User talk:Billinghurst/Archives/2021

From Wikidata
Jump to navigation Jump to search
This page is an archive. Please do not modify it. Use the current page, even to continue an old discussion.


Mary, Mary

Hello. You created Mary (Q50809379) a while back. Is it supposed to be different from Mary (Q734578) (perhaps as a male counterpart to the more common female name)? If not, they should be merged. If so, the distinction should be made more apparent. The New Q5 tool seems to prefer the item you created when populating fields, which creates confusion. Cheers, -Animalparty (talk)

@Animalparty: it was three years ago! It would have been created to a need and replicated what was current style back then when there were non-gendered, male and female gendered first names. Align it with whatever the current schema may be, simply not important to me. Maybe the tool needs to be tweaked, rather than item itself.  — billinghurst sDrewth 02:08, 6 February 2021 (UTC)

Structured data across Wikimedia is starting!

Hello there, Billinghurst! I am writing to you because a new WMF project, Structured Data Across Wikimedia (SDAW), is about to start.

SDAW is a grant-funded programme that will explore ways to structure content on wikitext pages in a way that will be machine-recognizable and -relatable, in order to make reading, editing, and searching easier and more accessible across projects and on the Internet.

The reason why I am contacting you is because of your work across the Wikimedia projects, and your opinion would be of great interest for us to kickstart the discussion. We defined a series of questions we would like to have feedback on. If you want to ask a question on your own, or just want to share with us your ideas or opinions, please feel free to do so!

Also, if you know some other user(s) that can be interested, please let me know or invite them to join the discussion. The more, the better!

Hope to hear from you soon! -- Sannita (WMF) (talk) 15:03, 20 March 2021 (UTC)

@Sannita (WMF): Thanks for the note, I will have a look. The most particular question that I will be wanting to address is how Wikisource can best structure their pages and subpages (specifically the respective namespace header templates) to have the best metadata. Also knowing that we have a file as a copy of an edition at WikiCommons; edition(s) of a work at one Wikisource , possibly translations at other Wikisources, and often no parent item, and to which WP and WQ could reference—while knowing that many old works are likely to be part of the writer's article, rather than wikipedia articles themselves. AND wikilinks at Wikidata are not.

Plus starting points and completing circles. Upload file at Commons is typically where we start and then the flow to get the data to follow easily. How I do things now is (though not all circumstance) IA-upload > Commons file > Wikisource Index page > Wikisource transcription through to main ns transclusion > Wikidata item creation using WEF framework > back to commons file to add WD item > back to WD to add IA data. Other things then done by me are add to our new texts for front page display (and add locally to various thematic pages and user pages, again no WD tie-in); usually tweet about it (local template, no WD tie-in); usually add something to WP (local template, no WD tie-in.). At times work is transcluded it is available for download as an etext [download button]. So you can see that we have plenty of avenue for improvement and application of structured data.

Conundrums, and I sometimes cannot suitably explain it in the typed medium. Oh, and s:WS:S#Google often unable to find works at Wikisource might have some context.  — billinghurst sDrewth 23:21, 11 April 2021 (UTC)

To also note that I am a rubbish cataloguer, and with a limited subset of categories and works at WS, trying to portray a volunteer provided set of works to a complex catalogue world is umm interesting and challenging. We are passionate transcribers!  — billinghurst sDrewth 23:39, 11 April 2021 (UTC)
Hey, thanks a lot for your feedback. I passed it on to the tech team, and hopefully will let you know in the future. Also, if you have other ideas or questions, please ping me or write directly on our meta page (which, actually, might be better, because the team also regularly watches the page, and there's a higher chance of a direct answer). :) Thanks again and hope to hear again from you soon! Sannita (WMF) (talk) 17:30, 14 April 2021 (UTC)

statements, properties, qualifiers

I personally started to clean up journals and their articles after reading Wikidata:How to use data on Wikimedia projects and then doing s:en:User:RaboKarbakian/WD article which caused s:en:User:RaboKarbakian/Trial and error.

You can add qualifiers to statements, if you like, but keep the statements also. The less that the data modules need to be used, the simpler template authoring will be.

My cleanup started at the point I was going to use the WikidataIB module to link "Vol. #" to an existing volume at source.

