User talk:Valeriummaximum

Welcome to Wikidata, Valeriummaximum!

Wikidata is a free knowledge base that you can edit! It can be read and edited by humans and machines alike and you can go to any item page now and add to this ever-growing database!

Need some help getting started? Here are some pages you can familiarize yourself with:

• Introduction – An introduction to the project.
• Wikidata tours – Interactive tutorials to show you how Wikidata works.
• Community portal – The portal for community members.
• User options – including the 'Babel' extension, to set your language preferences.
• Contents – The main help page for editing and using the site.
• Project chat – Discussions about the project.
• Tools – A collection of user-developed tools to allow for easier completion of some tasks.

Please remember to sign your messages on talk pages by typing four tildes (~~~~); this will automatically insert your username and the date.

If you have any questions, don't hesitate to ask on Project chat. If you want to try out editing, you can use the sandbox to try. Once again, welcome, and I hope you quickly feel comfortable here, and become an active editor for Wikidata.

Best regards!

Yale University Art Gallery CC0?

Hi Valeriummaximum, I noticed you added Commons compatible image available at URL (P4765) to a lot of files like for example Glass Vessel Balsamarium, Yale University Art Gallery, inv. 1938.5999.899 (Q100729690). Here you added copyright license (P275) -> Creative Commons CC0 License (Q6938433). I don't recall the Yale Gallery releasing anything as CC0 and I can't find any release on their website. Can you please point me to to release? To be clear, Creative Commons CC0 License (Q6938433) is a license and only the copyright holder can assign a license. Multichill (talk) 22:43, 6 December 2020 (UTC)

Hi, @Multichill:, I left a question on the discussion page for Commons compatible image available at URL (P4765) because I wasn't entirely sure what exactly was required for this. But if you visit the website page for the item [1] you will see a public domain declaration beneath the image. The image links I uploaded all have a public domain in the internal records for the gallery and were uploaded with the permission of the gallery metadata director. I was very confused by the discussion of CC0 and public domains and licenses in this property, so just let me know what I need to change. Valeriummaximum (talk) 22:56, 6 December 2020 (UTC)

Copyright, licensing and releases is always hard and confusing. These files are copyrighted and need an explicit release.

At the moment https://artgallery.yale.edu/using-images just mentions public domain, but not an explicit release. If you're in contact with the gallery director, maybe you can get it updated to explicitly mention that for any works believed to be in the public domain, the images are released in the public domain too? That's how other institution like the Metropolitan Museum of Art (https://www.metmuseum.org/about-the-met/policies-and-documents/open-access ), Cleveland museum of Art (https://www.clevelandart.org/open-access) and Smithsonian (https://www.si.edu/openaccess) are doing it too. Would be great if you can get that done and I will be more than happy to upload all these images to Wikimedia Commons. Multichill (talk) 23:11, 6 December 2020 (UTC)

You can find the gallery's open access statementshere. As I said, if you go to the item's page, it has a public domain declaration beneath the image. The only circumstances in which Yale University reserves copyright is when the image and object are owned outside the institution. The item you point out explicitly has the words public domain. Valeriummaximum (talk) 23:19, 6 December 2020 (UTC)

reproduction rights are addressed here [2]Valeriummaximum (talk) 23:26, 6 December 2020 (UTC)

Public domain is not the same as CC0 so you adding that statement to a lot of items is incorrect. Making incorrect copyright claims shouldn't be taken lightly (and it's the kind of thing that gets people blocked from editing on Wikimedia projects). Either https://artgallery.yale.edu/using-images needs to be updated to explicitly mention CC0 or these 13.000 statements needs to be removed. Of course I prefer the first option. If you need some assistance from local Wikimedia volunteers who have done these kind of releases before, please let me know so I can get you in touch with them. Multichill (talk) 23:32, 6 December 2020 (UTC)

My apologies if I give the impression of taking this lightly. I believed in good faith that I was making a correct statement and was acting in consultation with the institution itself. I'm happy to bring this up with the gallery (many people at the gallery, and you yourself, have uploaded Dura images from the Yale gallery). I would be happy to change the claims for these images to say that they are public domain and available for free use as Yale policy's lays out. Just let me know what I need to do. Valeriummaximum (talk) 23:38, 6 December 2020 (UTC)

