User talk:Clements.UWLib

From Wikidata
Jump to navigation Jump to search

Inception vs publication dates for theses[edit]

Hi, Sorry for the lack of edit summaries on the batch edit I submitted! I was finding theses that had an identical publication date (P577) and inception (P571), and consolidated to just publication date (P577) (query that yielded the list: https://w.wiki/f25). I've assumed that if the inception and publication are different that they're indicating something other than publication date (e.g. inception might indicate start of phd, or start of thesis writing period, or thesis submission date). T.Shafee(evo&evo) (talk) 02:14, 8 October 2020 (UTC)[reply]

Ah, that makes some sense. So, if the publication date and date of inception happen to be the same, only the publication date should be included? Crystal Clements, University of Washington Libraries (talk)
Yes, that seems to be the way it's been handled by the majority of thesis items (though there's still a bit of a mix). About 16,000 theses state only publication date and 3,000 state only inception date with only a few stating both. T.Shafee(evo&evo) (talk) 04:57, 9 October 2020 (UTC)[reply]
If the inception date is known, I don't see why it should not be included even if it is the same as the date of publication. If someone were to search by inception date, they would not find the thesis unless they searched by the publication date...they would need to know the date was the same. Should this practice be an enforced rule for dates in theses? Is it already? Thanks for taking the time to answer my questions--I'm relatively new to Wikidata and used to library cataloging, which is quite a bit different. Crystal Clements, University of Washington Libraries (talk)
No problem, I'm still relatively new myself to many of the details. So the reason for using publication date (P577) rather than inception (P571) is that very searchability question - i.e. publication date (P577) is the more frequently used parameter for theses (as well books and journal articles) so is the more likely search (indeed it was the search that I'd been running until I manually noticed items missing that I'd expected to pick up). Also, for many of the thesis repositories, the actual stated date is a publication date (example with inception date in wikidata, but stated as publication date in the repo). T.Shafee(evo&evo) (talk) 22:43, 12 October 2020 (UTC)[reply]
It makes sense, if you only want to use one property per date, to use publication date. But I don't see why that means one should not include both if they are available and determined to be useful. They are two separate statements and each adds information to Wikidata. Crystal Clements, University of Washington Libraries (talk)
I see what you're saying about having both present in cases where they're the same. I did a quick check for whether the same scenario happens for books and found 1800 book items and 45 journal articles that use inception date too (https://w.wiki/gdY and https://w.wiki/gda). Sadly, that's probably the extend of my knowledge on the topic, but it could be worth asking at Wikidata_talk:WikiProject_Source_MetaData or Wikidata_talk:WikiProject_Books. T.Shafee(evo&evo) (talk) 03:52, 13 October 2020 (UTC)[reply]
Having through a little more, my hesitance might be that I find inception (P571) ambiguous for these items. Some of the book and journal article items use inception to mean date published, others mean date available online, or date when draft was submitted, but it's not possible to tell without going over to the item's catalog URL. Using a more specific property might be clearer for cases where inception doesn't mean publication (example). T.Shafee(evo&evo) (talk) 04:00, 13 October 2020 (UTC)[reply]
Hmm. I don't think that's a good enough reason to delete unique, correct statements about the inception date of a thesis. Is there an application profile somewhere with rules for describing theses in Wikidata? I don't ask to be adversarial--I ask because we are hoping to eventually upload metadata about our institutional theses and dissertations to Wikidata, so we need to understand how the metadata should look. What is usually done when there's disagreement? It seems like such a tiny thing, but I'm new to the Wikidata community. It seems that you think inception dates should be deleted in favor of publication dates when those dates are the same. I disagree. Who decides? How does it work? Thanks again Crystal Clements, University of Washington Libraries (talk)
The most relevant guideline is probably:
There's also a couple of other relevant pages at:
The best way to get feedback is probably to summarise the options and ask for third opinions at either the Wikiprojects (Source MetaData / Books), or in the general project chat. People there may also think of angles or situations that neither of us have. That'll also help in updating the existing guidelines. And no worries, I didn't interpret as adversarial (though always good to confirm). It's useful to hash these things out to avoid us (or others in future) accidentally unravelling other people's data! T.Shafee(evo&evo) (talk) 05:37, 14 October 2020 (UTC)[reply]
Thanks for all the feedback! We will definitely check out those resources when we're getting ready to put more theses/dissertations into Wikidata. Crystal Clements, University of Washington Libraries (talk)
  • According to WikiProject Books, inception (P571) should be used on work items. The item for a thesis is a work item. We don't create separate expression/edition items for individual editions of a thesis, which, for electronic ones, would not be appropriate anyway, as there is only one manifestation and expression of the work that would be freely accessible. Online theses are treated as published books in our cataloging rules, so we should be following the guidelines of WikiProject Books. UWashPrincipalCataloger (talk) 06:41, 8 February 2021 (UTC)[reply]
Thanks UWashPrincipalCataloger. So, for the sake of electronic theses, we should be including (generally) the properties for works on one Wikidata item, and exclude information only relevant at the expression/manifestation level such as publication date/publisher?--Crystal Clements, University of Washington Libraries (talk) 16:09, 9 February 2021 (UTC)[reply]
This is an interesting question, because with theses there generally is just one expression of the work, and one manifestation of the expression. However Proquest does actually publish our ETDs in addition to we putting them into ResearchWorks. Work items should only have the work elements specified by the WikiProject Books model, so no publication date, no place of publication, no publisher. Since we can still include the handle of the ETD and the full work available at URL property, I don't really see a need for a second expression/edition item. UWashPrincipalCataloger (talk) 16:45, 9 February 2021 (UTC)[reply]