The little citation tool there is kindof cool, no?--RaboKarbakian (talk) 13:00, 5 May 2021 (UTC)

@RaboKarbakian: Try the WEF framework gadget available at enWS, way better if you are putting data in on a per page/edition level. Personally I have that globally activated for all sites, makes my life way better.  — billinghurst sDrewth 14:44, 5 May 2021 (UTC)
Preferences? Not until I can get language selection to work here, I need Fr, De, Nb; I have Es, and two varieties of Chinese and nothing affects this. Not #babel, not alternate languages.
No thoughts +/- about my template? Just an advert for another javascript inter-something: -face, -ference, -cession. It is difficult not to blame trans/inter lopers for these problems. The changes in what the editor does, the sudden redrawing of a tab into the mobile format, that this device has never acknowledged my media settings. Even now, I delete to correct a typo and it adds a space, which it does irregularly.
If you are advocating js gadgets, how do I harden javascript? How do I get languages to work?--RaboKarbakian (talk) 15:05, 5 May 2021 (UTC)
@RaboKarbakian: WEF has language switch ability, with it autoselecting language based on the language of wiki where you are adding the work. If you are wanting to add multiple headings and descriptions then there is a WD gadget for it, it worked for me last time I tried.  — billinghurst sDrewth 15:26, 5 May 2021 (UTC)
The message above this is about machine-readable. I am here because of the qualifiers you added to Made in Germany (Q81846368), which are fine, I left them, but the whole should easily fill a citation form for the article. Do you remember the discussion of deprec.?--RaboKarbakian (talk) 16:28, 5 May 2021 (UTC)
@RaboKarbakian: How I set up "Made in Germany" is the way to proper way to show the properties, they are not independent properties they all relate to and qualify the "published in", see Help:Qualifiers. Tell me how something like "page" is an independent property; example how many "page 9" are there in the all time? It makes no sense unless it is a qualifier. Just because other people have done it wrong is no reason to change to replicate sub-par usage.  — billinghurst sDrewth 23:52, 5 May 2021 (UTC)
Can you adjust s:en:User:RaboKarbakian/WD article so that it will make an article citation for that particular article? Don't worry about the second part, that pulls in the scan....--RaboKarbakian (talk) 12:42, 6 May 2021 (UTC)
Page 9 on is the page number of the first time the article was published. It would be the page number of the edition being datafied, if that were the case. Qualifier pages will be great for "has editions" and "has parts", but the first edition should be easily citable. The biologists have been passing information like this for centuries, the effort, greatly hampered by 18th century technology.--RaboKarbakian (talk) 12:51, 6 May 2021 (UTC)
I know to what the page number refers, I am not a blithering idiot. The property PAGE only has any relevance with regard to the "published in", it is not an independent statement it solely relies on the publication, that being the case, it is a qualifier for the publication.

You can have the "published in" without other components, it is therefore an independent statement; you cannot have "Page" or "Volume" as statements without reference to the publication, so at that point that means that they are qualifiers. It is only relevant qualifying the publication, and you cannot wish that away. Please do it properly.  — billinghurst sDrewth 13:27, 6 May 2021 (UTC)

I get into this "fugue" (I think that is the word) where I try to make beautiful data pages. That all changed when I tried the simple task of using that data page for its purpose, which is to, at the very least, make a citation for its subject. The qualifiers make human reading easier. Machines can get thd info from a list, where real eyes can use the help when looking for one of many.
I think at best you have that fugue. Argue logic with a version of my template as these data are for machines. At worst, you failed with the template and are taking it out via chat to a nobody. Either way, your logic argument is surely lacking in the machine-readable department.--RaboKarbakian (talk) 00:29, 7 May 2021 (UTC)
If you are not handling qualifiers, then the issue is with your design and implementation, not my logic. Those elements belong to the publication, and you are not addressing those aspects, and your denialism is troubling.  — billinghurst sDrewth 03:50, 7 May 2021 (UTC)

Help request

Hi. I saw your correct opinion at the deletion request discussion of Erceg (Q5385187). There are some users who merge name items with disam pages, like User:Araz Yaquboglu. (I am pinging them so that they will not fail to think I am speaking "behind" them.) I am very disappointed with cases like this and this, which are only a couple of examples of great mistakes. The first one to the issue you are concerned about (mixing and merging a name item with a disam item). The second one is about "how to destroy" a name:given and family name item. I give only these two examples for two different kinds of disruption, examples are not so few. (Please also have a look at the talk page of the last referred item.) I need help from colleagues not only to revert but also to stop these mistakes..