Just to clarify, exactly how would you tag images as public domains because P275 does not permit this value, so I am just wondering exactly how Yale's open access policy and release from copyright is to be modelled.Valeriummaximum (talk) 00:03, 7 December 2020 (UTC)

Multichill you also asked whether images, as opposed to the works, arereleased for public domain but that is stated, 'No permission from the Gallery is required to use our images of public-domain works...in support of Yale’s open-access policy, we hope that students, scholars, artists, and anyone interested in the collection will freely use the Gallery’s resources for presentations and publications, as well as for personal enjoyment.'[3] so again I'm just wondering how to represent this correctly in WD and, as I said, I am more than happy to make any corrections you see fit.Valeriummaximum (talk) 00:15, 7 December 2020 (UTC)

Multichill just to continue this point, almost all of the images currently uploaded for Dura objects at Yale were uploaded by your bot [4], so I am very confused about what exactly is required for copyright license (P275) in terms of public domain licensing. Is it simply that the qualifier of CC0 has to be deleted? Valeriummaximum (talk) 03:27, 7 December 2020 (UTC)

Multichill you uploaded a Yale gallery public domain image Dead Christ (Q49216767) as a commons compatible image, so I am just wondering what exact public domain documentation you specifically require from Yale when you have uploaded hundreds of the gallery's images already. I would be happy to simply delete the CC0 qualifier--I misunderstood how to document public domain status of images in wikidata.Valeriummaximum (talk) 12:46, 7 December 2020 (UTC)

Copyright is quite complicated. A creative work is either copyrighted or in the public domain. We model this on Wikidata an Commons by setting copyright status (P6216) to copyrighted (Q50423863) or public domain (Q19652).

If a file is copyrighted, it needs a free license to be uploaded to Commons. On Commons we model this with copyright license (P275) set for example to Creative Commons Attribution-ShareAlike 4.0 International (Q18199165). A license can only be assigned by the copyright holder. A copyrighted files has to be released explicitly otherwise it would fail the precautionary principle on Commons.

One of the special cases in copyright is the faithful reproduction of 2D artworks. If the original artwork is in the public domain, this faithful reproduction is also in the public domain in the US. See When to use the PD-Art tag for background information. That's the reasoning I used for File:Quentin Massys - Dead Christ - 1871.114 - Yale University Art Gallery.jpg: Because Dead Christ (Q49216767) is in the public domain, the photo is too regardless what the museum claims.

The example I mentioned earlier is not 2D so although no copyright exists on the original work, the photo is copyrighted. The release on https://artgallery.yale.edu/about/rights-and-reproductions is not explicit and enough for Commons. That's why I suggested earlier that the best way to solve this is to make it an explicit CC0 release. Multichill (talk) 20:10, 8 December 2020 (UTC)

I'm not sure if a CC0 release will be an easy solution but will keep you updated. My understanding is that Yale gallery intends these images to be public domain (and I'm not sure why you say that they are copyrighted) so we will see if there is an easier way to express this on the website shortly and I will let you know. It seems to me though that there are hundreds of images (2D and 3D) that have already been uploaded into Wikidata by yourself and others, so this is an issue that affects many Wikidata items. Valeriummaximum (talk) 20:25, 8 December 2020 (UTC)

Also, I should also say thank you for taking the time to explain things for me Multichill. I am removing the CC0 claims from the items for now because this is clearly not correct documentation but will leave the image links for now. As I said, I will get back to you shortly about how the gallery can handle this. I wonder too if the documentation for this can be improved for users?Valeriummaximum (talk) 20:38, 8 December 2020 (UTC)

No need to alter the links now while you're still sorting it out, but it won't hurt. When (if) we get the copyright sorted out, I can have one of the robots do the actual upload to Commons so that all these items have a nice illustration. In my experience these kind of releases are hard and take quite a bit of effort, but it pays off in the long run.

One thing I try to prevent is images being uploaded without a proper release and getting them all deleted. That's not a very nice experience.

