 September 2013 (UTC)
  • In the future, we maybe have to include namespace 2 on Commons to be able to add "Author" to files. -- Lavallen (talk) 05:58, 23 September 2013 (UTC)

Reword[edit]

I suggest "It contains at least one valid sitelink to public-facing content on a supported wikimedia family." This handles both the exlusion issue above and the need to add every supported member of the family. 130.195.179.107 23:48, 22 September 2013 (UTC)

I support this text to be added to Wikidata:Notability but I think we should still add DangSunM's language to Wikidata:Notability/Exclusion criteria. Filceolaire (talk) 23:56, 22 September 2013 (UTC)
I likewise support this wording, as well as adding DangSunM's exclusion examples, as well as explicitly excluding user category pages on other projects as well. It came up a while ago at RfD and I've been forgetting to propose it ever since. Ajraddatz (Talk) 23:59, 22 September 2013 (UTC)
Even if we do that, we should still re-evaluate this whenever a new project comes on board. --Rschen7754 00:32, 23 September 2013 (UTC)
Please start a separate RFC for user categories. We shouldn't make a major change for what is otherwise a routine update.
The proposed change by the IP would also remove support for every other non-mainspace namespace, so I would certainly oppose it, as an aside. --Izno (talk) 00:52, 23 September 2013 (UTC)
I don't even understand what it says. Before I read Izno's comment, I had no idea what "public-facing content" meant. And while I know what a "supported wikimedia family" is, someone else might wonder what this means - are all technically supported Wikimedia wikis "supported", or is only some subset of them called "supported" here? Why not just change to "It contains at least one valid sitelink to a Wikipedia or Wikivoyage page." to "It contains at least one valid sitelink." --YMS (talk) 06:00, 23 September 2013 (UTC)
What is "a Commons page"? If it is everything without a prefix, we are most likely fine. But we obviously do not want to include files as notable, and I think we do want to include categories. And what about the Commons: namespace?--Ymblanter (talk) 07:16, 23 September 2013 (UTC)
Sorry, I should have had coffee before writing this. "A Commons page" is everything hosted on Commons which does not fall under exclusion criteria. We still need to look carefully at the Commons: namespace, there are user galleries there and other stuff which we do not want to host. It might be a good idea to notify the Commons community, they know much better what they have.--Ymblanter (talk) 07:19, 23 September 2013 (UTC)
As I said above, User-pages maybe in the future have to be included, since users are Authors of a lot of files there. What we "want to" host is maybe not what we should look for. -- Lavallen (talk) 07:48, 23 September 2013 (UTC)
Do you have a complete overview what is currently in the Commons: namespace? I do not, despite being a pretty active Commons user.--Ymblanter (talk) 08:08, 23 September 2013 (UTC)
Is it really a problem ? Technically I don't think so, a bot can easily add something like a meaningless class to those items to sort them, and we can make it easy to exclude them from a standard query with query engine results. I don't think it's a problem to have those items. TomT0m (talk) 09:30, 23 September 2013 (UTC)
Hmm, after re-reading it I agree with above that the text suggested by the IP isn't sufficient. Perhaps it would be good to look at each project being added on a case-by-case basis. Ajraddatz (Talk) 22:52, 23 September 2013 (UTC)

Succesion of administrative units[edit]

I was working with the history of Botkyrka municipality (Q113718) here, when I now faced some trouble. It was founded 1971 when Botkyrka rural municipality (Q10433726) and Grödinge rural municipality (Q10509949) merged and formed a new entity. The pre- 1971 municipalities did not only had different names, they where of a different kind, following other laws than the new founded. The neighbour Salem municipality (Q1255130) was founded in the same way at the same time. All of this is easy to describe with preceded by (P155)/succeeded by (P156).

Salem was split 1974 into two parts. 10,4 km2 (9,3 was land) and a population of 249 was transferred to Södertälje municipality (Q516336). The rest was merged into Botkyrka.

Botkyrka was split 1983, and a municipality with the name Salem was again founded. This time, the parts transferred to Södertälje 1974, was kept by Södertälje, only the parts belonging to Botkyrka created the "new Salem".

Is it at all possible to describe this correctly on Wikidata? Transfer of small parts of land and population between administrative units happens all the time. Should I create separate items to describe "Salem 1971-73" from "Salem 1983-" to avoid two sets of date of foundation/creation (P571) in one item? -- Lavallen (talk) 08:17, 23 September 2013 (UTC)

I think significant event (P793) with the object <border change> (new item needed for this) will work. It's designed to track changes in an item over time. When we have a way to link to geographical objects for borders then these can be used with multiple values and date qualifiers showing each version of the unit. Filceolaire (talk) 17:38, 23 September 2013 (UTC)
In this case, at least for Salem, is there two "founded date" and one "dissolved date". It opens to a lot of possible misunderstanding for robots and coding in WP et al. In juridical terms, Salem (1971-73) is most likly not the same "person" as Salem (1983-). They have the name, type of administrative division (P132) and large parts of the geographical area in common, but they are not the same "person". If they would have had two different names, they should have separate artices on WP, just like the pre-1970-municipality has. -- Lavallen (talk) 18:25, 23 September 2013 (UTC)
Exactly. The defining issue when deciding whether to we need one page or more is: Can we describe this on one page using wikidata properties? and Could we describe it using wikidata properties if we split it into multiple pages? Filceolaire (talk) 23:02, 23 September 2013 (UTC)

Language link for Strawberry[edit]

In pl.wiki two articles (Poziomka and Truskawka) refer to the same English article, Strawberry. The link in English version of Strawberry is Poziomka. The links should look like this: Truskawka = Strawberry Poziomka = Fragraria I would change it myslef, but when I try to do this, I am always denied, because "Site link Fragaria is already used by item Q13158. Perhaps the items should be merged and one of them deleted? Feel free to ask at Project chat if you are unsure." This is curious for me, since there are already two Polish articles linking to the same English article.

Properties for family names and personal names[edit]

I just looked whether there are properties to tell us the personal names and the family names (and patronyms and middle names and whatever other naming systems there are) of persons. I did not find any and I also did not find any discussions at the property proposals. But I'm sure there were discussions about this at some time in the past. Could anybody point me to these discussions if there are any? Properties like these would for example be useful to automatically collect lists with the bearers of the name at the Wikipedia page about the name. --Slomox (talk) 14:37, 23 September 2013 (UTC)

I think surname (P734) and given name (P735) are what you are looking for. I'll just add some aliases to make them easier to find.
See Wikidata:Property_proposal/Person#patronymic_.2F_Patronym_.2F_nom_.C3.A0_suffixe_patronymique for a proposal under consideration for a patronymic property - feel free to comment or vote. Filceolaire (talk) 17:20, 23 September 2013 (UTC)
Yep, thank you, these properties are what I was looking for. Almost no entries use the properties so far, but I hope that will change in the future. --Slomox (talk) 07:49, 24 September 2013 (UTC)

Property needed for title specifier?[edit]

I'd like to get some thought on a qualifier idea. Basically I'm adding some details on Francis (Q450675) and it turns out some titles/office held (namely Titular bishop (Q948657), Auxiliary bishop (Q75178)) are inappropriate to writing in full with the associated location, which is nonetheless very relevant. In theory diocese (P708) can be used, but I'm not entirely sure it is correct in this case, plus this solution leaves several other cases unsolved. Members of legislatures, deans/director of universities, leaders of other civil or religious organizations (such as congregations)... run into the same issue, you want to specify the riding/district, but there is no article for the title of member of the United States House of Representatives (Q13218630) for the Tennessee's 7th congressional district (Q7699995), so you'd need a qualifier property. Pope Francis' titular church as a cardinal (San Roberto Bellarmino (Q959683)) can similarly not be listed. Is a property ("associated organisation or area"?) that would encompass these situations appropriate? Circeus (talk) 14:52, 23 September 2013 (UTC)

Can't we simply use of (P642) ? Simple and easy to retrieve in a variety of contexts --Zolo (talk) 16:16, 23 September 2013 (UTC)
Thanks for pointing it out. I was clueless to its existence because as far as I can tell it is not documented in any of the lists of properties! Circeus (talk) 20:26, 23 September 2013 (UTC)
I'm pushing in my RfC on constraint to switch to a class based model for proposals on constraints. This would allow a class based documentation : the noble person class would be the place to see how to use properties on noble people items ... I think the endless list of properties approach is far from beiing good enough, and it will not scale : the more it will grow, the worst it will be. see Wikidata:Requests_for_comment/Constraint_violation_technical_bases (it's on constraint, but it's the same problem for documentation, and clever use of constraint templates display could be the documentation by itself) TomT0m (talk) 20:36, 23 September 2013 (UTC)

Commons is here![edit]

Heya folks :)

It's done! We've just enabled interwiki links for Commons. \m/ Please let me know if you have questions or spot any issues. --Lydia Pintscher (WMDE) (talk) 19:52, 23 September 2013 (UTC)

Great! IMO should be better if we would not need to choose the language, because it's always English. --Stryn (talk) 20:21, 23 September 2013 (UTC)
Äh... I just looked at the IRC, and you're working on it. --Stryn (talk) 20:35, 23 September 2013 (UTC)

