About this board

Previous discussion was archived at User talk:Marcus Cyron/Archive 1 on 2017-02-26.

https://www.wikidata.org/w/index.php?title=Q247365&oldid=prev&diff=1352910464

5
Kam Solusar (talkcontribs)

Hi Marcus, die "Quelle" hab ich gelöscht, weil es IMO keine wirklich verlässliche Quelle ist. Ist einfach eine von recht vielen ähnlichen Seiten, die ihre Infos blind von anderen Seiten übernehmen. In diesem Fall woohl von TMDb, die selbst halbwegs okay ist (auch wenn sie IMO öfters mal gerade bei Veröffenlichtungsdaten von anderen Quellen abweicht). Die Seite beschreibt sich auch selbst als "image discovery platform". Wenn man in diesem Fall die Daten von TMDb als verlässlich ansieht, sollte man IMO dann lieber direkt https://www.themoviedb.org/movie/73134-tendres-cousines nehmen.

Wir haben leider im Bereich Filme ein recht großes Problem mit zweifelhaften und teilweise schlicht unbrauchbaren Quellenangaben. Da ist im Laufe der Jahre leider extrem viel relativ achtlos in Wikidata hineingekippt worden, vor allem mittels unausgereifter Tools. Darunter alle möglichen Torrent- und andere Filesharing-Sites, Foreneinträge, Spam- und SEO-Seiten, 08/15-Webshops, Wikipedia-Klone und sehr viel Murks der wohl in den Google-Suchergebnissen aufgetaucht istk, auf denen die fraglichen Suchbegriffe aber nie in einem zusammenhängenden Kontext zu finden waren. Da bin ich im Moment dran, das Gröbste auszumisten und diiese Seite war dabei mit auf meine Liste geraten.


Gruß, Kam

Marcus Cyron (talkcontribs)

Hallo, wenn diese Quelle fragwürdig ist, dann ist doch aber die ganze Aussage fragwürdig und sollte dann komplett gelöscht werden. So ist das doch weiter da und gammelt als fragwürdige Aussage hier rum. Wenn man davon ausgeht, dass das nicht sauber belegt werden kann, sollte man es dann doch komplett löschen?!

Kam Solusar (talkcontribs)

Tja, eigentlich schon. Ist aber leider eher ein generelles Problem des Projekts, dass ein recht großer Teil der Angaben im Filmbereich immer noch nicht oder nur rudimentar (importiert aus Wikipedia) belegt ist, da machen diese paar Angaben auch keinen großen Unterschied. :-/ Und solcherlei Belege scheinen oft erst nachträglich ergänzt worden zu sein, d.h. die Angaben stammen meistens gar nicht aus diesen Quellen selbst und wurden von andren Usern eingefügt.

Da bin ich recht zögerlich einfach komplette Aussagen zu entfernen, vor allem weil man keine Begründung/Zusammenfassung für Edits angeben kann und ich es ehrlich gesagt vermeiden möchte, dutzende Diskussionen mit unterschiedlichen Usern über die Entfernungen solcher Angaben zu führen. Wäre natürlich schön, wenn ich die Zeit und Energie hätte, für alle diese Filme die Angaben jeweils einzeln zu überprüfen, wenn nötig zu korrigieren und mit Belegen zu versehen. Ist aber leider bei der riesigen Zahl von Items und niedrigen Zahl von aktiven Usern nicht annähernd machbar. Da versuche ich in letzter Zeit lieber, zumindest erstmal größere/übergreifende Probleme wie problematische Quellenangaben, fehlerhafte Modellierung der Daten mit falschen Properties/Eigenschaften, fehlende externe IDs etc. zu korrigieren - in der Hoffnung, dass wir in Zukunft bessere Möglichkeiten haben werden, unsere Daten mit externen Quellen abzugleichen um Fehler/Unterschiede aufzuspüren.

Marcus Cyron (talkcontribs)

Ich bin mir wirklich sehr unsicher, ob es sinnvoll ist, nicht dann doch die Daten komplett zu löschen, statt nur der Belege. Wenn wir solche Hoffnungen haben, sollte das dann auch wirklich komplett so abgeglichen werden können.

Kam Solusar (talkcontribs)

Eigentlich schon. Werd in Zukunft mehr auf die WP-Artikel und externen IDs schauen, ob die Angaben in Ordnung scheinen und sie ansonsten mitsamt den unbrauchbaren Quellenangaben löschen.

Reply to "https://www.wikidata.org/w/index.php?title=Q247365&oldid=prev&diff=1352910464"
MisterSynergy (talkcontribs)
Marcus Cyron (talkcontribs)

Für mich als aktuelles Beispiel ist es der Amerikanisch-Samoanische Handballverband, der eben nicht nur den klkassischen Hallenhandball, sondern ganz besonders und ausdrücklich auch Beachhandball vertritt. Kannst du aber auch noch weiter oben ansetzen. Fußball schließt mittlerweile diverse Disziplinen ein, wie Futsal, Beachsoccer, Hallenfußball, Blindenfußball etc. Somit ist die Angabe der Disziplinen die auch wirklich vertreten werden durchaus von Bedeutung. Denn nicht jeder Verband beitet alles. Der DFB etwa nur den klassischen Fußball und Futsal. Viele Handballverbände nicht Beachhandball. U.s.w.