Good luck and thanks for putting the effort in! Multichill (talk) 20:47, 8 December 2020 (UTC)

Multichill we have been in conversation with the rights department of Yale University gallery. They are in the process of implementing a CC0 release of all images but they confirm that all images related to items excavated at Dura in the gallery are already released into the public domain under a CC0 license [5]. Please clarify if any further clarification is needed to import these images into Wikicommons. Valeriummaximum (talk) 19:42, 17 February 2021 (UTC)

That's very good news. Well done! https://ydea.yale.edu/about-ydea/rights-and-reproductions is excellent for uploading it to Wikimedia Commons. Multichill (talk) 17:38, 18 February 2021 (UTC)

Thanks Multichill, would it be best to add this reference as a qualifier for P4765?Valeriummaximum (talk) 18:21, 18 February 2021 (UTC)

Multichill Just to nudge, I re-added the CC0 statements but I just want to check if there's anything I can do to help upload these images into Wiki commons? I am a little unsure of how P4765 is meant to work as a placeholderValeriummaximum (talk) 20:37, 22 February 2021 (UTC)

I should have shared Wikidata:WikiProject sum of all paintings/Automated image uploads! Does that help? Multichill (talk) 21:32, 22 February 2021 (UTC)

Multichill Thank you! Would you be able to confirm that the image metadata has been appropriately modeled in WD? I think there are some elements like "creator" and "operator" that I was not sure how to model in an archaeological context. query is here.Valeriummaximum (talk) 21:45, 22 February 2021 (UTC)

Yeah, the standard bot probably won't work. I'll have a look if I can make a slightly modified one to handle these images. I recently also did that for the Cleveland Museum of Art. Give me a couple of days. Multichill (talk) 21:51, 22 February 2021 (UTC)

Multichill Thank you, let me know if there's anything I need to change as well. It takes a while to reformat all the items, so any help would be much appreciated! Valeriummaximum (talk) 21:53, 22 February 2021 (UTC)

Hi Multichill, I just want to check to see if you've had any progress on this or have any suggestions for what I can do. Many thanks Valeriummaximum (talk) 18:29, 11 March 2021 (UTC)

Thanks for the ping. I haven't forgotten about this, I justed wanted to get one of my other upload projects (Commons:User:GeographBot) on the road again before I picked up something else. Multichill (talk) 17:12, 15 March 2021 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ Ok, let's get ready to do some uploading! I think https://w.wiki/36rX returns all the files you want uploaded, right? I'll make a new source template on Commons to link to https://ydea.yale.edu/about-ydea/rights-and-reproductions . Makes it also easier to check usage and image views. I guess all the artifacts should be categorized under Commons:Category:Art from Dura Europos? Multichill (talk) 20:12, 17 March 2021 (UTC)

Multichill Thank you!! That sounds perfect to me! Thank you so much again for this!Valeriummaximum (talk) 23:27, 17 March 2021 (UTC)

I uploaded all the files. To track all the files I created Template:Yale Digital Dura-Europos Archive which puts the files in the hidden tracker category Category:Images from Yale Digital Dura-Europos Archive. For regular by topic categorization I put the files in Category:Objects from Dura Europos in the Yale University Art Gallery. Someone might start sorting these out in subcategories. Multichill (talk) 10:55, 8 May 2021 (UTC)

WONDERFUL! Thank you so much! I should add that all Yale Gallery images have been now been released under a CC0 license with a IIIF manifest. See here for example. So it should be much easier for users to ingest gallery images into the Wikidata now. Thank you so much again!! Valeriummaximum (talk) 14:22, 8 May 2021 (UTC)

Inception is generally not unknown

Hi Valeriummaximum, I was looking at cases where inception (P571) is set to unknown value Help, see https://w.wiki/qCz . This shouldn't be used a lot. If the time period is known, the date should set to the average of the earliest and latest date with a low enough precision to cover the both of them. Say for example the source says a work was made between 1882 and 1886, we set the inception to 1884 with precision decade. When the precision is set to decade only the 188x part is significant so the last digit doesn't really matter. We just set it to the average so when sorting by date, it makes more sense. Because most (all?) of the items you edited cross the year 0, the output of the interface is a bit awkward. I updated Fresco, Yale University Art Gallery, inv. 1938.5999.2269 (Q49313813) as an example: The average is -33 and I set the precision to a millennium. Maybe you can update the other Yale frescoes? Multichill (talk) 16:24, 11 December 2020 (UTC)

