Wikidata talk:WikiProject Genealogy

Wikimania 2016[edit]

Only this week left for comments: Wikidata:Wikimania 2016 (Thank you for translating this message). --Tobias1984 (talk) 11:47, 25 November 2015 (UTC)[reply]

Is there anyone active out here? I'm trying to make use of Wikidata item ids with Werelate.org. Mostly, I just want to use the Wikidata IDs as a language independent way to uniquely identify people. But I'm beginning to find some defects and was wondering how free I should feel to start editing links over here... --Jrm03063 (talk) 15:33, 1 February 2016 (UTC)[reply]

@Jrm03063: Hi! If you see mistakes you are more than welcome to fix them. If you are unsure you can also post some items here so other people can double-check. --Tobias1984 (talk) 17:33, 26 April 2016 (UTC)[reply]

Properties - don't simplify too much; see Familypedia's as a starter[edit]

Familypedia uses over 900 properties, though not all are for individuals. The intention is to have at least one property for every item that could be included in a GEDCOM file and then some.

Examples of properties that are used or are created but currently unused: day, month, year, street_address, locality/city, county, state, nation, latitude, and longitude of each marriage plus people in attendance plus five properties for the children of each marriage or liaison (list; notes; sources; primary sources; secondary sources). That's over a hundred for someone who had merely seven marriages, but at least two people out of the current 51,000 who have semantic articles had twice that number. Add the same time and place items for birth, baptism, death, burial, with some extras such as cause of death. Add derived properties such as each event's location (a string combining the locality, county, etc) and date (combining day/month/year - we've not encouraged people to code hours and minutes yet!), "Age maternal grandmother at birth", and the number of generations down from Charlemagne!

The system is only partly developed. We still have to code facts such as education, occupation, residences, hono(u)rs, and community service, to name only a selection. Some of the properties for them have been created but are unused. http://familypedia.wikia.com/wiki/Special:SemanticStatistics is meant to show the latest stats but is sometimes unrecognized even though it is listed at http://familypedia.wikia.com/wiki/Special:SpecialPages

Incorporating those and using many of the specially-developed templates (notably the "Showfacts ..." group) along with the three currently-used extensions, i.e. Semantic MediaWiki, Page Forms (which was called "Semantic Forms" until November 2016 but which Wikia hasn't yet upgraded to), and Semantic Drilldown, maybe with other semantic extensions that are not supported by Wikia, anyone could theoretically produce a good free genealogy wiki. If it were to have the status of a WikiMedia project, it could probably get more contributors, eventually, than Familypedia has.

Starting with fewer properties would be simpler but might lead to time-consuming revisions once the full potential of all possible ways of displaying genealogical data were to be exploited.

I hope we have people with programming and/or wikidata experience to put it all together.

Robin Patterson (talk) 12:39, 6 March 2017 (UTC)[reply]

Wikidata (genealogy) <-> WeRelate[edit]

The WeRelate Genealogy Wiki (Q7983244)contains "Person:" pages for about 2.8 million people. Of those, about 23,000 have been previously determined to be the subject of biographies on the English Language version of Wikipedia. The intent of that effort being to help boot strap the database with information for people who were apt to appear commonly in uploaded GEDCOM files - such that those pages become helpful in alignment/registration of overlapping GEDCOM loads. In addition, such pages would not be fleshed out with a basic extract from English Wikipedia, so they would not be as vapid as they might otherwise be.

Later in that initial effort to boot strap WeRelate, it could be seen that a limitation to the English Wikipedia only was not desirable. Also, problems associated with reliance on stable page names and longer term persistence of particular subjects began to appear as the English language Wikipedia changed over time.

It has subsequently became apparent, that the Wikidata identifier provides a superior means of establishing alignment between WeRelate Person pages and the wider Wikipedia biography set. The Wikidata identifier is language neutral and stable over time. Additional potential benefits beyond alignment with the wider Wikipedia biography set have also become apparent - almost too numerous to contemplate.

So I have begun a project intended to establish alignment between WeRelate Person pages and Wikidata Genealogy information. This WeRelate page contains introductory remarks on the effort for the WeRelate community.

Besides improving information presently on WeRelate - it now appears that benefits should accrue to Wikidata as well. While Wikidata is a great repository for information of many sorts - it doesn't seem like a great place to study and carefully organize such information. The presentation is not visually convenient for genealogical study. Further, critical bi-directional genealogy properties (for example - parent <-> child) need to be separately established for the parent and child. WeRelate was created specifically to deal with genealogy information so bi-directional relationships automatically propagate as appropriate.

