Romaine,

thank you for that insight into the Dutch Wikipedia. This is really important! Thank you a lot.

All of this sounds very, very positive. I hope that the community will remain as positive (and yet cautious) towards Wikidata as they are.

Again, thank you for making this really insightful write up.

Cheers,
Denny



On Tue, May 19, 2015 at 11:13 AM Romaine Wiki <romaine.wiki@gmail.com> wrote:
Hi Egon,

The main use of the codes is to test it in the Wikidata sandbox or in the user sandboxes. Other usage is so far I know not implemented.

The basic reason for that is that there is no consensus about the usage of Wikidata data in templates on the Dutch Wikipedia. The usage of properties in templates is considered a big change, and big changes need consensus first, as it is not easy to go back if it gives problems or unwanted situations, and it has impact on many users.
The usage of the codes directly in articles is not the what parser functions are for, parser functions do not belong in articles. Usage of these codes directly in articles is not accepted.
If there is consensus of using Wikidata, the question also will be where this would be welcome and where not. And also which data may be used. (In the past we had for example discussions about the number of inhabitants, and there it came up that CIA Factbook data is is not allowed for quality reasons.) Also some guidelines need to be defined to avoid problems and to assure quality.

The focus of the Dutch community is (concerning Wikidata) currently mainly to get all the articles with basic statements on Wikidata. This we do by hand to make sure all the basic statements are actually there, as most of them can't simply be added by a bot. Also this makes sure that there are no/less duplicate items on Wikidata. As Wikidata is pretty new, most users need to get a bit acquaintanced with Wikidata in the first place.
In April a user organised a voting, which resulted in the situation that all local interwikis have been removed from all the articles. This was completed in April. So we do not accept local interwikis any longer.
When all local interwikis were gone, I wrote in the central discussion page a call for all users to add there articles on Wikidata + to add certain basic statements as described in the message. Since then, a group of local users is working on getting the number of not connected articles in Wikidata down. From the 4000+ unconnected articles already 75% have been done.
At the same time, new articles have been checked for being linked on Wikidata. If users forget to add their article, we add their article for them on Wikidata and a personal message with basic information is added to the user talk page who created the articled. The message says that maybe you have been forgotten, or was still intending to add the article to Wikidata, then the message is not needed, but it is needed that the author of an article adds his/her article to Wikidata. In the message is also described why this is needed, where someone can find links, and how to add an article to Wikidata.
There are two types of messages: one for users who should have added their article to an existing item (assuming the existing item already has the basic statements), and one for users who should have created a new item with the basic statements.

The basic statements include instance of (for everything), (for places and objects:) coordinates, country, located in the administrative territorial entity, (for people:) gender, date of birth, place of birthdate of death, place of death, occupation, (music:) performer, date of publication, (animals/plants:) taxon rank, taxon name, parent taxon, and anywhere if exists: Commons category. These are the basic statements for the most written articles. With these statements it is possible in case of multiple items with the same label, to add a sitelink to the right item and to easily check if an item is added to the right item.

By adding their own article for them to Wikidata with the basic statements, they have a good example (close to home) what is expected. We ask them with the next article to do it themselves. Also, something that is important, I say they can come to me with questions any time. Adding a sitelink mostly works (only 3 people do not want to do this or find it too difficult). All other users which have been informed, add their articles now to Wikidata, mostly, if they do not forget as they still get to be used to it more. Only authors that only write once in a while need to get informed. The last group of users, the new users, should then still be acquaintanced with Wikidata, but that is something that is probably only useful when they manage to write an article that follows the conventions of Wikipedia.

But when informing the users about it is needed to add their articles to Wikidata, the most heard reaction is "I didn't know".
The second most heard reaction is that they do not find it easy or intuitive to get to Wikidata, add the article and add some statements.

This feedback is given, but having users add their article on Wikidata is a great success. We notice that users like it to have someone who looks over their shoulder and helps when it is not added properly. In this way already some users have grown to do it completely by themselves.

But I also see the signals of users that they are afraid to be asked to do more than that. The usage of Wikidata for interwikis is now accepted, to add some statements has a large understanding as they see the benefit of this. But doing more than that seems mostly not needed, and also seems to go beyond the maximum of acceptance. That some users are capable of using codes is great, but the large majority must feel themselves comfortable with codes as well to be able to allow it, and to me it seems they mostly are not comfortable with it. It seems for most users a bridge too far.

And having automated lists with data directly from Wikidata seems also not acceptable in the article namespace, reading the reactions in a current discussion. Automated lists created by the software seem much like special pages which have a special namespace. Being able to edit an article (lists included) is considered to be very important for the functioning of Wikipedia, if it is not a basic rule for Wikipedia.

If such automated lists are wanted, it would be more likely to have them accepted if they are added only to a (new) special namespace for automated lists.


And yes, Lydia, has a point, as nlwiki is an early wiki where the arbitrary data has been made available, users from other wikis have added codes to articles, which have been undone as such is not acceptable on nlwiki.

Periodically multiple times all parser functions and magic words have been removed from articles on nlwiki as they are not acceptable in articles, and are considered to limit and disturb the ability to edit Wikipedia too much.


Greetings,
Romaine










2015-05-19 14:41 GMT+02:00 Egon Willighagen <egon.willighagen@gmail.com>:
Dear Lydia,

On Wed, May 13, 2015 at 5:20 PM, Lydia Pintscher
<lydia.pintscher@wikimedia.de> wrote:
> The rollout of arbitrary access on Dutch Wikipedia

Is there an overview of Dutch WP pages where it is being used? The
Berlin/Germany use case experienced resistance, needed further
discussion and consensus first? Has it been adopted on other Dutch
pages? How was it received there?

Egon

--
E.L. Willighagen
Department of Bioinformatics - BiGCaT
Maastricht University (http://www.bigcat.unimaas.nl/)
Homepage: http://egonw.github.com/
LinkedIn: http://se.linkedin.com/in/egonw
Blog: http://chem-bla-ics.blogspot.com/
PubList: http://www.citeulike.org/user/egonw/tag/papers
ORCID: 0000-0001-7542-0286
ImpactStory: https://impactstory.org/EgonWillighagen

_______________________________________________
Wikidata-l mailing list
Wikidata-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikidata-l

_______________________________________________
Wikidata-l mailing list
Wikidata-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikidata-l