For a lot of reasons, that's just not a good approach for these objects. If you think this should be the standard, wikidata isn't a good place for us to host GLAM data Valeriummaximum (talk) 17:16, 11 December 2020 (UTC)

I would add that we sought a lot of community recommendation from many many wikimedians about representing period ranges -- some suggested changing the date precision but averages was never suggested. Can you point to a discussion page where this standard of not using unknown value too often is presented? There's good reasons for using unknown value (it's recommended for this specific use, it represents the curators' intentions, it allows us to distinguish absolute dates from period judgements, and it's the closest thing that reflects archival practices). Valeriummaximum (talk) 17:16, 11 December 2020 (UTC)

You missed the point. If the precision is set to decade, 1882, 1884, 1887, etc. are all the same value in the Wikidata data model, if the precision is set to century than 1920 and 1990 are also the same. As a convention for paintings we usually set it to average so the imprecise dates don't clutter at a start of a decade on lists like Wikidata:WikiProject sum of all paintings/Creator/Vincent van Gogh.

So set a date and set the precision to something big. Help:Dates has most info on it, but not sure if it's complete. Multichill (talk) 22:28, 11 December 2020 (UTC)

I rechecked the definition of the data type. I guess crossing the year 0 messes up precision because whatever precision you set, it always starts (or ends) at year 0 so I guess that's the only case that you're better off using "unknown value". So nevermind.

Many people did initially suggest changing the precision date and we experimented with that, but the issue of crossing centuries and crossing between BCE and CE (and also how this affects SPARQL queries) caused many many issues. It's clear that Wikidata needs to rethink a lot of its modelling for time (because using unknown value with date qualifiers also very much restricts how people can access that information as a SPARQL query). Ultimately 'unknown value' was recommended to us. I genuinely look forward to replacing this placeholder in the future but for a lot of reasons I think it is much safer to retain and I am very reluctant to average date ranges because, while it is sometimes safe to fill missing data, it is useful in this case to keep track of these particular objects for the reasons given above (to track absolute dates vs. ranges). Also millennium is a bit too imprecise when talking about the history of art in Dura.Valeriummaximum (talk) 23:00, 11 December 2020 (UTC)

Just out of curiosity, the years 323 BCE and 256 seem very specific, what are these based on? Multichill (talk) 22:37, 11 December 2020 (UTC)

These are based on known historical dates for the site of Dura Europos. Generally, if a painting (or any object) shows Greek influence (in technique, religious cult, etc) or was exacavated from a site datable to Greek settlement, this would suggest the Greek period of occupation (i.e. 323-113BCE), but many objects can only be roughly dated and of course dating by Parthian, Greek and Roman style can be an unreliable method. The qualifier 'circa' is inconsistently given in the catalogues so these are not always perfect dates. Valeriummaximum (talk) 23:00, 11 December 2020 (UTC)

Coming back to this oldie. If you don't cross year 0 like on 1 Nummus of Constantine I, Emperor of Rome, from Trier, Yale University Art Gallery, inv. 1938.6000.3017 (Q100585727), it's better to do it like this. Logic is generally: (earliest + latest)/2 with manual precision set to include both earliest and latest. Qualifiers stay the same. These could be updated. Multichill (talk) 11:12, 8 May 2021 (UTC)

User talk:Multichill Personally, I'm still not sure I feel comfortable with this edit. When curators give a very specific, definite date range, I'm not sure what value we add by including this kind of imprecision.Valeriummaximum (talk) 14:27, 8 May 2021 (UTC)

That's how things are modeled on Wikidata. The source states "323–24" so it's quite weird to state that the date is unknown, that's even more imprecise. We modeled exactly what the curator put in. Multichill (talk) 14:39, 8 May 2021 (UTC)

My concern is that if someone queried for an object created anytime before 322 or after 325, this coin would come up. Can you point me to where these decisions are publicly documented so I can better understand the motivation? Valeriummaximum (talk) 14:44, 8 May 2021 (UTC)

You added earliest date (P1319) and latest date (P1326) so that someone would be doing an incorrect query if it shows up while looking for anytime before 322 or after 325.