The program being used to explore Wikidata, such that additional identifiers can be established in the WeRelate data, displays differences between the information stored on WeRelate and Wikidata. Most of the time, such differences are due to information present on Wikidata but not yet present on WeRelate. However, it has also become clear that a number of defects are present in the Wikidata genealogy information. Not a crime of course - any large data set is certain to have flaws. But this project provides an opportunity to more conveniently review Wikidata's genealogy information such that errors are more apt to be detected and resolved.

Would appreciate hearing from others interested in Wikidata genealogy information. --Jrm03063 (talk) 18:43, 11 May 2017 (UTC)[reply]

Hi I requested the Wikidata Property:P2949 for WIkiTree and I am Aleš Trtnik born in 1969 in Ljubljana, Slovenia.

I am maintaining WikiTree person ID (P2949) identifier. I add several hundred new entries each week derived from data comparison. I also handle redirects caused by merges of Last Name at Birth changes and remove the deleted and bogus entries.

I am also adding References to statements matching WikiTree data. has created a WikiTree project to check WikiTree profiles with Wikidata profiles see Project:Database_Errors. Question: does Werelate has an api so that you can retrieve all profiles with a Wikidata ID and check if there is any mismatches between WeRelate and WikiTree? - Salgo60 (talk) 21:34, 12 May 2017 (UTC)[reply]

I can see where such a property might eventually be useful, but I don't see it as a priority just now. It's trivial to do a search on WeRelate for a Person with a particular Wikidata ID.

If there were multiple WR people with the same Wikidata ID, they would all be returned by the same search.

There could - or may be an API to simplify access to WR data - but I've been getting by with tools that screen scrape. I routinely run a script that retrieves all the WR Person pages that have a Wikidata ID, looking for duplicates and comparing/contrasting the results with what's already present on Wikidata. I suppose someone could extend that to look at the Wikitree page referenced by Wikidata.

@Jrm03063: Wikidata currently doesn't have nice visualizations of genealogical information but there's no barrier to someone creating a visualization. Data entry could also be optimized.

I think the Wikidata model of having multiple kinds of data in the same database is useful. I recently researched Albert Einstein's wider family and meet Raphael Einstein (Q30302383). Being able to note that he is a partner (owns/is employeed) in B. Baruch und Söhne (Q28061405), which is the company founded by the father and brothers of his wife, makes sense for purposes of family history.

The Wikidata Alias function seems very helpful. I frequently notice people who are called differently in different databases. The Jette in one database might be called Annette in another. Wikidata is very easy to use in languages besides English. Currently my first impression of WeRelate is that it's very English focused.

Sorry, I didn't see your observation about language focus previously. I understand that we currently support English and Dutch, but others could be added. I would like to see us do something for pages that align with Wikidata IDs - such that the user would see a wiki extract in their preferred language (if a WP page in that language was available). --Jrm03063 (talk) 14:37, 27 June 2017 (UTC)[reply]

Being able to cross-use tools like QuickStatements and our query service is useful.

Nevertheless, I create a proposal for an authority control property for WeRelate. ChristianKl (talk) 21:20, 24 June 2017 (UTC)[reply]

The problem of a better genealogy interface for Wikidata really has both a technical part and a policy part. The technical part - defining associations that result in multiple property claims for example - seems like a straight-forward software problem. The policy part takes us into the question of how you do genealogy when you're restricted to working with people who satisfy notability requirements. I seem to recall seeing a compromise that suggests that adding a person that creates a link between existing notable people may be acceptable - but that's still too limited. I offer something like WeRelate as a better interface, because it solves the family connection problem by design, and isn't limited to people who satisfy notability requirements.

As far as learning relationship information that ought to be added to Wikidata, I should think Wikitree ought to be able to do exactly what I've been doing with WeRelate. The down side of Wikitree being that it doesn't protect the structure of GEDCOM uploads - so it can't export a GEDCOM that retains much of the useful GEDCOM structure. In so far as WeRelate preserves GEDCOM fact, source and note structure - WeRelate should be able to usefully exchange factual information with Wikidata beyond just familial relationships. Wikitree would need to do some work to serve similarly.

I've already had very good success, running code that grovels over people in WeRelate who have an associated Wikidata ID. From that information, I'm able to infer what the corresponding Wikidata relationship properties should be - which I then compare with what's actually present on Wikidata. --Jrm03063 (talk) 16:50, 26 June 2017 (UTC)[reply]

I think the biggest downside of Wikitree is that it doesn't have an open license. It's also a company run with a profit motive. I do consider WeRelate a big improvement over Wikitree.

