Hallo Matthias,
ich denke, dass es sich bei Neue Deutsche Biographie (Q67224894) nicht um die NDB, sondern nur einen Lexikoneintrag handelt.
Matthias Winkelmann
Joined 26 November 2015
Hallo Matthias, du hattest bei Q65219791 die Beschreibung von Straßenbahn auf S-Bahn geändert, aber das "ist ein" auf Straßenbahnhaltestelle belassen. So passt das nicht, ich habe die Beschreibung zurück geändert, denke auch dass das etwas anderes ist als Q32661483. Okay?
Hello, you did some importing from YVNG database of holocaust a year and a half ago but you seem to have been adding wrong days of death in some cases.See for example Kalman Drach (Q95110319) where the source says "1944" but you added "1944-01-01". This problem seems to appear in thousands of items you have edited... see https://w.wiki/4TyQ . Can you do something about it?
You are apparently not active here anymore so I did it for you.
I had to revert a bunch of your changes to St Mary's Star of the Sea Catholic School (Q7594538) because they were for the wrong school. It seems that the wrong MySchool ID was added some years ago. I've fixed it now, but you might want to check things.
Hello, you add years of birth for several biographies (Q3131628, Q523844, Q1450815) when there is already a date of birth with a reference, why double it? Bien cdlt
P.S. : FYI, with some records on my watchlist, I moved the Open Library identifier to the specific date.
See the other topic... I was adding Open Library as a reference, but QS decided to add the date a second time, then fail at adding the actual reference.
I found the solution by searching. When we add a year with an identifier and there is already a specific date, it is necessary to put "privileged rank" to this last, so that the infoboxes work.
If you add a second birth or death date that is less precise than the previously present date, please mark it as "deprecated" (since it is less precise. Duplicate dates of birth or death get marked as problematic and cause issues for Wikipedia and other sites pulling data from the entry. For example, at Simon Dach (Q77285), there were birth and death dates giving the day, month, and year, but you added a second data entry for just the year. I have marked your additions as deprecated, and you should do this if you are duplicating dates with less precision.
For entries where there is already a date exactly matching what you are adding, you should simply add the reference, not duplicate the date. You did this at Richard Josiah Hinton (Q61643575) and this creates a problem of duplicate values. Simply add the reference rather than duplicating the data value.
I would also point out that using OpenLibrary as a reference adds nothing. Most of the dates of birth and death at OpenLibrary were added anonymously without any source for the information, so it is not a valuable or reliable source of information.
Hi. For one person I saw that it added exact replicate dates of life, and without references. Can you please double check that your run is not replicating existing data. Also, references are preferred when adding dates. Thanks.
It's supposed to add the year with a reference to Open Library. QS is supposed to/usually does catch the case where the statement already exists and just adds the reference. For some reason, it fails to do so here. I'll go over it and delete the superfluous statements.
Okay... The reason is that, even with precision=9 (i. e. "Year"), the dates 1921-01-01:00:00:00[etc.] and 1921-00-00:00:00:00[...] are considered to be different.
For my own archive and those finding this 10 years from now, here's a query that finds these items:
SELECT ?item ?itemLabel ?dob ?precision ?dob2 ?precision2 (YEAR(?dob) as ?y) (YEAR(?dob2) as ?y2) ?timenode ?timenode2 ?timestate2 ?olid2 WHERE {
?item wdt:P648 ?ol.
?item p:P570/psv:P570 ?timenode.
?timenode wikibase:timeValue ?dob .
?timenode wikibase:timePrecision ?precision .
FILTER(?precision = 9).
?item p:P570 ?timestate2 .
?timestate2 prov:wasDerivedFrom [ pr:P648 ?olid2 ] .
?timestate2 psv:P570 ?timenode2.
?timenode2 wikibase:timeValue ?dob2 .
FILTER(?timenode != ?timenode2).
?timenode2 wikibase:timePrecision ?precision2 .
FILTER(?precision2 = 9).
FILTER(YEAR(?dob2) = YEAR(?dob)).
SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
} ORDER BY ?dob LIMIT 500
I'm running the cleanup statements now but those are, once again, just giving me random errors half the time. So.. maybe later, I guess.
Please have a look at Talk:Q2763.
electric power transmission (Q200928) publication interval 2 month doesn't make sense. Can you check the whole batch?
Hi, I just restored the item for Oskar Schindler (Q27518152). It is ok (and actually also ideal) to create own items for fictionalized versions of historical persons in order to not muddle information about the fictional character (e.g. narrative role (protagonist, antagonist), but also information about the relationship to other characters) with information about the real person. The presentation of characters in fictional works can differ substantially from what may be considered historically accurate. I added fictional analog of (P1074) to make clear that Oskar Schindler (Q27518152) is actually meant to represent Oskar Schindler (Q60029). Best wishes!