Look at Güllü Yoloğlu (Q12839922) also, please. Due to the former error above, she has a red link surname in Commons. (I mean mistakes here reflect in other places also...) Thanks for your time and interest. --E4024 (talk) 01:05, 27 May 2021 (UTC)

@E4024: I am guessing that you are having some troubles somewhere to end up on my talk page, as I am not typically involved in issue resolution here at Wikidata. Hsarrazin is a far better exponent of given and family names and for resolving the big issues, I just trip over things at some time and have an opinion. I reverted the first sample that you pointed to as merging disambiguation pages and family names is wrong and problematic; and I have left a comment on the user talk page of the merger. Of course where there is a systemic problem, then alerting the talk page of Wikidata:WikiProject Names is the best course of action.  — billinghurst sDrewth 06:48, 27 May 2021 (UTC)
Hi again. I'm not having or interested in problems somewhere in internet (I have real life problems that I have to attend); as already stated, I ended up in your TP due to your comments at the deletion request discussion of Erceg (Q5385187). Without having looked at what you did about the above issues, I thank you very much in advance for sparing time to them. I have no other requests from you. If I can help you with anything, such as translating something from Turkish almost better than Google, please do tell me. Take care. --E4024 (talk) 13:55, 27 May 2021 (UTC)

Call for participation in the interview study with Wikidata editors

Dear Billinghurst,

I hope you are doing good,

I am Kholoud, a researcher at King’s College London, and I work on a project as part of my PhD research that develops a personalized recommendation system to suggest Wikidata items for the editors based on their interests and preferences. I am collaborating on this project with Elena Simperl and Miaojing Shi.

I would love to talk with you to know about your current ways to choose the items you work on in Wikidata and understand the factors that might influence such a decision. Your cooperation will give us valuable insights into building a recommender system that can help improve your editing experience.

Participation is completely voluntary. You have the option to withdraw at any time. Your data will be processed under the terms of UK data protection law (including the UK General Data Protection Regulation (UK GDPR) and the Data Protection Act 2018). The information and data that you provide will remain confidential; it will only be stored on the password-protected computer of the researchers. We will use the results anonymized (?) to provide insights into the practices of the editors in item selection processes for editing and publish the results of the study to a research venue. If you decide to take part, we will ask you to sign a consent form, and you will be given a copy of this consent form to keep.

If you’re interested in participating and have 15-20 minutes to chat (I promise to keep the time!), please either contact me at kholoudsaa@gmail.com or use this form https://docs.google.com/forms/d/e/1FAIpQLSdmmFHaiB20nK14wrQJgfrA18PtmdagyeRib3xGtvzkdn3Lgw/viewform?usp=sf_link with your choice of the times that work for you.

I’ll follow up with you to figure out what method is the best way for us to connect.

Please contact me using the email mentioned above if you have any questions or require more information about this project.

Thank you for considering taking part in this research.

Regards

Kholoud

@Kholoudsaa: Happy to talk. Though, I can say that the vast majority of my work here emanates from work at Wikisource. Things like s:Thom's Irish Who's Who <=> Thom's Irish Who's Who 1923 (Q60485568) and special:WhatLinksHere/Q60485568 for all the biographies. Author pages at Wikisource will get researched and then get their data here, and we extract as much as possible back. enWS pushes our metadata here to prevent rotlink, and interlinking works (enWP links) <=> editions (enWS), and you need to understand the complexity due to WD:Books, which leads to data fixes for the interwikis. I also do anti-vandalism, anti-spam, and anti-conflict of interest edits, so they have an impact here. Also check the potential merge reports for number of English language wikis, though that is say six reports once a week, and that is pretty well low counts these days. [Preface all of that with multiple admin rights across wikis.]  — billinghurst sDrewth 13:48, 27 May 2021 (UTC)

Mechel

Hello, I've seen this rollback; sorry, but have you seen what is there in the Commons category? It's about the village, not the mining company. --Syrio posso aiutare? 08:20, 8 June 2021 (UTC)

Hi, I was unpacking the other linking through to the Category:Mechel (Q8614994), which was all wrong. I have properly attached the Commons Cat which was otherwise unattached to the item.  — billinghurst sDrewth 12:43, 8 June 2021 (UTC)
Ok! --Syrio posso aiutare? 16:05, 8 June 2021 (UTC)