MisterSynergy (talkcontribs)

Okay, verstehe.

Ich denke aber, dass wir anstelle von sports discipline competed in (P2416) hier einfach sport (P641) verwenden sollten. Die Trennung zwischen "Sportart" und "Sportdisziplin" ist ja eh nicht in jedem Falle sauber hinzubekommen und in der Fachwelt gibt es verschiedene Definitionsansätze dazu. Die allgemeinere Eigenschaft sport (P641) ist für solche Fälle eigentlich besser geeignet. Der Sportverband selbst nimmt ja auch nicht in einer Sportdisziplin an Wettbewerben teil, was der Zweck der Eigenschaft sports discipline competed in (P2416) ist.

Marcus Cyron (talkcontribs)

Kann ich au mit leben.

MisterSynergy (talkcontribs)

Alles klar, ich habe das entsprechend umgestellt und die Eigenschaftseinschränkung auch wieder entfernt. Viele Grüße!

Reply to "Special:Diff/1266579899"
Dorieo (talkcontribs)

Ok. Thank you for the warning. Greetings

Marcus Cyron (talkcontribs)

Not a warning! Just a hint ;). I made this "mistake", what's not really a mistake, also for years in the beginning. Were you ever in the Archaeological Museum of Athens?

Dorieo (talkcontribs)

I don't know English, so I've written it using a translator.  As to whether I've been to Athens, the answer is yes, in 2007. I spent two days photographing works at the National Archaeological Museum. It's a pity that many pictures were taken by my son, then a child, and they were blurred. I also took photos at the Acropolis and the Agora Museum and at the Agora itself.

Clearly you have been to Athens. There are hundreds of pictures of you in Commons, by the way, all very good.

Best regards

Reply to "Re: Gender"
Jarekt (talkcontribs)

Marcus Oinochoe by the Shuvalov Painter (Q834119) is in Public Domain. In EU once an author is dead for 70 years the work falls into PD. The object is usually own by someone and that someone might put all kinds of restrictions on access to the objects or on photographing it, but they are not using copyright laws to justify it. Photographs of such objects only carry photographers copyright. If you are not convinced you can do a simple test: If that object is still copyrighted than the image of it can not be hosted on Commons. File Deletion request for the image and you will get second opinion.

Marcus Cyron (talkcontribs)

Jarekt - we talking abou the object. Not about the images of the object. And the object is owned by an Institution. The object is not Public Domain. We're not talking about Copy right here. We talking about the object itself.

Jarekt (talkcontribs)

Marcus, Let me quote 1st sentence of [[en:Public Domain]: "The public domain consists of all the creative work to which no exclusive intellectual property rights apply". The statement you have removed only stated that this work in not covered by any "exclusive intellectual property rights" (copyrights). Most valuable objects are own by some institution, I am sure Mona Lisa is or US constitution, but both are also in Public Domain. PD just means "no copyrights" and is unrelated to ownership.

Reply to "Oinochoe by the Shuvalov Painter"
2001:7D0:81F7:B580:DE6:2213:E282:2C48 (talkcontribs)

Hi! What exactly is there nonsensical about Special:Diff/1065442417? This item to my understanding is about type of marble (i.e. a class), so the "instance of" relation doesn't really apply. Now, after merging, this item is more accurately set as a subclass of (subclass of) marble. See also Help:Basic membership properties.

Marcus Cyron (talkcontribs)

"Hymettian Marble" is a marble. Marble is a stone class. Point. And it's a subclass of "greek marble". Point. I don't see you problem.

2001:7D0:81F7:B580:1166:1356:EBF9:ECC2 (talkcontribs)
Reply to "Hymettian marble"
Mike Peel (talkcontribs)
Marcus Cyron (talkcontribs)

Hallo Mike. This is a very, VERY complecated field. You have to imagine some 1000 gods and the researchers today still find new ones. Lists for example with the names of deities are some of the oldest writings in history of mankind. But this is not static. This changes. Nintinugga is an eary goddess, that later is part of a transformation of sumerian gods. It's a process, that did not heppens from today to tomorrow, but was ongoing until the end of this kind of religion with the christian and muslim expansion. Some of the gods merged, others become part or an aspect of an other deity. So yes, Nintinugga is at a point Gula and Bau/Baba. But before this moment, it was a independently godess. Over the centuries and in contact with other people, babylonians, assyrians, cassites, syrians and so on the system changed ofte, sometimes more, sometimes less. It was always fluid. So we need to have here a item for Nintinugga, one for Gula, one for Bau, one for Baba and at least also for their cross-entities. And this is truely a very problematic field. Here we would need a real expert. This would not be me. The whole system would need a scientific structure. Wolrdwide you don't have that much of those possible persons. On the other hand, Wikidata would give a phantastic structure for a catalogue of ancient mesopotamian gods. I try to think about it, maybe we can start here a project with external partners. Have to think about it.