Off-topic: I think that article should be changed to link. (I wrote at translatewiki.net, but didn't get any answers). --Stryn (talk) 20:21, 23 September 2013 (UTC)

Well done to the development team! :) Three down, seven or so to go? Del♉sion23 (talk) 22:06, 23 September 2013 (UTC)

I added an interlink here BBC (Q9531), but it doesn't let me add more than one. The "add" button is grey now. Danrok (talk) 00:48, 24 September 2013 (UTC)

This new link on BBC (Q9531) is linking Wikipedia articles to a Commons category : precisely the type of link that was non-controversially listed as Time2wait.svg On hold in Wikidata:Wikimedia Commons. What's up exactly? - LaddΩ chat ;) 01:49, 24 September 2013 (UTC)
I was expecting to be able to add one item per language to the list. See Category:Albert Einstein, on that page there is an English commonswiki link there, but no way to add all the category pages for other languages. The add button is grey, along with a message saying "List of values is complete.". Danrok (talk) 19:00, 24 September 2013 (UTC)
You can only add one link to Commons because the project is multilingual and only has one page for all languages. The "English" label is misleading and is supposed to be removed (bugzilla:54492). The Anonymouse (talk) 19:11, 24 September 2013 (UTC)
I see now! I think my confusion was down to looking at the Wikipedia interlinks on the Commons category page (the links to the related Wikipedia articles). I was thinking that this was a replacement for that. Danrok (talk) 21:58, 24 September 2013 (UTC)
Just like for Wikipedia and Wikivoyage you can only link to one Commons page in an item. --Lydia Pintscher (WMDE) (talk) 05:53, 24 September 2013 (UTC)
What happens if you need to link to both a commons gallery and a commonscat? --Jakob (Scream about the things I've broken) 11:37, 24 September 2013 (UTC)
Wikidata can't handle that. The best option I can see is to pick the one with most viewers, so as to inconvenience the fewest people when interwiki links on the page you don't pick fall out of date. Maybe bots will pick up the slack. (Ah, interwiki bots, deja vu!) --Avenue (talk) 12:24, 24 September 2013 (UTC)

I saw the Commons interlink is listed in q937, as English language. As the Commons is multilingual project, language should be displayed as "Multilingual", not English. Kwj2772 (talk) 11:36, 24 September 2013 (UTC)

See above. We are working on that. The bug for it is bugzilla:54492. --Lydia Pintscher (WMDE) (talk) 11:58, 24 September 2013 (UTC)

What is the different from "Commons category" [1]; Xaris333 (talk) 12:17, 24 September 2013 (UTC)

It can store the same data, but with a different function. See commons:Commons:Village_pump#Wikidata_is_here.21 for more details. --Avenue (talk) 23:24, 26 September 2013 (UTC)

entries[edit]

Hello. I would like to translate the word "entries" that appears in all wikidata pages with the greek word (for the greek version) but i can't find it anywhere in translatewiki.net. Xaris333 (talk) 19:57, 23 September 2013 (UTC)

It's this one (change the code en). --Stryn (talk) 20:04, 23 September 2013 (UTC)

Thx! Xaris333 (talk) 20:29, 23 September 2013 (UTC)

Language link: Julio Alak[edit]

I tried to link Julio Alak to خوليو حلاق to no avail. It's the same guy, I swear! (check with google translate if you don't believe me :) ). I get this: An error occurred while trying to perform save and because of this, your changes could not be completed. Details: Site link خوليو حلاق is already used by item Q12210978. Perhaps the items should be merged and one of them deleted? Feel free to ask at Project chat if you are unsure. --FiliusRosadis TheRealOne (talk) 20:17, 23 September 2013 (UTC)

→ ← Merged. --Stryn (talk) 20:44, 23 September 2013 (UTC)
THX :) --FiliusRosadis TheRealOne (talk) 13:36, 24 September 2013 (UTC)

From a community standpoint, today's Commons deployment was a tremendous failure[edit]

From a technical standpoint, this appears to be a successful deployment. Hats off to the developers for that.

From a community standpoint, however, today's deployment was a failure. We knew that this was coming for weeks and did not prepare for it. When Wikidata launches on new projects, there are several things that the Wikidata community needs to do before launch:

  • The Wikidata community, in consultation with the sister project(s) that Wikidata is being deployed to, needs to determine what data is coming over, how the data is going to be transferred, and when the transfers are going to happen.
  • The Wikidata community needs to set up a page explaining Wikidata going live on the sister project(s) means to those sister projects(s) — how will it effect those projects, where will they see changes, and how they can help. A list of people involved in both projects that are ready to field questions or offer assistance would also be helpful.
  • The Wikidata community should have infrastructure for the deployment ready to go before the deployment happens. This means both that the necessary policy changes need to be amended (after the necessary policies are decided on), and that things like interwiki link importing bots should be selected and largely ready to go ahead of time.

Instead, what we have is:

  • Wikidata:Wikimedia Commons, a proposal page that pretty much only indicates that no decisions have been made and everything is on hold
  • A thread on the Commons village pump that went stale a week ago as the primary means of communication between the two projects
  • BotMultichill (talk • contribs • logs) doing interwiki imports, with only the tacit approval of the people that were in the #wikimedia-wikidata IRC channel to go on, and no established set of policies to govern what interwiki links he should and should not be importing
  • A majority of Commons users are oblivious that this is happening, and because they have not been brought into the loop, are not monitoring the effort

In many ways Commons is the closest project to Wikidata among all the WMF projects, and potentially has the most to benefit from a close working relationship with Wikidata. If Wikidata is to be successful we need to do better at creating a bridge to the projects we're deployed to, and we need to improve our preparation skills. If we don't, Wikidata will become a walled garden, and that's something I don't think anyone wants.

Sven Manguard Wha? 23:06, 23 September 2013 (UTC)

As the user who started the actual version of Wikidata:Wikimedia Commons, I totally quote Sven: this is the first time, out of three, that we just don't know what to do. And this is absolutely awful. I also thought that links such as Wikipedia page = Commons gallery and Wikipedia category = Commons category would have been non-controversial points in this operation, but it turned out I was wrong... --Sannita - not just another it.wiki sysop 23:35, 23 September 2013 (UTC)
+1. For a site that is in development like this one, we all need to take efforts to stay on top of current developments, and figure out how they will affect us. --Rschen7754 23:37, 23 September 2013 (UTC)
A new thread on Commons: commons:Commons:Village pump#Wikidata is here!--Ymblanter (talk) 10:36, 24 September 2013 (UTC)
I have pointed out an example in that thread (the Albert Einstein pages) of the sorts of problems being caused already. Personally I don't think all the blame belongs with the Wikidata community; unlike Sven, I think the developers also deserve a big share. Their tendency to deploy (mis)features before the relevant communities have confirmed they want them and can handle them is IMO the root cause of the immediate problem here. --Avenue (talk) 11:19, 24 September 2013 (UTC)

So, how do we fix this?[edit]

See header. --Rschen7754 03:03, 24 September 2013 (UTC)