Alys Lowth

Thank you for adding the biographical information about Alys Lowth; could I ask what your reference was for birth and death dates, baptism, various different names and so forth? —Giantflightlessbirds (talk) 10:43, 11 June 2021 (UTC) Oops, no I can see the references you've added to the Wikisource author page; many thanks. —Giantflightlessbirds (talk) 10:45, 11 June 2021 (UTC)

@Giantflightlessbirds: Typically I will put the refs on the author's talk pages at enWS, though very occasionally if I am doing research for others, then I will put them at enWP; or the item talk page. Always happy to dig up such information if you need it.  — billinghurst sDrewth 11:06, 11 June 2021 (UTC)

article links and a question about tools at commons

s:en:The Land of Enchantment#7 tells where the first publication was. If the journal is ever published at source, where does the article link to source go?

Perhaps you can stuff 5 to 10 automobiles into a garage, but if you cannot get them out again then all you have is a garage with a bunch of stuff in it and not really having automobiles. Where will the [[en:s:Young Folk/Volume N/Ben the Sailorman]] link go?

About tools at commons, do you have tools available to move images into and out of category that I do not have access to? I am simply making the categories there match the Main at source better. To me, "Illustrations for" at commons is a silly redundancy, like "data at wikidata" or "source at wikisource" and if you have tools that will more quickly do this than I do, can I ask some moves of you?

Well, that is a lot of ifs. But the most relevant "if" is the answer to where the link will go.--RaboKarbakian (talk) 12:50, 12 June 2021 (UTC)

@RaboKarbakian: My understanding of the model is that where a written work by an author that is not implicitly part of another work (as happens in compiled works), it can have its own item as a conceptual item (linked to enWP) and a separate/physical item that is the edition item linked to enWS and these are joined through is<->has an edition. For editions, the first published information is interesting, though not directly relevant to this edition's metadata, that information is pertinent to the conceptual, and that other edition.
Re automobiles. If the metadata mechanics is not working, then we fix the model's mechanics not put in (dis|mis)information to fix up what is not working. It may take a little while, and that is why phabricator: and the Wikidata tag exist to line up fixes, or upgrades to Module:WikidataIB, etc. This is a wiki, these things will develop, so put the right data now, and the fix the model, rather than wrong information that needs fixing later. All editions at enWS of "Ben the Sailorman" will point to the conceptual item for Ben the Sailorman which will also be a versions page at enWS when we have more than one edition. (Our versions pages link to conceptual pages, which interwiki to enWP as pages exist)
Re Commons, there is a bulk category move tool there, check c:Com:Tools (I have it loaded through my global.js at metawiki). There is also the ability to request bulk moves at c:User talk:CommonsDelinker/commands. Always seek consensus for bulk moves that may be contentious. I wouldn't ever use "Illustrations of ..." in a nomenclature, 1) because it is a hard search through HotCat, 2) redundant for my needs; though cannot be definitive if use case without looking at the cases, there may be a valid reason that it is used in places. See c:Com:Category name for guidance.  — billinghurst sDrewth 00:56, 14 June 2021 (UTC)
@billinghurst: thank you for your explanation, I confess, it is going to be a challenge for me to know when one situation applies or the other. I am forward looking (having had to update many things throughout my years) and if there is a chance that a publication might appear at source and the original article with it, then in my minds eye, they each get different data. There is a book "The Princess Porquoi" (or something like) whose short stories were all published in Scribners or Atlantic Monthly or etc. and the space is there for those originals and the images are different (the magazines had colors (not colours, as USA). So, I think that any short story or scientific article should be treated the same, with or without the images or the already existing wikisource link. The instructions and understanding of how to unfurl the data when needed (move the automobile out of the shared garage) also seems, without it existing yet, confusing.
About commons, I can recat better without the tools, although, hotcat was nice on the touch screen. The admin there have not had a problem with my reorganization of things and that they match the source source better is probably a plus overall. It is nicer, really, having Cat/Subcat. Anyways, you caught me trying to toss off a dreary job on you. There is a bot I could "trick" into making the moves for me, but not all cats are to be cleaned out entirely, so that would not work.... ENOUGH typing through this problem on your talk page. Apologies.
Is there anything that you can do to make the differences and when one applies or the other easier to understand? Also, nice style on the TOC at source. It looks much better! --RaboKarbakian (talk) 02:18, 14 June 2021 (UTC)

WSes throws up so many variations to WD due to the diversity of what we reproduce. I am not the best at explaining it, there will be others that are better.