As far as notability goes, Wikidata criteria of serious and public sources aren't very strict. If you do allow data about people for which there isn't any public record you open yourself up to the insertion of a lot of false data. ChristianKl (talk) 18:17, 26 June 2017 (UTC)[reply]

I'm trying to get some help using one of the more efficient methods of adding information to Wikidata. I tried using QuickStatements without apparent success. I've done a lot by hand, but that's time consuming and has its own perils. I could easily create a dump that follows the QuickStatements syntax as I understand it - and I know that I have thousands (perhaps tens of thousands) of claims that still need to be added. Thanks... --Jrm03063 (talk) 23:36, 4 July 2017 (UTC)[reply]

What problems came up when you tried to use QuickStatements? ChristianKl (talk) 23:42, 4 July 2017 (UTC)[reply]

I created a line that should have added a property. Nothing moved until I filled in the field at the top indicating something about where the data came from with a string indicating english wikipedia. The screen then cleared - but the page/person that should have picked up a claim did not - nor did I see anything appear on my list of contributions. --Jrm03063 (talk) 18:49, 5 July 2017 (UTC)[reply]

Maybe I spoke too soon. I just took another look at that page and noticed I had to establish some account settings so that QuickStatements would actually work. I'll try again... --Jrm03063 (talk) 01:09, 6 July 2017 (UTC)[reply]

Ok, I just tried again, placing the following in the quick statements text field, first column articles indicated as from "enwiki". Pressed "Do it", and the fields cleared.

Q6254449        P22     Q7287505
Q7342139        P22     Q16850394
Q7342139        P25     Q5297559
Q7342139        P3373   Q6129844
Q7342139        P3373   Q527871
Q7342139        P3373   Q5649831

Nothing seems to have occurred. A delay wouldn't surprise me - but this long - for so tiny a request? --Jrm03063 (talk) 14:57, 6 July 2017 (UTC)[reply]

@Jrm03063:, did you use one tabulation to separate fields ? Because I don't think white spaces work. --Melderick (talk) 12:32, 25 July 2017 (UTC)[reply]

@Melderick:, I was careful to use one, and only one, tab as a separator. I have some certainty on this, as the claim syntax is written by a program. I'm running my program again at this moment - and I'm planning to regenerate the claims again. I should have a fragment for testing shortly. --Jrm03063 (talk) 17:45, 25 July 2017 (UTC)[reply]

Property_talk:P4159 +24000 items re now uploaded and we have +40 000 items for Property_talk:P2949 - Salgo60 (talk) 16:54, 8 December 2017 (UTC)[reply]

Still actively working on WeRelate/Wikidata alignment. Over 32,000 WeRelate Person: pages are now tagged with Wikidata identifiers (more than 1% of the total WeRelate Person: page set). I have also been working the Wikidata side of things to add/correct genealogy claims - over 10,000 claims added since I began working over there. --Jrm03063 (talk) 21:39, 12 August 2018 (UTC)[reply]

Template talk:Marriage at Wikipedia[edit]

Notified participants of WikiProject Genealogy

Please come respond at w:Template talk:Marriage at Wikipedia on how marriages are encoded in infoboxes. I do not care which way you vote. This issue affects what data is imported from infoboxes into Wikidata. --Richard Arthur Norton (1958- ) (talk) 17:47, 25 October 2017 (UTC)[reply]

References to civil registers?[edit]

Notified participants of WikiProject Genealogy. What is the best way to add a reference for Western Australian Registry of Births, Deaths, and Marriages (Q42333722), given that each reference needs to indicate registration district, year, and number. I'm testing on Anne Eliza Back (Q42333974) with location (P276), publication date (P577), and inventory number (P217), but I don't think these are the best. Thanks! —Sam Wilson 04:38, 31 October 2017 (UTC)[reply]

I've created Wikidata:Property proposal/civil registration district. Sam Wilson 02:12, 22 December 2017 (UTC)[reply]

Don't forget to update her in Familysearch (free registration) See: https://www.familysearch.org/tree/person/details/KNXP-HQJ It is the rare free genealogy site that has actual documents or minimally indexes to documents, and has room to write a biography. --RAN (talk) 13:38, 13 November 2019 (UTC)[reply]