The project as a whole suffers from a critical lack of willingness to discuss things. Participation rates at most Requests for Commons are bad. Participation rates at Properties for deletion discussions are outright abysmal. Major decisions, from how new datatypes are implemented to bot approvals, are conducted in IRC channels, far too often including the admin-only channel, because many of the most active users (myself included) have come to the unpleasant realization that IRC is the only place where you can get a half dozen people to discuss an issue within a reasonable amount of time. We have project critical RfCs that have been open for several months, and in the absence of a community consensus on those issues, people are just doing what they think makes sense, even if they wind up contradicting each other. The failure of Wikidata to prepare for the rollout of Commons is a symptom of a greater problem — the project's inability to make decisions. How then do we fix this immediate issue then? We could start an RfC, but would people participate? I don't think that they will, but right now it's the only real solution we have. Sven Manguard Wha? 03:32, 24 September 2013 (UTC)
It's hardly surprising that this happens though. There are tens of RfCs open at any given time, half of which are on something that I don't understand, and neither does anyone other than a select few other people. I get the feeling that if Wikidata was somehow more easy to understand, then more people would be involved - but a major obstacle to that is a lot of stuff on Wikidata isn't even defined. That's what all the RfCs are about. Combine that with the fact that discussion happens on any one of the 52 project chats and the RfC pages, and nobody knows what discussions are even going on. Then there are noticeboards on top of that too! My personal preferences on a solution then look more at what I see as the root of the problem:
  • Consolidate policy-making discussion onto one page, project discussion (help, how do I do this, can someone fix this vhow andalism, etc.) onto another.
  • Make the policy-making discussions somewhat followable. Translation isn't a big deal, but scope is. I have almost no idea what is going on here, as an example. These massive sweeping RfCs obviously aren't working, so trying to make RfCs that address a specific question would be very, very beneficial for dumb people like me anyway. An RfC could follow the principles of a good essay or paper: present the topic, present your thesis, discuss from there. Example: The topic is instance of (P31). The thesis is that Q5 be used for any human people with P31. Support or oppose.
  • Unfortunately, the above isn't possible because of the insane system of determining which properties exist and which don't. Sven accurately summed up the two schools of thought on lots of properties vs. few properties, but all but maybe five users here have no clue what that means - either the definition or the implication.
  • Which leads to my next point, all properties should have very clear guidelines on their use prior to being created or used. Then a discussion over what qualifier to use for humans with regards to P31 would be easy, because it would be clarifying a specific area, whereas right now we need to clarify the entire property and properties as a whole. Am I even using the word qualifier correctly? Because it isn't documented well anywhere. I'll call it the "thing in the property" since that better describes it.
Honestly, I'm not sure if any of those ideas were solutions. More just what I see as parts of the massive problem. Ajraddatz (Talk) 04:14, 24 September 2013 (UTC)
What I think and try to push by RfCs : we should stop focusing and properties and start focusing on classes. Every items might have 10th of claims, each of them with several qualifiers : we should start to model things at the infobox level in you don't like the term class or type. It's a paradigm shift which will allow to think of a lot of usecases at the sametime : how do we model a city ? how do we model a computer scientist ? how do we model a math theorem ? Versus how do we use the <serie> property. Things are slow and confusing and ill documented on WIkidata, we need this paradigm shift to scale : it's a lot faster to decide for several property and a lot of items in one discussion than property and claims separately ... TomT0m (talk) 08:08, 24 September 2013 (UTC)
One more thing : discussing property by property community wide is exactly as discussing infobox field by infobox field community wide. From what i've seen, on WIkipedia, if someone wants to create an infobox, he creates an infobox. This is a major difference that explains a lot, in my opinion. TomT0m (talk) 09:08, 24 September 2013 (UTC)
I totally agree with all of you. Why not give this RfC regarding RfCs a try? And whenever we propose properties here or whole data models there we could use http://test.wikidata.org to create working examples. Both, simple as well as complex examples could be created as reference and as a basis for the descision. Perhaps this would help in getting a feeling of the look and feel of the proposal instead of abstract prose descriptions which are hard to follow in many cases.  — Felix Reimann (talk) 11:48, 24 September 2013 (UTC)
It seems overkill to launch a RfC each time we want to model a class, it would lead to tenth (and more) or RfCs. We need a model/class proposal/amendment' page, or something weaker. This would imply the creation of the needed properties all a once, maybe bypassing the property proposal page. TomT0m (talk) 14:39, 24 September 2013 (UTC)
One word about policy decisions : I don't think we need a better RfC process, we need a community agreement so that the RfCs go fast, and discussions so that the notions are clear enough (with a RfC rereading process prior to their launch ? I'm not really a good teacher so if someone could reread them and make them understandable prior to their launch it would be a good thing). TomT0m (talk) 16:26, 24 September 2013 (UTC)
As far as project launches, I've started a checklist at Wikidata:Sister projects. Feel free to add more questions that we should ask ourselves whenever we add more projects. --Rschen7754 01:34, 25 September 2013 (UTC)


Well, so what I meant is how to fix up our integration with Commons, and fast :P Though the big picture issue is worth discussing, right now we've got a lot of confusion, and meanwhile people are adding stuff without any guidance. --Rschen7754 05:55, 24 September 2013 (UTC)

I think we should (1) stop import immediately; (2) open an RfC on Commons (the main issue apparently being links of commons categories to Wikipedia pages); (3) in the meanwhile, discuss other issues outlined by Sven. I definitely like Multichill's proposal, but I do not think we should go forward at least before it has been accepted on Commons.--Ymblanter (talk) 07:17, 24 September 2013 (UTC)
I'm not totally in favour of stopping import. I think we can go on with the two non-controversial points I set up in the draft: items for WP/Wvoy articles should link to Commons galleries and items for WP/Wvoy categories should link to Commons categories. About the discussion with Commoners about "cross-links" (i.e. WP article -> Commons category), definitely +1: we need to talk with them about it.
Meanwhile, I made a "bold" decision and modified WD:N according to the fact that now Commons is supported - even if we still don't know how to deal with "cross-links". @Ymblanter:, can you please link the Multichill's proposal about that? Thanks. --Sannita - not just another it.wiki sysop 11:16, 24 September 2013 (UTC)
Well, I disagree that we should just charge ahead willnilly with making same-namespace interwiki links. Doing so can cause big problems, as I've explained in the Commons Village Pump thread linked above. Please wait for a consensus both here and on Commons before causing more damage. --Avenue (talk) 11:25, 24 September 2013 (UTC)
I probably can link the Multichill's proposal, but I do not quite see where to link it from.--Ymblanter (talk) 12:36, 24 September 2013 (UTC)
One suggestion - the front page Wikidata:Main_Page could be improved. As it is, much of it is just static information/links. It needs to be more informative about what is going on today, where help is needed, where important discussion is needed, etc. Something like a to-do list for the community. Alternatively Wikidata:Community_portal could be used for this. And, presently the same problem exists there, it's just some static links. We need the links, but we also need something that is a bit more lively looking. So you can see at a glance what people are chatting about, etc. Danrok (talk) 04:05, 25 September 2013 (UTC)

[edit]

Request for consultation on this community logo

First, I’d like to apologize for the English. If you can, please help to translate this for other members of your community.

The legal team at the Wikimedia Foundation would greatly appreciate your input on the best way to manage the "community logo" (pictured here) to best balance protection of the projects with community support. Accordingly, they have created a “request for consultation” on Meta where they set out briefly some of the issues to be considered and the options that they perceive. Your input would be invaluable in helping guide them in how best to serve our mission.

Thank you! --Mdennis (talk) (via the Global message delivery system). 03:24, 24 September 2013 (UTC) (wrong page? You can fix it.)

Import Interwiki doesn't work for Commons[edit]

E.g. try doing it for Q4654740: it says "unknown language". It Is Me Here t / c 13:00, 24 September 2013 (UTC)

It has been solved if I look at the right place. Earlier today, I typed "English" in the corresponding window, and it worked (for a different item).--Ymblanter (talk) 13:12, 24 September 2013 (UTC)
I think It Is Me Here means slurpInterwiki gadget. And why it doesn't work is propably because it does not yet support Commons. Someone should notice Tpt about that. --Stryn (talk) 13:52, 24 September 2013 (UTC)
This is what I meant, yes. It Is Me Here t / c 10:54, 25 September 2013 (UTC)

Litteral data and import tools[edit]

I've been looking into how we could massively import lots of data and how a tool for this should work. It seems like at least one such tool exist and it is probably a lot better than those we are using now when it comes to big datasets. I leave the name out for now, a lot of work is necessary to make it work properly in our context.

The tool is based upon an idea of transforming source XML by small code snippets, and then we could use the generated target XML to make the necessary API requests and JSON structures for upload to Wikidata. One of the methods for importing the source XML is to use OAI-PMH, which means a lot of data is available. Some libraries publish their complete catalogs with this protocol, not to forget that Europeana uses this protocol so a lot of data is available. Some such feeds can contain millions of records.

The tool act as a collaborative repository for transformation rules, so if some user set up the rules an other user can reuse the rules weeks or months later to update our imported version. One small note, we must be careful to update what we know is the specific source idea about correct data and not mess it up with data from other sources.

That is all the fun facts. The problem is that we (the community) do not have any clear understanding about what kind of information we do want and where to stop. This will not only be the preferred name of a book author, but it could very well be complete lists libraries that have a specific book. Likewise it could be data about lakes that not only contain various cross-sections and coverage, but also water level as time series in hour intervals for the last 200 years. The amount of data could be very big, and in fact so big that if we run such a "bot" it will overload everyone. We need some rules on what kind of data we should import and what we should just link to at external sites.

As a very coarse rule of thumb I propose we (nearly) only import literal data that can be used in our sister projects. We do not import data just to create a lot of statements in Wikidata. If we import data we should always add sufficient references to the source, if necessary both as machine readable and human readable pages. That includes links that double as identifiers, but also those cases where we chose to use identifiers that is not uris by themselves.

Does this make sense? Or should we disallow tools like this? Jeblad (talk) 15:34, 24 September 2013 (UTC)