Any published work, any short story and any article could have a dual conceptual and physical, it is just going to be total overkill and take you away from transcribing if that is what you do. So, for me it is best value effort, so those works most likely to be replicated are the fiction works as you described, so they are more likely to have that dual set of items. Single use/throw away representations will get the one as "life is too short". I don't understand your example about illustrations vs colour, etc., I reproduce the work as is and record the data as is.

The use of both wikidata and metadata for enWS, or for pulling enWS data into other fora is definitely imperfect, and all we can do is pound and pound and pound the developers. It comes in time. It took five years but now we have w:en:Template:cite Q and while it could be better that is about the pounding, and remembering "it is a wiki", we are volunteers. From where we were 13 years ago when we started this is a vast improvement.

Whereas a mini-biographical entry in many works will unlikely ever see a reproduction, so will only have the one physical item, and be part of a larger conceptual item of the work in which it was published. There often will be many mini-bio through multiple works.

If it has a physical representation, which is the case for enWS, it will be part of an edition of a work. It may also be part of some other set of metadata that sits outside our edition. For example, an edition item linking one of Dickens serialised works can pick up the parts from multiple of editions of a journal, and each of those is an article of those issues. It is all bloody tricky when we get out of the single physical.

Re Commons, it is just a gadget for selective moves, it is what you are wanting if you are doing category re-orgs, it is what I used for the WS author occupations restructure.

Re TOC at enWS, Inductiveload has styles that sit with each Index: page and that allows for a lot simpler editing than those complicated TOC begin and subsidiary templates. It has allow me to strip a whole lot of cruft from my ToC and tables. Part of that is guidance to set a max-width in the css that allows for alternate displays and better exports. I couldn't get it working with classes in the "TOC begin" template though as you say the parameter gave a decent display.  — billinghurst sDrewth 03:01, 14 June 2021 (UTC)

@billinghurst: I made a really nice title page for Q19028421. I used a different scan of the same publication because there was a super small tiff at the source version and only single color, etc. It is the same version of this edition so all should be good there. I put all of the relevant information on the record there and I very much encourage you to use the data item at the scan (via the book template) to see the problems there!! Perhaps a concensus will be as great at writing the phab ticket as it is at voting! Who ever did the images for the Jacobs books (well, English Fairy Tales....) there has my respect. They just stroke my eyeballs with their goodness! --RaboKarbakian (talk) 16:51, 14 June 2021 (UTC)
@RaboKarbakian: Our work is on editions of each published work, not the reliance on the scans as provided by Google or IA. If we have to scavenge different scans of the same editions to build our djvu, I am all 👍👍👍, we did that for DNB in the old days. What I am uncertain (and undecided, and I wax and wane) about is whether we should be inserting colour images where a work was published in black and white. That is not a discussion for here.  — billinghurst sDrewth 00:03, 15 June 2021 (UTC)

Use of automated tools and correct instance of (P31)

Hi, please be careful when using scripts to edit Wikidata in bulk. I noticed some of your edits about A Naval Biographical Dictionary (Q16055052) are problematic: [1][2]

Accordingly, please undo or fix your overwrites. If you think changes to existing items about A Naval Biographical Dictionary (Q16055052) are needed, please propose these on Help talk:Import_NBD_from_enwikisource first. Thanks. --- Jura 12:19, 6 September 2021 (UTC)

Not using scripts all manually selected in WEF tool which just allows the input at enWS. I must have just miss selected with the drop-down list for Murphy, thanks for picking that up.  — billinghurst sDrewth 13:39, 6 September 2021 (UTC)
Also, none of Wikisource main namespace pages will be "Wikimedia list articles", we are reproducing works, so there is nothing Wikimedia-created in this space. Tables are tables.  — billinghurst sDrewth 13:45, 6 September 2021 (UTC)

Charles Aubyn Beach (Q108451212)

Hello! That might be a bit pedantic, but it is not a good practice to change the date of birth and keep the reference. The source DOES NOT state that the subject was born on 1 December 1827. Proper way to do it is to deprecate existing claim with the reference and add the new one. Thank you, Henry Merrivale (talk) 11:13, 8 September 2021 (UTC).

Retrieved from "https://www.wikidata.org/w/index.php?title=User_talk:Billinghurst/Archives/2021&oldid=1792609344"