Marcus Cyron (talkcontribs)

PS: end of this month I will be in UK. I'm invited to Warwick University and have an appointment at the Beazley Archive. To the second one becaus I want to inspire them with the possibilities of Wikibase.

Mike Peel (talkcontribs)

That sounds good, but complex! For now I've moved the enwp sitelink over to the other item. Have fun in the UK - it would be good if you could connect with @MartinPoulter: or Wikimedia UK while you're there!

Reply to "Nintinugga"
Wikielwikingo (talkcontribs)


Sorry. I didn’t intend to do such mess. I know that the german and italian articles refer to a sculpture which is in Altes Museum in Berlin, and the english, portuguese, spanish and swedish articles refer to a sculpture typus. I know that merging all of them is not exactly a fair procedure, but I thought it could be useful to have all that information gathered and available to readers and editors and later, as the articles would be made, change the links.
I see that now there’s no link between the german and italian articles, and it is needed. The german one is Apollo Kitharoidos (Berlin SK 44). The italian one is Apollo citaredo.

Sorry again, english is not my mother tongue.


Wikielwikingo

Marcus Cyron (talkcontribs)

Is OK, was just my frustration in the first moment. The problem is - and this has happened already to us all - that sometimes we do things in a good meaning, that not just fix errors, but also cause some problems. Problem with such things is, that there's mostly more than one version of a statue. You have the typus (and yes, you were intentionally right, here must be a central object). And then the ancient copies of the sculpture and the modern ones could all have their own data item. I think I have sorted it now for the existing items in a way, there are not longer problematic issues.

Reply to "Apollo Citharoedus"
Epìdosis (talkcontribs)
Reply to "Ancient Greece - period?"

Hungarian labels for Hungarian people

1
Tacsipacsi (talkcontribs)

Please do not add Hungarian labels for Hungarian people (like here). Hungarian uses Eastern name order, so the person-in-question’s name is Zsiros Janka, not Janka Zsiros. (I also had to dig a bit to ensure her surname is Zsiros and not Zsíros, as diacritics is often just dropped in foreign sources.)

Reply to "Hungarian labels for Hungarian people"

<no value> for P570 for the living?

5
Tagishsimon (talkcontribs)

Do we really add <no value> to date of death (P570) for living people? If so, why? And if so, do we also add <no value> to the 4001 other properties the person has no value for. I don't think we do. Your edit on Ulrike Peter (Q26839360) is honestly the first occurrence of a NV in P570 for a living person I have seen... so much so that I write a report to see how frequent it is. 77 occurrences.

select ?item ?dob where {  
  ?item wdt:P31 wd:Q5 . 
  optional {?item wdt:P569 ?dob.}
  ?item p:P570 [rdf:type wdno:P570] .}

Try it!

Marcus Cyron (talkcontribs)

@MB-one:, please take over. I can't.

MB-one (talkcontribs)

Hi @Tagishsimon,

whether living people should have a no value statement for date of death (P570) or no statement at all, is currently an undecided issue. A good reason to have the statement is data completeness and that in this way we have recorded the fact, that said person is indeed currently living, because they haven't died at the time when the statement was made. Not to have any statement can lead to the confusion whether the person is currently living or the date of their death just has not been recorded in Wikidata yet. That means less confidence in the actual status from the data at hand. On the other hand reasons against the statement could be, that some living people, who have Wikidata items could find it creepy, to even have a statement related to their death at all.

If you are interested in this topic, I would suggest that we bring this to the Project chat and discuss it there with the wider community. Until then I guess, we should preserve the status quo, which means, some items have a no value statement and others have no statement at all.

Cheers

Tagishsimon (talkcontribs)

Hi @MB-one: The status quo, if the SPARQL query is to be believed , appears to be 77 items with the statement, and ... how many living people have we on wikidata ... without. So I think use of <no value> to denote "is still alive" is an innovation; currently I think we tend to presume that the lack of a date of death indicates living. I don't yet see how or why <no value> improves on the current situation.

The trouble I have with it is that my experience of <no value> tends to be for data which will not ever exist - e.g. the director (P57) value for 1900-1920 films which did not record who directed them, or the publication date (P577) for a film that was cancelled or not released. I have not seen it used to say "we have no data /yet/". And as I indicated above, there are 10s or 100s of properties for which we could say <no value>.

So, yes, maybe we should take it to project chat. Equally, it's no biggie; I was merely that surprised to come across it that I thought it worth discussing. thx.

Tagishsimon (talkcontribs)
Reply to "<no value> for P570 for the living?"