I recall a discussion on the mailing list about a tabular data datatype (csv file for example), who could be used to store water level of different places at different datatype, so we don't have to create statements about that. TomT0m (talk) 15:42, 24 September 2013 (UTC)
See meta:DataNamespace or [2], possibly more. TomT0m (talk) 15:49, 24 September 2013 (UTC)
Ripping the structured feeds apart and storing the data as static files doesn't really make sense. I think we should build on linked data structures like Wikidata itself, and link out to other machine readable sources. The question is where do we draw the line and what kind of literal data do we store locally in Wikidata. Jeblad (talk) 15:57, 24 September 2013 (UTC)
Tabular data is not unstructured datas, it's a structure to store datas that can be stored as a table and could be useful in certain cases. It's structured if the metadata about the column and lines of the tabular datas are well defined, which does not seem to be a problem into Wikidata framework. TomT0m (talk) 16:06, 24 September 2013 (UTC)
There are currently no plans to do this and I don't see this changing anytime soon. --Lydia Pintscher (WMDE) (talk) 17:33, 24 September 2013 (UTC)
Thanks for the insight, of course the dev team did not say it was on any list, just pointing the link. TomT0m (talk) 17:42, 24 September 2013 (UTC)
Why do we need tables ? We have a structure property:value, no need of table: tables can be built from the property:value structure, it is just an extraction problem to solve and this is what will be done in phase 3 to create lists from Wikidata. Snipre (talk) 00:20, 25 September 2013 (UTC)
One reason would be to simplify the management of sources (and licenses, if it comes to that). If we have a hundred data values, all from the same 25x4 table, and some detail of the cited source for the table needs correction, it would be better to be able to do that once for the whole table than having to do it 100 times. Another reason would be to enable easy editing of row or columns as a whole. If some qualifier should be added to a column, it would better to be able to do this once for the column than having to do it for every value in the column. This is not just because doing it for each value is more work, but also because it's more error-prone. --Avenue (talk) 02:33, 25 September 2013 (UTC)
If you have a table to import use a bot and you avoid the error of manual import. There are some bots ready for that job. My problem is about use of only one piece of data from that table: you will need an extra tool which will specific to table format in order to digg into the table. Snipre (talk) 07:46, 25 September 2013 (UTC)
It's not just about importing, it's about maintenance too. Yes, bots can help, but I think tabular data is common enough that native Wikidata support would be worthwhile. Regarding data access, I think the advantage of being able to simply access entire rows or columns would outweigh a little more complexity in accessing individual values, but I take your point. --Avenue (talk) 15:43, 25 September 2013 (UTC)
Not if the format is specified, ie. we know what is in column and rows. We ca even imagine that with that knowledge, Wikidata automatically translates the table into a set of statements qualified by relevant values of row and colums, ie. that a table of population of a countries city at specific year is translated into statements with the city as subject, the population property, and the year as qualifiers, and that these statements could be antively accessible by the query engine. TomT0m (talk) 15:50, 25 September 2013 (UTC)
Jeblad External data sets can in XML, cvs, html,... there is no unique tool but for each data set you have to create a tool which is specific to the data set. Each data set is different so an unique tool is necessary. And once you have the adapted tool you can start creating statements: we have the necessary structure to source so no need to ask something. Just a detail: discuss with persons having a strong knowledge in the field of the data sets about the quality of the data sets. There are plenty of free databases which are bulls**t. Snipre (talk) 00:29, 25 September 2013 (UTC)
This is about data sets published with OAI-PMH which is basically XML with some constraints. That makes it possible to create a standardized tool for mapping the data sets to new forms like the one we are using in Wikidata. But to do so require some basic rules on what kind of literal data we want. We obviously don't want to store everything, we want to store enough for our purpose and to facilitate reuse. Jeblad (talk) 05:58, 25 September 2013 (UTC)
A tool for that would be great. I think we should then discuss what to import dataset by dataset, at least until things are clearer on Wikidata, just like for any other source. --Zolo (talk) 07:15, 25 September 2013 (UTC)
I assume from this that there is no constraints on the literal data for now, except for missing properties that may depend on missing types. There are also no limit on the overall quantum uploaded from a source. There can be constraints in the future on type, value and quantum, but no one has made any attempt on specifying any such constraints as of this writing. Jeblad (talk) 13:41, 28 September 2013 (UTC)
Constraint based class definition is extensible to do that, it is actually a feature of class definition in OWL. I saw also a proposal by Ficeolaire to restrict values based on the class of the object, by it does not apply to litterals. TomT0m (talk) 17:43, 28 September 2013 (UTC)

Focus on classes.[edit]

It's kind of hard to know what Wikidatians thinks, so i'll ask a simple questions : is there a broad concensus about the idea we have

  • to switch from a property based reasoning and voting to a class based discussions, which includes : which property use with the concerned classes, how to qualify them?
  • We also switch documentation to document classes instead of properties

 – The preceding unsigned comment was added by TomT0m (talk • contribs).

  • I have no idea what this means. Sven Manguard Wha? 16:33, 24 September 2013 (UTC)
    Ok, great, something to start with :). I'll take an analogy : the infoboxes. Roughly, an infobox template like infobox scientist is the equivalent of a class on Wikidata (see Help:Basic membership properties. Any item which is is linked to a scientist, like Albert Einstein (Q937) is in the human class (and in the scientist class if that exists), for example, and the article linked to it can have the infobox scientist template in their sourcecode. What comes with the infobox template is a way to display some informations about Einstein and persons in general. Which we do not have really right now for classes ... We reason on property proposal, not in term of which properties and qualifier should I use for a scientist ? I think the latter option is really something we need, and that we should start focusing on that. That that is usually called a model for the items on that classes. Constraints can be defined such that a bot can check if any scientist fit in this model, and reports statements and items that do not fits. Questions ? /o\ TomT0m (talk) 16:49, 24 September 2013 (UTC)
    I think we should focus on associating a model to classes, understand how to express this model, how this model can help user when they edit thingd (think of a gadget that understands that model and says to you you should use the scientific domain property with one of those value for example ...
I didn't understand a lot of what you said there Tom but the bit I did understand I disagree with. For me an infobox template is not the equivalent of a class. I think you need to rewrite the above with more examples of how this might work in practice. Remember that if it is all about constraints then most people don't need to worry about it as they don't operate bots which check constraints. Filceolaire (talk) 22:41, 24 September 2013 (UTC)
@TomT0m There are too many RfCs open on subjects closely linked so it's difficult to keep an overall view of what is wikidata structure and what is the most important ~element to decide first. The main problem is we aggregated data and data structures from different WPs without any clear idea of how we will structure this data. Right now everybody is working alone in his side without thinking global and application in wikipedia.
If we can find a solution to the GND migration (which were the first seven classes) and if we can get a list of classes and subclasses, then we will be able to go further with properties definition for each class and documentation. or the best way is to create an infobox and each class and specific properties can be tested directly in wikipedia. That's the only way to create a move because right now what we are doing as just cloud sculpting. Snipre (talk) 00:15, 25 September 2013 (UTC)
@Filceolaire : of course they are not totally equivalent, but what I could say is that discussing things property by property, like we do in property creation, is roughly the same as discussing community wide every field of every infobox. And having a RfC for a class (or a couple) is exactly at the same level of having a RfC on wikipedia for every infobox. TomT0m (talk) 09:30, 25 September 2013 (UTC)
@Snipre: I think we do not need to have a clean class hierarchy before we create classes. Having a process and tools for that will realease the lock of the need to make a big decision any time we want to do this. But first community needs to understand what a class is. TomT0m (talk) 09:30, 25 September 2013 (UTC)

Members of Parliament & Congresspersons[edit]

What is the standard way of describing the position of a member of a legislative house (MPs, MLAs, Congresspersons, Senators)? I've seen it done a couple different ways. The property member of (P463) doesn't seem right because the description says it's for membership of an organization or club. The property office held (P39) seems like it should be for a specific office, like president or party whip, although it might be okay if it's qualified as "office held -> member of house of commons -> electoral district -> Anytown". However, that solution wouldn't work for legislatures with at-large members. I'm wondering if it would be better to have a parameter specifically for membership in a decision-making body (this might also include membership of councils or boards). Arctic.gnome (talk) 16:42, 24 September 2013 (UTC)

"However, that solution wouldn't work for legislatures with at-large members." Why not? -- Lavallen (talk) 16:53, 24 September 2013 (UTC)
It wouldn't work if the property "office held" was supposed to refer to a specific office because there could be a hundred people with the office "at-large MP of Anystate" at a given time. If the "office held" property can be used like that, then I'll start adding it to the pages of MPs and congresspersons. I just figured that I should first ask if a dedicated property for members of legislatures would be better. Arctic.gnome (talk) 17:19, 24 September 2013 (UTC)
I made some tests in Q518776 some time ago. To me, it looks like P463 does not fit, I would move it to P39 instead. A "property for members of legislatures" will always come to a point when I will not know if it fits or not. I think P39 will do the job for us. -- Lavallen (talk) 17:32, 24 September 2013 (UTC)
Okay, I'll use P39. Follow up question: there do not appear to be many pages for the office of member of a legislature. On en.wp, titles like "Member of Parliament of X" or "Member of Parliament (X)" redirect to either "Parliament of X" or "Member of Parliament". Should I create a bunch of Wikidata pages without associated Wikipedia articles for the titles of "Member of Parliament of X"? Arctic.gnome (talk) 17:48, 24 September 2013 (UTC)
It was pointed out to me that adding of (P642) to office held (P39) will usually work for most elected officials that represent constituencies. Circeus (talk) 18:13, 24 September 2013 (UTC)
For the ones that represent a constituency, if "office held" is good enough, I was planning to use electoral district (P768) as the qualifier. Arctic.gnome (talk) 19:43, 24 September 2013 (UTC)
NOO ! Use of (P642) because it is a general property which can be used for executive and legislative bodies. Do you think how you will extract data in wikipedia if for some political positions you have to look for qualifier of (P642) and for others electoral district (P768) ? Snipre (talk) 20:30, 24 September 2013 (UTC)
If we use "of" as the qualifier of "office held -> Member of Parliament", should we include "electoral district" on MPs' pages as a seperate parameter from "office held", or should it be avoided alltogether? Arctic.gnome (talk) 23:08, 24 September 2013 (UTC)
Sorry I have no solution right now: we need to discuss that in details but each time you propose a specific way to classify data you have to think how difficult it would be to extract that information later because the best classification system is useless if you can't use it later. We need a RfC, a new one, to fix that because right now people are doing what they want without any thought about the extraction feasability of their contributions.
Just an attempt to answer your question: why do we need "electoral district" property ? I understand now your question
  • X
office held property -> Member of Parliament property value
of qualifier -> France qualifier value
start date qualifier -> xx.xx.xx qualifier value
  • Y
office held property -> President property value
of qualifier -> USA qualifier value
start date qualifier -> xx.xx.xx qualifier value
  • Z
office held property -> Governor property value
of qualifier -> Texas qualifier value
start date qualifier -> xx.xx.xx qualifier value
  • A
office held property -> Prime minister property value
of qualifier -> Australia qualifier value
start date qualifier -> xx.xx.xx qualifier value
  • b
office held property -> Senator property value
of qualifier -> USA qualifier value
electoral district qualifier -> NewYork qualifier value
start date qualifier -> xx.xx.xx qualifier value... Snipre (talk) 23:27, 24 September 2013 (UTC)
In example Y, would the page "Barack Obama" have no direct link to the page "President of the United States"? Arctic.gnome (talk) 23:48, 24 September 2013 (UTC)
"President of the United States" is typically the worse thing to use because you have two informations in the same value: president and United States. How can a machine or a query read this kind of information ? And then if for mayor you apply a different structure where the position and location are stored differenttly this would be a mess.
In programming language there is a key rule: unique format. So either we use "position+location" in an unique property value for all political position (and we create for each mayor a specific item) or we separate position from location for all politicians (president and mayor). If you want to understand the problem just try to think how you will extract the information for a president which was once mayor. And then if you ad that this person was an senator and you add a new qualifier electoral district, then in the same code you have to figure 3 different cases, this is a mess. You need an unique structure able to be read by an unique code for each "office hold" and able to build the correct infobox line with the same elements. Without that you will need to test every possible structure for each "office hold" and this will consume an important part of server capacity. Snipre (talk) 00:54, 25 September 2013 (UTC)
I think in that case 'office held' could link to 'President of the United States' but for other less prestigious offices you could have:
  • c
Some Guy item
office held property -> President property value
of qualifier -> Bally-go-backwards municipality qualifier value
start date qualifier -> xx.xx.xx qualifier value Filceolaire (talk) 00:27, 25 September 2013 (UTC)