See Help:Dates. Your understanding of the data model might be slightly different to what it actually is. Multichill (talk) 19:44, 8 May 2021 (UTC)

User talk:Multichill If you know in advance to refine your query to the qualifier level searching along earliest date (P1319) and latest date (P1326), then yes, there is no problem; but if you simply query inception (P571), looking for objects created after 325, then this coin will still appear because you have specified a precision level of 320s. I am concerned about this latter case and I don't understand the need to create imprecise dates. To repeat, can you point me to public documentation where these standards have been debated and approved? Valeriummaximum (talk) 20:02, 8 May 2021 (UTC)

Yes, if you do imprecise queries, you get imprecise results. And no, we're not creating imprecise dates. If I want to get all the coins from a century these won't show up. We have a loss of data here.

It's an entirely reasonable query. The question for me still is, what is the need for imprecise averages, a standard that does not reflect any exiting curatorial standard.Valeriummaximum (talk) 22:31, 8 May 2021 (UTC)

I already pointed out Help:Dates to you. What else are you looking for?

Actually, that does not help: it doesn't explain when it is suitable to average out dates (contrary to your suggestion, it is not just a problem when the date range passes over year 0, but also when it passes any decimal e.g. It would be misleading to say that a coin minted 16-26 was minted in the 20s). It also does not say that date ranges must be averaged anywhere. Nor does it spell out where these conventions came from or that they must be applied by every user.Valeriummaximum (talk) 22:31, 8 May 2021 (UTC)

By the way, why do you keep liking to "User talk:Multichill"? What are you trying to achieve with that? Multichill (talk) 21:57, 8 May 2021 (UTC)

Have your published your pywikibot code?

Hi, I saw https://www.wikidata.org/wiki/Help_talk:QuickStatements#Property_with_multiple_values and wonder if you published the code somewhere?--So9q (talk) 11:20, 17 January 2021 (UTC)

• HI So9q, the documentation for Pywikibot is super useful but it's always good to be able to see what others have done. Here is a link to some of the initial upload. Basically, I just did a sparql query, retrieved the results as a json file, and then iterated over a data frame with all the values I wanted to upload. If the item in our data frame was in Wikidata, we updated it with new values; if it wasn't, the line item = pywikibot.ItemPage(repo) creates a new page. You can check the Pywikibot documentation on Wikidata for a lot of help. I hope what I wrote is clear generally (it has a lot of redundancy). Valeriummaximum (talk) 13:00, 17 January 2021 (UTC)

Connecting coins with the emperor

I noticed quite a few coins in the upload like for example 1 Nummus of Constantine I, Emperor of Rome, from Trier, Yale University Art Gallery, inv. 1938.6000.3017 (Q100585727). Wouldn't it be nice to connect these to the emperor? Maybe also the other side if known? Multichill (talk) 11:02, 8 May 2021 (UTC)

User talk:Multichill Yes, I've brought this up in several different forums. See here for one of the discussions. The issue is that the emperor is not always depicted on the coin (coins can sometimes be identified with the emperor alternatively by legend, iconography or findspot) so we've been thinking about alternative properties. We thought alternatively we could model the relation by "authority" or "regnal years". I'd really value your suggestion about how this can be modeled appropriately.Valeriummaximum (talk) 14:19, 8 May 2021 (UTC)

Maybe break out the easy cases first where we know who or what is on the coin and use depicts (P180) for these? For more generic links significant person (P3342) could probably be used with a qualifier. Multichill (talk) 14:43, 8 May 2021 (UTC)

I'm afraid that that's a very tricky business (I can't even be sure that the coin you just edited depicts Constantine I) and coins will often conflate the portrait of gods and emperors and past emperors, so determining who or what is a coin can always be controversial. It would be best if there were some generic property for the authority overseeing currencies. Valeriummaximum (talk) 14:48, 8 May 2021 (UTC)