Strongly disagree...
There are dozens of movies and books that portrait Abraham Lincoln. Except for one opera, where the character is remarkably not the actual Abraham Lincoln, we never considered any discrepancies to the real person worthy of making that distinction and creating items for the character.
That includes Abraham Lincoln: Vampire Hunter (Q587707).
The 23 listed roles on Apollo 13 (Q106428) all link to the actual person they are supposedly representing. And so on... If you check the list of fictional characters based on real people, it is dominated by Shakespeare and Opera, two genres where the divergence of, say, the historical Macbeth and the one from the eponymous play is rather obvious.
So what does it say if we stray from common practice in the case of Oskar Schindler (Q27518152) and Schindler's List (Q483941)? Quite obviously, it raises doubts about the historical accuracy of the movie. It did this even in my mind, as I expected the enwiki article to include some sort of controversy that prompted this. To my surprise, there really aren't any accusations of factual inaccuracies made against the movie! (Apollo 13, btw, has a lengthy section discussing accuracy).
Comparing the two items, I noticed that there actually are no differences in any of the statements. That shows both that the depiction is accurate, and that there is no need for the split. Where the character is left-handed, the real person was as well!
With these two remaining good-faith interpretation of this modelling choice falling away, it can only be ascribed to the ever-present background noise of arbitrariness. And another interpretation, which I am sure anyone reading this far has already thought of: namely, that it falls somewhere between "alludes to" and "is instance of" Holocaust denial.
I'm merging them again. Feel free to undo after you've created items for each cultural representation of Abraham Lincoln, Hitler, Jesus, Jim Lovell, and New York.
The decision to differentiate between fictional characters and humans they are based on has nothing to do with denying that the events they were involved in happened.
It is to prevent the intermixture of fact and fiction generally. As @Jheald: expressed it at Wikidata:Project chat/Archive/2020/01 (Splitting and merging fictional characters): "The point about having separate items for fictional counterparts of real people, distinct from the items for the actual real people, and only letting fictional characters interact with other fictional characters, is important. Otherwise we find statements like this getting batch-added to the items for real people."
There are always differences between the characterization of a person in fiction and the historical person. The historical accuracy of the depiction of Oskar Schindler in Schindler's List is contested, see https://www.theguardian.com/world/2004/nov/25/germany.film. Yes, there are no differences in statements, yet (apart from the instance of (P31) and narrative role (P5800)-statements you deleted), but there may be if somebody sets out to model the narrative of Schindler's List (Q483941). I will leave the item now as it is as the narrative of Schindler's List (Q483941) is not modelled at all so there is no loss of information that could not be easily restored if somebody sets out to model the narrative.
"Feel free to undo after you've created items for each cultural representation of Abraham Lincoln, Hitler, Jesus, Jim Lovell, and New York."
If this were English Wikipedia, I think a link to WP:ALLORNOTHING would be appropriate here. Since we are instead in Wikidata, I think it suffices to say that each aspect of editing has to be considered separately. For example, as you noted, Wikiproject Performing Arts has been majorly behind the effort to have separate items, in part because there was an import workflow in effect. It worked for them; whether it works in a given other instance relies a lot on context, I think.
Hello, I see you removed P31 instance of Holocaust victim.
I don't understand why you did this.
Could you explain please?
Yes, it was in line ihr the discussion at Talk:Q2763#Modelling of holocaust victim. We just don’t use instance of (P31) that way and I’ve also had feedback (offline) that it’s possibly offensive to „allow the perpetrators to define“ that these people’s‘ identity is that of a „victim“.
I’m not entirely sure there is consensus of how to model it yet. For now, it’s relatively straightforward to getan list by searching for the respective Yad Vachem or Memorial Book statements.
I've got a problem. Not all the people known to be Holocaust victims have one unique identifier in Yad Vashem so relying on the Yad Vashem idenfier misses them in Wikidata queries. Talk:Q2763#wikidata queries for Holocaust victims and survivors
Role has got the same problem philosophically as P31 - role: Holocaust victim vs instance of: Holocaust victim
Plus I'm not sure which P code to use for role as there are at least three possibilities. And no clear consensus emerged from the discussion.
As long as there is not a clear consensus on an approach that works for all the Holocaust victims and allows them to be found in queries, could everyone let the P31 Holocaust victim remain? That way, in queries, we can retrieve all entities with either a Yadvashem code or a P31 Holocaust victim.
If you remove P31 for Holocaust victims who have no clear Yad Vashem code, it accidentally disappears them.