So, having titles like "president" and "mayor" are useful on pages, but it would also be useful to have links to notable offices, like "president of the united states". Should we just include both? Barack Obama would look like this:

office held
president
of -> United States
start date -> 2008
President of the United States
start date -> 2008
Senator
of -> United States Senate
electoral district -> Illinois
start date -> 2005
end date -> 2008
Senator
of -> Illinois Senate
electoral district -> Illinois 13th District
start date -> 1997
end date -> 2004

Does this meet everyone's needs? The one problem I foresee is how to avoid the title "President of the United States" showing up twice on an infobox". Arctic.gnome (talk) 16:40, 25 September 2013 (UTC)

I've amended this (additions in bold) so we know if he was in the senate or the house. For most administrative units, which only have one council, we can link to the item for the admin unit. Filceolaire (talk) 21:58, 26 September 2013 (UTC)

Branch discusion[edit]

We need a general decision about politician (and perhaps for other positions). There are 2 possibilities for the property office held (P39):

  1. Use the assembly item
    Advantage: all assemblies have an item office held (P39):Presidency of Bosnia and Herzegovina (Q844944)
    Disadvantage: the grammatical relation is not correct (in English and in French at least)
  2. Use/create an item for each assembly which correspond to the person. United States Senate (Q66096) -> United States Senator (Q13217683). But in that case we double the number of items associated with assembly and we will create quite a lot of empty items like this one Member of the Swiss Federal Council (Q11811941)

If this is quite easy for national assemblies we have a problem when going to low levels in the political systems like mayor: we can't create an item for each position and here the unique solution is office held (P39):mayor (Q30185) with qualifier of (P642). As this solution is already applied then we should create a reduced list of item representing the different labels for political positions like senator (Q3510907), deputy (Q1055894),... and use each time qualifier of (P642) with countries as value. But we need a solution if we want to be able to create an extraction code as simple as possible and especially if we want to avoid to keep dozens of infoboxes only to describe political positions which are a mess for persons with several activities. Arnold Schwarzenegger (Q2685) is a politician, an actor or a sportmen ? Can it be possible to create only one infobox for all persons ? Snipre (talk) 19:46, 24 September 2013 (UTC)

As the item is 'senate of the USA' rather than 'Senator of the USA' does that mean that the property should be named elected to instead of office held? Filceolaire (talk) 22:52, 24 September 2013 (UTC)

Not all public officials are elected, so that wouldn't work. For example, in Canada, senators are not elected. I think that 'office held' should still work for that, even if it doesn't sound grammatically correct... though surely there is some other solution that I can't think of :/ Ajraddatz (Talk) 22:55, 24 September 2013 (UTC)
Though, two properties would work. 'elected to' for elected officials and 'appointed to' for the rest. Ajraddatz (Talk) 22:58, 24 September 2013 (UTC)
And how do you want to extract that information in wikipedia article or infobox ? Each time you create different possibilities this means additional code lines with if loops in order to extract the information. Or you assume we will have two infoboxes for politicians: one for elected politicians and another one for appointed politicians. Snipre (talk) 23:15, 24 September 2013 (UTC)
  • You would also need "inherited" for people who automatically assumed office upon the death of the previous holder. I think the best solution would be to keep using "office held" and create pages for "Member of Parliament of X" and "Congressperson of X". Those are notable offices, and there certainy could be standalone Wikipedia articles about them if anyone cared to write them. Arctic.gnome (talk) 23:23, 24 September 2013 (UTC)
but what about all the other less notable offices - state house and state senators for states in the USA and every other federal countryin the world - must be a thousand there. We also seem to be committed to listing the mayors of every one of the administrative divisions below that - counties and municipalities and boroughs. All these administrative divisions have items but their council/congress/senate usually/often don't. Where the council does have a separate item this is usually linked to articles filled with election results. In practice though, I guess 'of (P642)' can link to the item for the council where this exists and to the item for the administrative division where there isn't a separate item for the council. In the few cases where an administrative division has two levels of elected chamber (like the 'House' and 'Senate' most US states have) will need items for each of these. Filceolaire (talk) 00:16, 25 September 2013 (UTC)
@Filceolaire Did you already try to extract data with lua code ? Because each time you create a different way to store data you need a different code to extract them: so for president of US you have a specific code, for mayor another one. I know I am annoying but I think you have no idea about what are the consequences of your choices. But even if two formats are not a big deal when you do that for professor, sportman, actor, author,... This will be impossible to handle in wikipedia. We need an unique way to store data about politician because if you take all possible cases which can be describe by "office hold" you will see the difficulty to manage this property in an infobox. Snipre (talk) 01:06, 25 September 2013 (UTC)
Learning how to use Lua is on my todo list but for now I depend on others, like yourself, to advise me on what can be done and how to arrange things to make it straightforward to do things. I did try asking for advice on Wikidata:Contact the development team but never got much in the way of useful info there :( What is the best place to start learning about Lua?
Could you have a look at Wikidata:Properties_for_deletion#station_code? There is an issue there about how data is extracted for use in an infobox and I think your opinion on it could be useful. Filceolaire (talk) 11:15, 25 September 2013 (UTC)
It's not really related to Lua in itself, the same probem would arise in templates language. There is several possibility : If we want a generic infobox, for example, for every elected people in the world, it's better to have a unique format. That said maybe a french mayor and united states president deserves a different infobox as the elections are not the same, the office held is not at all the same ... This is a choice : a generic infobox, that can handle generic information, several infoboxes specialized to a subject. This is also a shoehorn problem : trying to be to much generic could lead to overly complex infoboxes, it should be an indication that we should split the infobox. On the over hand, we should try to keep the structure as clean as possible and not make special case in the same information can be handled in the generic model. TomT0m (talk) 15:07, 25 September 2013 (UTC)

Suggestion of articles merge[edit]

I don't know if this has already been suggested and is in progress. Since I do not know how to check this, I am suggesting to merge two articles: "Folk medicine" and "Traditional medicine". I believe they are about the same thing. Sincerely, Tatiana Grehan

Following the 'Data item' link of enwp for these 2 pages (after the search engine here could only find one of these) I found 'traditional medicine (Q771035)' and 'Folk medicine (Q3495328)'. The first has sitelinks to articles in 22 languages; the second has sitelinks to articles in 5 languages but 2 of these (english and portugese) also have articles linked to 'Traditional medecine'.
If the English and Portugese wikipedias decide to merge the separate articles they have for these concepts then we can follow suit of Wikidata but until then we have to have 2 wikidata items, reflecting what is on those wikipedias. Hope this helps. Filceolaire (talk) 23:04, 24 September 2013 (UTC)
If you want to propose these articles be merged then this has to be done separately on each wikipedia as each is a separate project with it's own policies. If you want to have a go then en:Wikipedia:Merge and pt:Wikipédia:Fusão have the info on how to do this. Contact me on my talk page if you need help - just click on the 'add topic' tab at the top of the page to leave a message. Filceolaire (talk) 00:03, 25 September 2013 (UTC)
As you are stating that one topic is divided into 2 items, it is an interwiki conflict. You can use page Wikidata:Interwiki conflicts for discussion. --Infovarius (talk) 12:39, 27 September 2013 (UTC)

Category interwikis. "Person from" vs "Person born in"[edit]

I doubt there's much that Wikidata can do about this, but an example of my problem here is that the French Wikipedia's "Catégorie:Naissance à Aix-la-Chapelle" had an interwiki link to Category:People from Aachen (Q8044420), so I moved it to Category:People born in Aachen (Q7906198) along with the Russian "Категория:Родившиеся в Ахене". The Belarusian and Turkish links also had to be swapped. I quite like being able to help out when I find two items which need to be merged, but it seems that the majority of French and Russian interwiki links are pointing at the wrong English categories. That isn't Wikidata's problem per se, only we are here to sort the problem out. The French Wikipedia doesn't appear to have categories for "people from X", only "people born in X", the English Wikipedia is the exact opposite, the Russian has both.