@Richard Arthur Norton (1958- ): I do try to keep FamilySearch up to date, but the main non-Wikidata project I work on is now WikiTree. :-) Anyway, I'm back wondering again how best to handle these references, and it seems like it's more common to use section, verse, paragraph, or clause (P958) for the registration number (judging by this sort of query). Do you think that's correct? An example is Frederick Samson (Q5498696). Sam Wilson 09:55, 7 March 2024 (UTC)[reply]

Patronyms[edit]

Notified participants of WikiProject Genealogy I also asked this at Project_chat: I am not sure where the discussion left off on patronymics. We now have Property:P5056 where we add in data like Jóhannesson (Q25935009) for Icelandic names. Norwegian and Swedish patronymics became surnames around 1900 and were adopted by families that migrated to the US earlier. Is the plan to create an entry for "Hanson (patronymic)" to distinguish it from "Hanson (family name)" or is the plan to just add "Hanson patronymic" as an also_known_as at the entry "Hanson (family name)"? You can see what that would look like here: Hanson (Q9145324).

What are the best modelled items for your areas of interest?[edit]

Hi all

Over the past few months myself and others have been thinking about the best way to help people model subjects consistently on Wikidata and provide new contributors with a simple way to understand how to model content on different subjects. Our first solution is to provide some best practice examples of items for different subjects which we are calling Model items. E.g the item for William Shakespeare (Q692) is a good example to follow for creating items about playwright (Q214917). These model items are linked to from the item for the subject to make them easier to find and we have tried to make simple to understand instructions.

We would like subject matter experts to contribute their best examples of well modelled items. We are asking all the Wikiprojects to share with us the kinds of subjects you most commonly add information about and the best examples you have of this kind of item. We would like to have at least 5 model items for each subject to show the diversity of the subject e.g just having William Shakespeare (Q692) as a model item for playwright (Q214917), while helpful may not provide a good example for people trying to model modern poets from Asia.

You can add model items yourself by using the instructions at Wikidata:Model items. It may be helpful to have a discussion here to collate information first.

Thanks

John Cummings (talk) 15:28, 17 December 2018 (UTC)[reply]

Identifier property clarifying it does not imply notability - for catalog references?[edit]

The property WikiTree person (P2949) was created with a clarifying "an identifier that does not imply notability (Q62589320)". Seeming contrary, this example suggests Western Australian Registry of Births, Deaths, and Marriages (Q42333722), a catalog, does imply notability.

I think the codifying clarity in terms of what imply notability or not is a good thing, and this could simplify rejecting trigger-happy editors looking to evade the intention of notability criteria. Anyone can elaboate on the state of this matter? Best regards, CarlJohanSveningsson (talk) 10:12, 12 November 2019 (UTC)[reply]

I am not sure such a property is necessary, since just being in an external database should not imply notability ever. I do think you need more than one angle for edge cases and I try to do this whenever possible, though I have noticed in the past that among my early additions to Wikidata there are items which up to today still only have one external link. I would like to see the "structural need" rule broken down into "notability clauses" whereby a clause could be "direct relation", which applies to this project, but also "creator" (of a notable work on Wikidata) or "founder" (of a notable institution or business on Wikidata) or "sitter" (of notable painting(s) on Wikidata), or "fill in other notable thing here". Breaking these down, you might be able to make a case the other way, by which I mean an external link to XXX does mean notability. So instead of a blacklist, keep a whitelist with explanation and comments. Otherwise known as a WikiProject! Jane023 (talk) 11:30, 12 November 2019 (UTC)[reply]

I am not positive, but am pretty sure that everyone in De Vroedschap van Amsterdam, 1578-1795 (Q28329667) is notable enough for Wikidata, just because of its importance in Amsterdam archives and Dutch history. So yes, there are (finite) databases that could be whitelisted per country/province/state/city, etc. Jane023 (talk) 11:40, 12 November 2019 (UTC)[reply]

"An identifier that does not imply notability" appears to be the project of a small group of people, and does not reflect the consensus of the general community. The concept does not trump Wikidata:Notability, which states "It refers to an instance of a clearly identifiable conceptual or material entity. The entity must be notable, in the sense that it can be described using serious and publicly available references.", which is the opposite of what is stated. I understand that there are several "public and serious" databases that allow unchallenged crowdsourcing, so that any living person can add themselves, and perhaps this is what the original concept sought to identify, and I agree with that narrow concept. --RAN (talk) 01:06, 13 September 2022 (UTC)[reply]
I agree that including identifiers from sites that allow anyone to create records is likely to not be enough to permit a Wikidata item be created. I mean, at extremes that could mean that every X username (P2002) should be included! However, it sounds like maybe the discussion here is aiming at something like "we shouldn't include every Wikitree person unless they're also needed for some other purpose". Is that right? I often create items for people here and link them to Wikitree, but generally there's some other purpose (e.g. depicts at Commons, or just making the connections between other already existing items). I hope that's okay. Sam Wilson 03:08, 13 September 2022 (UTC)[reply]

