User talk:GZWDer/2015

Hello. How can I request items for deletions? sorry for the mess. Mjbmr (talk) 05:35, 29 January 2015 (UTC)Reply

Why do you create new items without a label? (e.g. https://www.wikidata.org/wiki/Q19050558) Now it will be very hard to find these items. Michiel1972 (talk) 12:30, 9 February 2015 (UTC)Reply

@Michiel1972: Dexbot will add labels to these items [1].--GZWDer (talk) 13:01, 9 February 2015 (UTC)Reply

It would be cool, if you'd try to add at least one statement to each item created. --- Jura 17:11, 10 February 2015 (UTC)Reply

Q5849863 is a good place to visit, but there're too many authors to work on...--GZWDer (talk) 02:09, 11 February 2015 (UTC)Reply

Yeah, maybe. I generally try to create items for a category of specific field and then add a statement that applies to all of them. The problem with categories in some languages is that eventually subcategories include anything ;) --- Jura 05:37, 11 February 2015 (UTC)Reply

Hello, thak a lot for your help. I did need a hand to know if I am doing a good job. However, I want to improve my contributions here, but there is something I don't understand: Wikidata:Wiki import WikiProject. What I have to do there? Should I create new elements? Merge them? How can I help there? Thanks :) Sed you a big hug.- 你可以说 BlackBeast Do you need something? 00:07, 17 February 2015 (UTC)Reply

谢谢. Fix it alreay. Let me know if I might change something else.- 你可以说 BlackBeast Do you need something? 03:21, 23 February 2015 (UTC)Reply

Hi GZWDer, I understood that some test editions were necessary before requesting permission. I've just tested that the bot was able to edit, but I do think editions related to the actual work are necessary. Could you please explain me what should be done? Thanks --Discasto (talk) 10:52, 25 February 2015 (UTC)Reply

Sorry, I misread the policy :-( --Discasto (talk) 14:08, 26 February 2015 (UTC) PS: I'll ask for permission once I have the code readyReply

Hello,

I just stumbled on massive import of fr.ws texts that you did with your bot in february, without even noticing the wikisource.fr community... :((

We were working on a policy to import only validated and corrected texts, and to add claims on them methodically, so that it was possible to work, but with that mass import, you really flooded our capacity to treat those items (we're a very small group), and we have no tool to extract correct infos from wikisource for now (it is under construction), which means that the work has to be done manually !!!

Do you really think it is good practice to do that, without even noticing, and I discover it 5 weeks later ? Do you even understand French, do you have real possibility to help us now, or do we just have to deal with your crap...

I have the impression that a library has just been emptied on my doorstep, and now, I have to order and clean it up alone... and I am a librarian, so, at least, I know the job... :(( --Hsarrazin (talk) 12:55, 28 March 2015 (UTC)Reply

Hi GZWDer,

Would you create this property? It has been over one week that it has been up. --- Jura 19:03, 28 March 2015 (UTC)Reply

Hi! I have open a talk about your bot here. Sincerely Miniwark (talk) 12:25, 30 March 2015 (UTC)Reply

Hello. I see you have stated about P1477 that:

Should also be replaced with monolingual-text type. See talk page..-- 10:32, 4 September 2014 (UTC)

Umm... which talk page is that? And why would youWikiData restrict names to "monolingual", when there are plenty that aren't? -- Jokes Free4Me (talk) 21:29, 12 April 2015 (UTC)Reply

Okay, and based on that impression and no other feedback, it was just blitzed in?! Okay, i guess i should be bold too and rename the two, into "monolingual birth name" and "multilingual birth name". Definitely no reason for deprecation exists. -- Jokes Free4Me (talk) 08:45, 13 April 2015 (UTC)Reply

Hi GZWDer,

Would you look into hair_color and different_from, they seem to have sufficient support for creation. ----- Jura 12:24, 18 April 2015 (UTC)Reply

What do you think of it? --- Jura 09:02, 9 May 2015 (UTC)Reply

Thanks for doing the first one. BTW Dictionary_of_.. could be ready as well. --- Jura 08:11, 13 May 2015 (UTC)Reply