Then I found out that both the French, Italians and Russians have categories for people who have died in Y; for example Category:Deaths in Aachen (Q9220591). Basically, looking at the categories on all local Wikipedias, this is a real headache. Does anyone have any recommendations on how to deal with these?

The really annoying thing is that some people have been correcting the links, but not adding the labels or descriptions, which of course causes problems later... I have noticed that an awful lot of users have been emptying items manually, but then not proposing them for deletion. This leaves empty items. Can't these people be warned after merging two items that the one left over should be deleted? Or is this not a problem for Wikidata in the short/mid-term? I'm not even sure if there's a guideline on this; but it seems to me why help, if you're only gonna do a half-arsed job? I guess that's the way it goes, but I'd still be interested in any tips as I have my ear on the ground! Jared Preston (talk) 19:52, 24 September 2013 (UTC)

The category issue you noted about 'people from' and 'people born in' is a headache. Yes. Sorry I haven't any recccommendations.
The people correcting the links may not be English speakers and may be adding labels in other languages that you can't see.
Leaving articles with no items and not deleting them is slightly irritating but spotting items with no sitelinks is just the sort of job bots can do while sorting tricky merges is what bots cannot do so I would not discourage these helpers.
Does that help? Filceolaire (talk) 23:17, 24 September 2013 (UTC)

Pictogram voting comment.svg Comment "Категория:Родившиеся в Ахене" translates: "Category:Those born in Aachen". It Is Me Here t / c 10:58, 25 September 2013 (UTC)

  • JFYI. In Russian Wikipedia the 2 kinds of categories have a clear sense: "Category:Persons:Moscow" contains all persons strongly factually connected with Moscow (though it can be sometimes vaguely defined, I must confess) and "Category:Born in Moscow" is self-describing. Further: categories like "Persons:Moscow" always contains as a subcategory "Category:Died in Moscow". I can't figure out what English wikipedians mean by "People from" and why they don't have "Categories by deathplace". --Infovarius (talk) 12:47, 27 September 2013 (UTC)

We need your opinion[edit]

Only one week before the end of the decision part of the RfC Wikidata:Requests for comment/Guidelines for RfC process. I can only recommand you to have a look at some of the current RfCs in order to decide if the current RfC process needs improvement. And try to answer this question: how do I know which is the state of a RfC ? Snipre (talk) 21:20, 24 September 2013 (UTC)

Please make this a more neutral notification next time. --Rschen7754 01:35, 25 September 2013 (UTC)
You are free to provide statistics about succesful RfC's, number of participants (discussion and vote) vs number od wikidata contributors or discussion duration to show the success of the current procedure. I just want to know the opinion of contributors and I ask a direct question: no propaganda. Snipre (talk) 16:58, 25 September 2013 (UTC)
Project chat is not the place to openly canvass for the success of your RFC. Please do not do this again. --Rschen7754 18:46, 25 September 2013 (UTC)
Be more precise: what do I have to avoid in the future ? Trying to attract people in order to have more opinions about one topic ? I thought there was a large consensus (see above) about too low involvement in discussion during RfC. Snipre (talk) 22:39, 25 September 2013 (UTC)
"I can only recommand you to have a look at some of the current RfCs in order to decide if the current RfC process needs improvement. And try to answer this question: how do I know which is the state of a RfC ?" appears to me to be canvassing, because it posits that the question that is being asked in the RFC is the same question as you're asking here, which is not the case. Really, the best way to avoid that problem is to keep your notification short by saying something like "This RFC is closing—please participate soon!" and just that. --Izno (talk) 23:06, 25 September 2013 (UTC)
Rschen: It would help if you had precisely identified what you perceive to be canvassing. A neutral notification is not itself canvassing, nor is it canvassing to ask for further feedback, especially prior to the closing date. It seems to me that other wikis have different rules or customs on the matter, so please be a little more understanding in the future. --Izno (talk) 23:06, 25 September 2013 (UTC)
+1 Vogone talk 23:15, 25 September 2013 (UTC)
Well, considering the recent debate at commons:Commons:Checkusers/Requests/INeverCry about posts on the German Wikipedia, as well as our past universal condemnation of canvassing at requests for permissions on this very project despite our lack of an official definition of canvassing, I think it is fairly well known what the standard on canvassing is on projects such as meta/commons/wikidata, and something that I quite frankly would assume that all of the participants here are familiar with. I would assume that anyone who has edited long enough to have any rights related to the closing of discussions should know what canvassing is, especially since canvassing only takes place around discussions. I find it problematic to have any editor closing discussions on global projects who does not understand what canvassing is on global projects. That is why every request for adminship on this project where the participant has participated in what we consider canvassing has failed, even if their home wiki does not agree that it was canvassing—and rightfully so. --Rschen7754 02:34, 26 September 2013 (UTC)

Bot for deleting orphaned Item[edit]

I was just wondering, I do find people who do merging of two item and do not request the orphaned item for deletion. Is there such a bot already running? Where if an item has no wikilink, wikivoyagelink and commons link then, it would be automatically deleted? --Napoleon.tan (talk) 04:51, 25 September 2013 (UTC)

I thinked about this again. and there are item that have no link but are notable in wikidata. i guess there should be another condition to find all those orphaned items like if a link from the item was removed and was added to another item. --Napoleon.tan (talk) 05:09, 25 September 2013 (UTC)
Well, an orphaned (i.e. no link to it) item with no link to anything else surely does not serve any purpose, does it? That's a simple enough condition for bots to delete/report such items, and merges being done manually there are just a few of them, so why even bother manually requesting deletion if such items are properly orphaned and link-blanked? -- Bjung (talk) 14:18, 25 September 2013 (UTC)
There has been a request for adminship permission for Hazard-Bot to delete items like proposed here (the discussion linked there now is archived). There was, however, no consensus to give this permission to the bot. --YMS (talk) 14:36, 25 September 2013 (UTC)
Later on there can be orphaned items at least for authors and books if we want to supply data for references in client pages. Because of this I don't think it is possible to automate deletion of items… If we say that an author should always refer at least one book, we still can have orphaned islands of items. Jeblad (talk) 09:37, 28 September 2013 (UTC)

Turning "English" in Commons Sitelink to - with a hack for now[edit]

Hi, I just wanted to let you know: I abused my admin power to add a line to the Commons.js so that it doesn't display "English" as the language for Commons for the casual reader of the page (but rather a "-"). This was not an office action, and I explicitly want to tell you that I did this without the appropriate power vested in me, so you can simply remove it again if you don't like it. (This are the two diffs: https://www.wikidata.org/w/index.php?title=MediaWiki:Common.js&diff=prev&oldid=72510307 and https://www.wikidata.org/w/index.php?title=MediaWiki:Common.js&diff=prev&oldid=72510718 )

Also, this is not a proper solution. We are working on it, and we hope to have something there for the next deployment (October 14). It also does not work when you add a link to commons and neither after you have added it, so it really sucks, but it is a bit better than before. Sorry for that! --Denny Vrandečić (WMDE) (talk) 09:26, 25 September 2013 (UTC)

What's the planned-for solution? --Izno (talk) 23:00, 25 September 2013 (UTC)
We're still figuring out the details but it should definitely not say English. Possible solutions are 1) turning English into - or multi 2) making the group a group for all such special wikis that have only one link and replacing English with Commons 3) something completely different we have not thought of yet. It currently looks like we'll do 1 for now and later 2. --Lydia Pintscher (WMDE) (talk) 07:13, 26 September 2013 (UTC)
Ideally (3) there should be no column like "language" for Commons-like projects. Infovarius (talk) 13:05, 27 September 2013 (UTC)
Right. It'd say something like "Project" or so. --Lydia Pintscher (WMDE) (talk) 22:31, 27 September 2013 (UTC)

Translation of wiki items[edit]

Hi everybody,

I'd like to know, if and where I can download a database with the translations of an item.

For example: Republic of Kosovo. With the help of the gadget "Labels list" I can easily see the translations of Republic of Kosovo with lang_iso code and translation.

Is it possible to get these information in a database. Because my workflow is to get translations for all european countries for example.


Any help welcome. Thank you very much. -- ch

If you want to get the list of all label list for each european language. then maybe you should try to learn the wikidata api. You can call the api using a normal web browser. then choose a format to get the data. for example in xml format. If you know a little bit about programming you can parse those XML to get the translation for all language.