Award[edit]

The Wikidata Barnstar
Congratulation for the good work on Wikidata and the contributions!

Listeria issue[edit]

I've fixed Wikidata:WikiProject PCC Wikidata Pilot/University of Washington/Items Created or Edited: Organizations Only ... its SPARQL was the main cause of the problem. --Tagishsimon (talk) 02:24, 8 February 2021 (UTC)[reply]

Thank you Tagishsimon! I have no experience using SPARQL hints and appreciate your expertise. --Crystal Clements, University of Washington Libraries (talk) 02:48, 8 February 2021 (UTC)[reply]
My pleasure. Bring any future SPARQL or Listeria issues of any sort to Wikidata:Request a query, where we'll make short work of them. --Tagishsimon (talk) 04:28, 8 February 2021 (UTC)[reply]

Heuristic genders[edit]

Instead of reversing heuristic genders, they have a very high rate of correctness, why not do the research to prove them right/wrong? If a given name is used by both genders, the formula does not assign a gender to that given name. You can also look for a primary/secondary source that mentions gender. RAN (talk) 17:46, 24 March 2023 (UTC)[reply]

Joining this request. I will revert https://www.wikidata.org/w/index.php?title=Q109493917&curid=104639174&diff=1859901829&oldid=1856022069 . The rules for naming people in Russia and USSR were always strict and it never changed. Citing https://ren.tv/news/v-rossii/727697-nazvany-prichiny-po-kotorym-mogut-otkazat-registrirovat-imia-v-zags :
In the practice of civil registry offices there are refusals to register births and state registration of name changes in cases where such names contradict the traditions of Russian onomastics. In particular, names that are abbreviations, do not correspond to the sex of the child, or contain numbers, numerals or signs that are not letters will not be registered. Also, the use of swear words and indications of titles, ranks or positions will not be allowed.
Lockal (talk) 18:39, 24 March 2023 (UTC)[reply]
 Support Matlin (talk) 19:06, 24 March 2023 (UTC)[reply]
 Support --Tomukas (talk) 17:12, 25 March 2023 (UTC)[reply]
This may be the case in Russia, but not in other countries. The amount of research necessary to prove heuristic genders right/wrong is enormous, and the reason behind making blanket gender statements based on labels in the first place (heuristic genders are much much easier to assign flippantly than to actually research). If we are going to do research to prove individual statements wrong, we may as well just create the P21 statements by hand as that research is being done. It is not feasible to do all this research, and if it is undertaken will take a long time. It should not be on the rest of the community to clean up mistakes made on large scales by quickstatements batches like this. It should be the responsibility of those doing the edits. The "they have a very high rate of correctness" sentiment actively reinforces a false, harmful binary view of gender which is not an accurate reflection of gender in the world. Assigning gender to people across the world on Wikidata based on their labels is unethical regardless of the rules for naming people in Russia (one country). Crystal Yragui, University of Washington Libraries (talk) 22:52, 25 March 2023 (UTC)[reply]
Looks like we've got conversations in a couple of places about the same topic. For the sake of keeping things in one place, can we move over to the Wikidata project chat?. I think more people will have the opportunity to participate due to that page's visibility. Crystal Yragui, University of Washington Libraries (talk) 23:08, 25 March 2023 (UTC)[reply]
Retrieved from "https://www.wikidata.org/w/index.php?title=User_talk:Clements.UWLib&oldid=1861249160"