Hello, According to Wikidata:Database reports/Constraint violations/P234#Format, there are a few dozen of wikidata item with the following error at InChI (P234) property: the value has the unwanted prefix "InChI=". Only a few dozen of wikidata item have currently this error at this property, this is not an issue for me to correct them. However, many of those properties where added by you, and it could be that you used an import tool with a bug, which could have made the same mistake elsewhere, or could add much more that a few dozens errors in the future (especially if this buged tool is used by other contributors during several years). Could you take a look at how tou added those properties? Visite fortuitement prolongée (talk) 19:44, 18 April 2015 (UTC)Reply

Hi GZWDer. I have a problem with ProCyclingStats cyclist ID (P1663) on Alberto Contador (Q132738) (website) that I can't explain. Jérémy-Günther-Heinz Jähnick (talk) 16:41, 15 May 2015 (UTC)Reply

Hi GZWDer, Please stop creating items for Dutch articles. This is not the way the Dutch community wants to work. Romaine (talk) 05:32, 26 May 2015 (UTC)Reply

(o) That's the way Wikidata works. How do they want to work at nl.wikipedia? --- Jura 05:39, 26 May 2015 (UTC)Reply

@Romaine: Please specify which nlwiki articles should have items.--GZWDer (talk) 05:45, 26 May 2015 (UTC)Reply

(after edit conflict) No, this is how you use it. We have asked bots to stop importing articles from the Dutch Wikipedia. Adding articles to Wikidata without any properties almost useless and creates too much extra work. Not to mention all the double created items. The Dutch community is currently working on getting all articles connected on Wikidata to existing items or to new items, with human judgement. A month ago we had 4000 articles that were not connected, now we have 1100. All by hand with statements added. Your bot is making it a mess this way, junk that is not helpfull. Romaine (talk) 05:51, 26 May 2015 (UTC)Reply

@Romaine: Stopped.--GZWDer (talk) 06:01, 26 May 2015 (UTC)Reply

Thank you! Romaine (talk) 06:03, 26 May 2015 (UTC)Reply

• If you want to avoid that someone else does the same, it might worth writing this down somewhere at Wikidata, including how to determine the numbers. This way, we can do something about it when there are 20'000 articles without items. --- Jura 06:04, 26 May 2015 (UTC)Reply

• I have thought of that but I do not see that effective before the Dutch users have completed their work. Romaine (talk) 08:16, 26 May 2015 (UTC)Reply

• @Romaine: But Wikidata:Notability allows creation of items for articles. It's my opinion that absent consensus to change that, there is no policy basis for this prohibition.--Jasper Deng (talk) 07:44, 26 May 2015 (UTC)Reply
  • Jasper, I think it's a good idea letting them try to do it first. It seems they are quite efficient about it. The main problem we got is that most Wikipedia's don't do it. I'm sure GZW could work on the others first. Besides, I agree that adding at least one statement would help. --- Jura 07:48, 26 May 2015 (UTC)Reply

(after edit conflict) That there is no policy for that, it does not mean common sense can't be used. The question should be: what would benefit Wikidata most of: a bot which is frustrating an active group of users who are working on adding items properly with statements, or let that group of users continue to add quality to Wikidata? And a second question that is connected to the first one: do you want to create much extra work or is it better to get the work done efficiently and effectively? And if you think the first option is the one to choose: are you the one who will check all the items for duplicate ones? Are you the one to add basic statements to the items? And are you the one that will clean up the mess? We want to go for quality, I hope that can be supported. Romaine (talk) 08:16, 26 May 2015 (UTC)Reply

I don't get it, after their creation items can still be worked on the same wa, or merged … Would actually this community would be happy with a tool to find items with articles in their language with few or no statements or interwikis ? TomT0m ^(talk) 08:26, 26 May 2015 (UTC)Reply

The number one frustration of users is that they can't easily added an article to an item, because it is already connected to another item with nothing on it. For experienced users it is easy to merge, for most Wikipedia users from nl-wiki it is not that easy.

Every day a group of users from nl-wiki is checking the new articles for being added to Wikidata. We make sure that new articles are all added. With each new article we check if there is already an item about the same subject or not. For this purpose we have a special page on nl-wiki what we use every day to work from. There is no other way to check new articles to be properly added. You can say that we can merge them afterwards, but who will check for that? Nobody! The only proper way of having one item only for one subject is by manually looking. Massively adding items with new articles without checking if the article already exists in other languages makes it triple the amount of work to get it right. Also all Rijksmonumenten of the Netherlands already have an item on Wikidata, if an article is written it must be added to an existing item. Also many genus articles are written currently, while those items already exist. So as said, each day we check new articles for being added to Wikidata to ensure quality and to avoid extra work, and doing this we are ahead of the other Wikipedias (citing Lydia), we do our homework and we have our workflow in order.