--Napoleon.tan (talk) 15:11, 25 September 2013 (UTC)

Wikibase dates in Lua[edit]

Do we have any built-in function or something to make it easier to convert Wikbase dates like +00000001809-02-12T00:00:00Z to something loo~king more natural. It seems that the #time parser is confused by the leading 0s. --Zolo (talk) 11:40, 25 September 2013 (UTC)

I guess you already have seen the solution in eu:Module:Wikidata? It does not solve all cases, but it's a start. -- Lavallen (talk) 17:00, 25 September 2013 (UTC)
Thanks, I had not thought of looking into the Basque module ! Now at least I see the logic (apparently, we need to write the code to parse the date). Once it is parsed we can actually send it to another module like fr:Module:Date. --Zolo (talk) 21:15, 25 September 2013 (UTC)
Beware that you have to be careful with years before 0. Year 0 does not exist, but it does in the time datatype. -00000001809-02-12T00:00:00Z, therefor have to be interpreted as year 1810 BCE. -- Lavallen (talk) 12:05, 26 September 2013 (UTC)
Thanks for the hint. I have made a new version in fr:Module:Wikidata/Test. For now, it only works when precision is no greater than a day and no smaller than a year, but at least it seems to work, see fr:WP:Wikidata/Bac à sable. --Zolo (talk) 14:58, 26 September 2013 (UTC)
or string.sub(d, 9, 9) == '0' then will be true for all years before 1000. or string.sub(d, 9, 12) == '0000' then will identify year 0 but not year 537, I think. But also that will be positive for year '10000'. -- Lavallen (talk) 17:05, 26 September 2013 (UTC)
Oops, fixed thanks. --Zolo (talk) 17:47, 26 September 2013 (UTC)
Is it possible to import modules from other sites? For example, if on en-wikipedia, is it possible to import a Wikidata module like
local WBHack = require( 'd:Module:WBHacks' )
If yes, we should perhaps provide modules for common tasks, in this case: Module:Time to format Wikibase dates considering precision, before, after, remove leading 0s and so on. A Wikipedia chapter then only needs to define for example the default format like "June, 12 2013" vs "12. June 2013" and use the rest of the logic from the centrally developed Wikidata module.  — Felix Reimann (talk) 10:33, 26 September 2013 (UTC)
I've begun some month ago a module Time that is done to provide a nice interface to manipulate date and time in lua modules. It supports the Wikidata time format. If someone want to improve it (the user friendly output haven't been written yet) he's welcome. Tpt (talk) 19:29, 26 September 2013 (UTC)

Qualifier for astronaut mission[edit]

Do we have generic qualifier which could be used for astronaut mission to indicate mission as launch or return? Example: w:ru:Быковский, Валерий Фёдорович. --EugeneZelenko (talk) 14:17, 25 September 2013 (UTC)

What about spacecraft launch date (P619) and spacecraft landing date (P620) ? See their talk pages too. - LaddΩ chat ;) 22:16, 25 September 2013 (UTC)
These properties for mission items :-) I think something like generic type qualifier could be shared for similar purposes. --EugeneZelenko (talk) 14:28, 26 September 2013 (UTC)
instance of (P31) is the qualifier you are looking for. Filceolaire (talk) 21:42, 26 September 2013 (UTC)

Help[edit]

I can not link the greek (el) wikipedia category [[Κατηγορία:Ιστότοποι της Ελλάδας]] with the item Q8500430 as I can not see the "edit" in ``Wikipedia pages linked to this item``. I have been blocked? --Vagrand (talk) 16:17, 25 September 2013 (UTC)

I have the same problem to add the main article ``National Archaeological Museum, Athens`` in item Q6483858.--Vagrand (talk) 16:44, 25 September 2013 (UTC)
I was able to add the link for the first item. I'm not sure why the edit link is not working for you, although I think this issue has happened before. The Anonymouse (talk) 16:50, 25 September 2013 (UTC)
It has. Javascript occasionally just fails to load for some people, or worse, loads improperly and your edits wind up showing as your IP and not your account. The WMDE/Wikidata staff are aware of the issue, I think. Sven Manguard Wha? 17:54, 25 September 2013 (UTC)

Understanding Wikidata - your comments on Meta for a grant proposal[edit]

Dear colleagues from Wikidata, now I have been bold and drafted a proposal for an Individual Engagement Grant: Understanding Wikidata. It is an initiative to provide material making it easier to contribute to Wikidata. I would like to ask a favor from you: could you have a look, comment, give me tips to improve it? m:Grants:IEG/Understanding Wikidata Kind regards Ziko (talk) 21:07, 25 September 2013 (UTC)

Phase 2 in Vietnamese Wikipedia[edit]

I just notices here: vi:Hạt_nhân_Linux that at least the version is automatically filled with data from wikidata. Sadly, I speak no Vietnamese.

  1. What needs to be done, to get this automatism in other languages as well?
  2. Several articles in the English Wikipedia have extra Templates for the current stable version, e.g. en:Template:Latest stable software release/Linux and also for the latest preview, e.g. en:Template:Latest preview software release/Linux. Are there Bots in place, to copy at least this data+reference into wikidata? Maybe deleting the pages at the same time, they are not needed any longer?
  3. For the software licenses, how about using en:Software Package Data Exchange (SPDX), at least on wikidata?
  4. How important are the en:Template:Infobox software boxes? I could imagine, Infoboxes for cities and states have higher priority. ScotXW (talk) 22:50, 25 September 2013 (UTC)
Neither do I speek Vietnamese
1. The vi:Template:Thông tin phần mềm contains {{#property:P348}}, that makes the magic of automatic updates from Wikidata. I am not recomending that, I would instead propose you to develop a Module to call for the content in the Wikidata-items instead. It allows you today to be more flexible in what information is distributed to Wikipedia.
2. We have a lot of bots importing information from Wikipedia. Exactly which bot importing what am I not aware of.
4. Creating templates for Wikipedia and filling the items here with content is done by the active users and their bots, so the priority depends on their interest! Join us, and you will influence that priority!
-- Lavallen (talk) 10:02, 26 September 2013 (UTC)

Change property proposal process[edit]

I think it would be better if property proposal was not classified as they are now, but were linked to a class, such as rocket (Q41291). The property proposals should be discussed on the discussion page of rocket (Q41291), and announced on property proposal pages. Any thoughts ? TomT0m (talk) 09:22, 26 September 2013 (UTC)

Can you give an example of such a request, using, say, any existing attribute specific to that "class" rocket (Q41291)? - LaddΩ chat ;) 23:36, 26 September 2013 (UTC)
I think we should CREATE ONE PAGE for each request, such as Wikidata:Property proposal/<property name in English>.--GZWDer (talk) 12:19, 27 September 2013 (UTC)
I just checked some article about space rockets, like Saturn 1, and I found that low earth orbit workload would be a property that would fit (maybe it would fit another type of vehicles but it seems specific enough).
The procedure would look like : I want to add that property, maybe with several other one. I go to [[3]], where they might be a list of properties that fits, or more. I put a template, say {{Property addition to a class|max low earth orbit workload}} if it's a new property, or {{Property addition to a class|P001|P002|...}} if the property already exists. The template puts maintenance category on these page, and a bot fix the announce in some of the project pages. The asker might add some additional information about the property : how much we should expect (1 by launcher model), zéro or more for the number of launches with that type of launches, what qualifier we should expect on statements with that property ...
But the example might not have been chosen well enough : I should have chosen the launcher type item. Ariane 5, for example, is a rocket launcher type. It's also a subclass of rocket launchers, and all the actual Ariane 5 rockets are instances of that subclass. TomT0m (talk) 15:35, 27 September 2013 (UTC)

patroller[edit]

we need urgently patroller, yesterday and today I checked several ip contributions and about 50% are wrong or vandalism --Rippitippi (talk) 16:34, 26 September 2013 (UTC)

I think we urgently need better abusefilter. I have just reverted some unconstructive edits. Matěj Suchánek (talk) 17:23, 26 September 2013 (UTC)
Abusefilter don't block this for example [4] --Rippitippi (talk) 18:07, 26 September 2013 (UTC)
Abusefilter cannot block everything but I can see that many edits would normally be blocked by a filter but they are not [5] [6] [7] [8] Matěj Suchánek (talk) 18:19, 26 September 2013 (UTC)
[Off topic] why are those (etichetta, descrizione proprietà) in Italian language in Rippitippi's diff? --Stryn (talk) 18:24, 26 September 2013 (UTC)
Same here. Maybe because Rippitippi's language setting is set to Italian (although I don't know why that is affecting how it appears to other users). Regardless of why, it's definitely a bug. The Anonymouse (talk) 03:45, 27 September 2013 (UTC)
By far the most edits by non-autopatrollers are simply adding sitelinks. Other than removing sitelinks, changing labels/descriptions or doing stuff with statements, this is a very low-risk kind of edit. Of course it will happen that someone links an article to a meta-page (help page) item, or a biography to a disambiguation page item, but this does not have any tragic effect (other than e.g. removing a sitelink, where the connection might get totally lost if this edit isn't patrolled in time) and it can be detected easily later on, by Wikidatans, by a normal Wikipedia user, or by some script, bot or database report that detects conflicting statements or something the like. So what about automatically patrolling this kind of edits? Then the ones that really need patrolling would be much more obvious, and it would be a much more doable task to patrol all of them. (I don't know how this could be done technically. A bot would be a possibility for sure, but maybe there could even be a configuration that those edits don't need to be patrolled in the beginning). --YMS (talk) 18:56, 26 September 2013 (UTC)