With the importing of interwikis from Wikipedias there is still a massive backlog of many items with none, or only a few statements, with many essential statements still missing. There much work can be done that actually will help Wikidata go forward in quality.

With the special page on our wiki we can do what we need to do, together with Autolist and Tabernacle. For the rest we have suggested Lydia which tools we miss.

So please let us do what we are good at, that would avoid much annoyance and low quality, and achieves high quality work. Thank you. Romaine (talk) 09:00, 26 May 2015 (UTC)Reply

I'm just curious, that's all :) Can you tell more about the special page ? Is this specific to your wiki (and if it is, why ???) TomT0m ^(talk) 09:15, 26 May 2015 (UTC)Reply

@Romaine: FYI: Its's very easy to merge two items using the addlink widget in client.--GZWDer (talk) 09:50, 26 May 2015 (UTC)Reply

@TomT0m: No problem. It is w:Special:UnconnectedPages. Every wiki has it, even Wikidata itself. For those wikis who have an active community who is adding articles with statements to Wikidata, it is very usefull, without any alternative. Romaine (talk) 10:57, 26 May 2015 (UTC)Reply

@GZWDer: For you and me it is easy, for a lot other users it isn't. The whole interface works great (for me, you and many other) and at the same time users experience it as too difficult. To me it is logic, for them it is abracadabra. Romaine (talk) 10:57, 26 May 2015 (UTC)Reply

@Romaine, Jura1: The bottom line is that if you want this to be an enforceable policy it will have to be discussed by the broader community. Unless the community agrees, users should not be blocked for something that is within policy. The merging of items should be made easier in the future, I agree, but that is not a reason to restrict GZWDer from making valid items.

It does go without saying that creating actual duplicates is indeed not okay and should not be done.--Jasper Deng (talk) 08:00, 27 May 2015 (UTC)Reply

@Jasper Deng, Romaine, Jura1, Tobias1984, TomT0m, Edoderoo: More than one year later, I have to revive this thread to highlight the fact that the mass import from Wikisources has been done without consulting communities (nor the Wikiproject Books, AFAIK). Wikisource is different from Wikipedia and it took some time for the (small) communities to understand how Wikidata works and how to deal with books (long story short, one item is not enough, often). I think that GZWDer showed a "dangerous" behaviour: it frustrates people who want to do things their way because they know better the projects. I'm still not sure if the damage that GZWDer has caused is fixable (i.e. if it is not much of a hurdle to work with existing items instead of creating new ones), but I'd love to see a discussion about this kind of behaviour. Aubrey (talk) 10:02, 3 August 2016 (UTC)Reply

1. No items for any subpages are created. Nor any obvious page can be linked manually.
2. Having duplicate items are better than having no items at all; these items are notable and merging items causes no lost of data.

--GZWDer (talk) 10:18, 3 August 2016 (UTC)Reply

As @Hsarrazin: tried to explain to you over than 1 year ago, Wikisource books are divide by "quality", so maybe a community could have decided to import here just the best ones. Also, right now, with already existing items, we have to use different tools to:

• clean things
• populate items with statements
• import books you didn't import.

Wikidata is already very complex, and forcing people to clean up after you meand that we have to complicate further our workflow, using different tools (SPARQL query to understand what'al already here and what's missing, quickstatements or bots)

This is why I also question your second claim: running the first batch is easy, cleaning and maintaining is much more difficult. This is why it was important to involve directly the communities: this could have been dn in one batch, but with more statements and also thinking about updatability. Why didn't you bother to ask WikiProject_Books or Wikisource communities? Aubrey (talk) 13:27, 3 August 2016 (UTC)Reply

If your community doesn't want to use Wikidata data, having these item does not harm anything. Users may still add interwiki link via widgets and existing item will be merged.--GZWDer (talk) 13:37, 3 August 2016 (UTC)Reply

My community wants to use Wikidata data, but you didn't import any data, just empty items. I know it's possible to merge or using a bot to populate existing items, but it's just more difficult. I'm sure you understand that (but probably don't care). Aubrey (talk) 14:32, 3 August 2016 (UTC)Reply

Items are just a start and data can be added gradually. Having items are better than being unconnected.--GZWDer (talk) 14:35, 3 August 2016 (UTC)Reply

Please listen: we are working on a easy way to create good items with good data. Having these items (with good data) is far better than having items with no data and no easy way to add this data!

So: please stop "connecting" us in a very unhelpful way! by adding a mountain of more work to do. (and let be known: all wikisources are working *together* in that! So please, don't even start with another one!)

We want to "be connected", but we want in a good and useful way. Just creating items is not helpful to anyone. (not for Wikidata, not for Wikisource, not for any other case of application).

Thank you, WikiAnika (talk) 19:00, 3 August 2016 (UTC)Reply

@WikiAnika, Aubrey: Please point out related discussion about the manually (instead of bot) importing of data to Wikidata in de/it (and probably others) Wikisources. I'm not going to import any new items from Wikisources.--GZWDer (talk) 19:24, 3 August 2016 (UTC)Reply

ws.conference in Vienna? Hack-a-thon, meet-ups, discussions and multiple direct talks at Wikimania? Personal mails to and from people of interest from all lang.ws? (to most of us, data is not our homewiki, others are not able to contribute in english or on meta (or in that special nerdy wikidata language)... So far it is all about direct communication with each other to get the also very different ws-projects together. We are preparing to be able to add the data with joint forces. This is not easy. Having to prepare for multiple special cases (if that, do that, otherwise ) just makes it harder.

Let us know, if you are interested to join that group adding data, after finished preparation for import. Or even to help us with zh.wikisource.--WikiAnika (talk) 08:51, 4 August 2016 (UTC)Reply

When you create a property, such as CTHS society ID (P1961), please update the relevant proposal (in this case, here) with a comment, as I did in this case. See also [2], [3] & [4]. Thank you. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:25, 26 June 2015 (UTC)Reply

I like the page Wikidata:Statistics/Wikipedia/Type of content you created. It's great. How often is it going to be updated?--Pere prlpz (talk) 19:17, 24 July 2015 (UTC)Reply

Hello, I see that you can Chinese. Can you please check the label and the description of these two properties (canonical SMILES (P233) and isomeric SMILES (P2017)) ? Thank you Snipre (talk) 13:56, 6 August 2015 (UTC)Reply

This was already prepared. I saw you removed a lot of small translation chuncks to replace them by very large one. I don't agree, this makes the whole translations a lot less maintanable imho as small chunks are unlikely to be touched, P3articipants ir here once and for all, while in you included in a large chunk it is likely to be seen and reseen other and other by tranlators at any change on the page. 08:17, 12 August 2015 (UTC) https://www.wikidata.org/w/index.php?title=Wikidata:WikiProject_Names&diff=240100596&oldid=239906449

Human Metabolome Database ID (P2057) is ready. --Tobias1984 (talk) 10:17, 13 September 2015 (UTC)Reply

luminosity (P2060) is ready. Please add examples, and start discussion about needed units and qualifiers. --Tobias1984 (talk) 11:53, 13 September 2015 (UTC)Reply

HSDB ID (P2062) is ready. --Tobias1984 (talk) 12:54, 13 September 2015 (UTC)Reply

LIPID MAPS ID (P2063) is ready. --Tobias1984 (talk) 13:00, 13 September 2015 (UTC)Reply

KNApSAcK ID (P2064) is ready. --Tobias1984 (talk) 13:18, 14 September 2015 (UTC)Reply

NIAID ChemDB ID (P2065) is ready. --Tobias1984 (talk) 13:25, 14 September 2015 (UTC)Reply

molar fusion enthalpy (P2066) is ready. --Tobias1984 (talk) 13:32, 14 September 2015 (UTC)Reply

thermal conductivity (P2068) is ready. --Tobias1984 (talk) 17:28, 14 September 2015 (UTC)Reply

CDB Chemical ID (P2072) is ready. --Tobias1984 (talk) 20:19, 15 September 2015 (UTC)Reply

boiling point (P2102) is ready. --Tobias1984 (talk) 20:54, 25 September 2015 (UTC)Reply

Hi GZWDer,

quite amazing output the other day. Thanks for your help!

BTW the SFDb properties are probably ready to be finalized. If you want to complete it? --- Jura 15:04, 14 October 2015 (UTC)Reply