Need of mass deletions[edit]

There are probably a few thousands empty items left after I moved the sitelinks to the proper item (which shouldn't have been created in the first place, but that's a different story).[9] Are empty items deleted automatically after a certain time, or is it best to provide a list for someone else to perform the mass deletion? -- Edinwiki (talk) 10:25, 27 September 2013 (UTC)

You can start a deletion request, but in reality empty items doesn´t matter enough to care about them. --FischX (talk) 12:30, 27 September 2013 (UTC)
Add deletion requests at Wikidata:Requests_for_deletions. Filceolaire (talk) 15:50, 27 September 2013 (UTC)

Need to create My town near by Madurai[edit]

hello sir,

                  i just want to create a info about my town (Silaiman is located in Madurai District.but i couldnt

please help me to create it

en:Madurai_district#Divisions lists 13 talukas in Madurai district, none of which seem to be called Silaiman. ta:மதுரை_மாவட்டம் lists 7 but again no Silaiman. To create a new page on either of these ask at the project chat page for that wikipedia - these are listed at Wikipedia:Village pump (Q16503).

en:Wikipedia:Teahouse is another good place for beginners to get help. Wikipedia:Teahouse (Q11059110) has a list of similar pages on other wikipedias. Filceolaire (talk) 15:48, 27 September 2013 (UTC)

Adding parts of articles[edit]

Hi all. I wonder whether it's possible to add a part of an article (I wanted to add en:Walker_(mobility)#Rollators to Q14164010). Regards, Trijnstel (talk) 17:55, 27 September 2013 (UTC)

No, Wikidata does not support interwiki links with anchors. Matěj Suchánek (talk) 18:11, 27 September 2013 (UTC)
But you can create a redirect on enwp to "Walker (mobility)#Rollators", and add that redirect to the page. But you have to deactivate the redirect while the link is added, and thereafter activate it again. -- Lavallen (talk) 18:23, 27 September 2013 (UTC)
Note that even if this is possible, it will make the lookup of item fail from the clients and it will make the uniqueness constraint fail. That means the simple one to one relationship between repository items and client pages are broken, and as the system is right now it can't be reestablished. You can link to a fragment on a client page, but that client page can't use information from the repository item. In short; bad idea. Jeblad (talk) 09:22, 28 September 2013 (UTC)

Asking for assistance[edit]

WD:Mineralogy task force
Property:P138 (Named after)
For example: axinite series (Q664861), named in 1797 by René Just Haüy (Q316515) from the Greek αξίνα ("axina") for "axe"
Is it possible to state "named after this language word" on Wikidata?
WD:Database reports/Constraint violations
Property talk:P484 and Property talk:P579
Is it possible to check if all minerals (item) have the first two lines (en label and en description)?
Regards --Chris.urs-o (talk) 03:18, 28 September 2013 (UTC)
You can probably used named after qualified by the language property. That's the closest you're going to get I think. --Izno (talk) 04:15, 28 September 2013 (UTC)
Thx --Chris.urs-o (talk) 04:26, 28 September 2013 (UTC)

Restricting edit to logged in user[edit]

Hello, I have been patroling the recent changes and I keep on noticing non-logged user (ip user) are the ones who keep on vandalizing the wikidata. I know that wikidata followed the wikipedia ruled to allow even non logged user to edit. But I think since data is easily vandalized, I was wondering if wikidata community could vote to require logged user to be able to edit wikidata. I know that the community wants to be open to as many people as possible but I think the same rule should not be done in wikidata since a vandal could easily destroy the wiki link.

What does the community think? Is requiring only logged in data requirement possible with the current wiki software? --Napoleon.tan (talk) 15:18, 28 September 2013 (UTC)

The wiki software is MediaWiki, therefore it is possible to prevent it. I am strongly opposed to such an idea. While it would possibly be good to prevent IPs editing, it can be damaging to the knowledgebase as it defeat the purpose of having all Wikimedia Foundation's projects be freely and openly editable. If an IP is causing problems, report them to administrators through the noticeboard, talk pages or even via IRC and we will deal with them appropriately. Vandalism occurring from IPs is not a Wikidata only problem, every openly editable project whether operated by the Wikimedia Foundation or by some third party company, will always experience vandalism. Stopping every IP editing is not the way to deal a small bunch of IPs as generalisation of every IP being 'a vandal/non-constructive' is a very bad idea and if we prevent editing for all IPs, it will badly reflect on all of us. John F. Lewis (talk) 15:30, 28 September 2013 (UTC)
I do not understand why stopping IP editing would be bad on all of us? I think that an IP may as well be annonymous because some internet provider provide dynamic IP so a vandal would just move from one IP to another. Hence you cannot ban that user properly. I am saying that the nature of data that wikidata provides is more unnoticeable than the other wiki project. For wikipedia, you can assume a vandalism when there are foul words or massive data deletion to auto detech vandalism. For wikidata, if a vandal would want to swap wiki links, it is very difficult to detect against a legitimate wiki link edit. I think the argument that because wikidata is being used by all wikipedia (for example claims), would mean more people would notice vandals immediately. But I think most wiki still do not use the wiki data in their wiki. Hence, it is harder to detect wiki vandals. Also, for other wiki project, people go to wiki articles to read the articles once in a while, while in wikidata, people do not go to wikidata page to read them. --Napoleon.tan (talk) 16:51, 28 September 2013 (UTC)
I am about 99% sure that the WMF will not allow this. The ability of anonymous users to edit is part of the principle that anyone can edit. IPs are capable of making constructive contributions (as are most IP edits I've seen).--Jasper Deng (talk) 16:56, 28 September 2013 (UTC)
Yeah I guess that they will not allow this. But I think people would inevitably miss some vandals. --Napoleon.tan (talk) 17:19, 28 September 2013 (UTC)
Jasper Deng is correct. The ability of anonymous users to edit is a founding principle. However, we should (and do) find ways to combat vandalism, of course. There are a number of pre-existing tools for this (e.g., AbuseFilter, FlaggedRevs, etc.), but additional tools can also be implemented. --MZMcBride (talk) 18:03, 28 September 2013 (UTC)
I am actually wondering if WMF has done any cost-benefit of allowing using ip edits. For wikipedia, or other article wiki, I understand that some people that are shy and very private may opt not to create article if they are not anonymous. Hence the benefit of them being able to write their knowledge and share to everyone points on the positive side even with the treat of vandalism.

However, for the wikidata. I do not see that data being withheld and not shared freely to the world just because people are shy about their privacy. Data is data, and it would be objective or non emotional as long as there is claim. --Napoleon.tan (talk) 17:37, 28 September 2013 (UTC)

No, Wikidata content is disputable just like any wiki content. For example, POV-pushers might edit war over the nationality of a person. As with any public wiki, creating an account might take time, and anonymous editing allows hands-on experiences (as John pointed out).--Jasper Deng (talk) 18:10, 28 September 2013 (UTC)

I have personally been wondering if we should go to flagged revisions. But then, I fear that we may not be able to keep up with the reviewing backlog. --Rschen7754 19:18, 28 September 2013 (UTC)

Languenge link: new system is bugged (or absurd)[edit]

As for the strawberry with Polish pages (see above), in Italian wiki we have "Ambasciata" (Embassy) and "Missione diplomatica" (Diplomatic mission), while in English wiki there is only one page for the second. The new system to add a language link wants a correspondence 1to1 among the pages so it does not permit me to add a language link to "Ambasciata" page since "Diplomatic mission" is already linked to "Missione diplomatica". This is a really annoying thing, since there a lots of examples where in a language there is only page about an argument, while in an other language the argument is handled in several pages (for example, TV series). This bug has to be fixed soon. SkZ (talk) 19:09, 28 September 2013 (UTC)

This is by design. The phase 2 of Wikidata associates information (data) with a particular entity. That means that every entity needs to have a unique linkage to its description on an external wiki. --Izno (talk) 23:03, 28 September 2013 (UTC)

commons category links[edit]

When adding a commons category to an item i can either use the property commons category or i can link to it in the "Wikimedia Commons page linked to this item" section. Which should i prefer? I mean a category is also a page, so either the property could be any page instead of just categories or it's not needed at all anymore now that we have the section? Mutante (talk) 19:38, 28 September 2013 (UTC)

Abraham of Bulgaria[edit]

In attempting to add the Russian Wikipedia article to the side bar of Abraham of Bulgaria, I encountered the following error:
Site link Авраамий Болгарский is already used by item Q4056008.

I don't know what is happening here are the Russian Wikipedia link is definitely not showing up under the languages side bar of the English Wikipedia entry.

Is anyone able to work out what the problem is? Thank you, in advance, for any assistance rendered! --Iryna Harpy (talk) 23:00, 28 September 2013 (UTC)

✓ Fixed One of the items had a Bulgarian link to the article, and the other item had a link to a redirect to the same article, creating an interwiki conflict. The Anonymouse (talk) 23:18, 28 September 2013 (UTC)
Retrieved from "http://www.wikidata.org/w/index.php?title=Wikidata:Project_chat&oldid=73186878"