Wikidata Notability[edit]

Addressing: "just being in an external database should not imply notability ever"
Rule 2.0 of Wikidata:Notability appears to allow any actual person that existed to be Wikidata-notable: "[The Wikidata entry] refers to an instance of a clearly identifiable conceptual or material entity. The entity must be notable, in the sense that it can be described using serious and publicly available references." Any non-living person that can be tied to a document to show that they actually exist should be Wikidata-notable. Are we choosing appearances in various databases to show importance, or to show literal existence? Are we trying to weed out fake entries or weed out non-important people? Of course living people have their own rules, mainly about what information we should include in their entry. I just want to be sure we are not veering toward English Wikipedia standards of being famous, rather than actually existing. --RAN (talk) 13:25, 13 November 2019 (UTC)[reply]

@Richard Arthur Norton (1958- ): I agree with your point of view. I have started a project to show how we can go further on genealogical information here: Wikidata:WikiProject_Genealogy/Gipuzkoa -Theklan (talk) 10:02, 24 December 2019 (UTC)[reply]

Clearly we have priorities and are uploading the most useful people first. There would be very little use for uploading 7 billion entries for every living person, there would be no way to disambiguate them. We would just have a 10,000 entries for John Smith that would just make it more difficult to find the John Smith most people are looking for. We already have a list of The Peerage entries that are ambiguous and we have a list of ORCID scientific researchers that are ambiguous. Merging those entries with existing people is already exhausting. --RAN (talk) 18:32, 12 April 2020 (UTC)[reply]

As best I can describe it, we have a priority to load information dense people first and avoid information sparse people. As an example the VIAF database has millions of entries of authors, some very clearly defined and information dense, we have the names of books they wrote and their birth and death dates, and usually their middle name. VIAF also has over 1 million information sparse entries with a very common name of a person, no middle name, no dates of birth or death, and the name of a single book they wrote. These would be a very low priority to add to Wikidata. They may be a match for an entry we already have but the ability to match would be very low. That would not restrict a person to add some of the information sparse people by hand, but the community would most likely not allow a bot to load them all, at this time. Priority goes to the information dense entries. --RAN (talk) 16:01, 6 September 2020 (UTC)[reply]

Charlemagne[edit]

The number of descendants in Wikidata is now >100,000: see WikiProject Genealogy/numbers/descendants.--- Jura 19:55, 29 November 2019 (UTC)[reply]

property for deletion: Familypedia person ID (P4193)[edit]

Please see Wikidata:Properties_for_deletion#Property:P4193. --- Jura 20:13, 26 March 2020 (UTC)[reply]

Help:Import BLKÖ from wikisource#Genealogical tables[edit]

Notified participants of WikiProject Genealogy

Above a short note about genealogical tables transcribed at Wikisource. There are some 300 included in an Austrian biographical dictionary of the second half the 19th century. While we probably already have data about the many members of the Habsburg family it includes, others might be less covered. Item creation for articles/people in the dictionary is still ongoing. I find that for much content of the work it doesn't actually matter that it's in German, but sometimes the names used aren't the ones used in English/today. I'm trying to find useful statements to add to the items about the tables. --- Jura 18:22, 12 April 2020 (UTC)[reply]

Maybe not what you ask for but using Commons:SDC and mark the name in the picture with a link to the person would be helpful. I have tested it for gravestones see image position tool / c:File:Gävle_gamla_kyrkogård_14_03_05_838000.jpeg - Salgo60 (talk) 01:55, 13 April 2020 (UTC)[reply]

Thanks for the suggestion: full annotations might work well for the smaller sized ones and could give interesting illustrations of the content.

I'm asking as I'm not quite sure which way to go with them. The tables are bit like reports from a small, old genealogy database. --- Jura 07:43, 13 April 2020 (UTC)[reply]

Ancestors viewer[edit]

Is the Ancestors viewer designed to discreetly omit evidence of inbreeding? I'm sure that's not the intent but I have noticed when there are duplicate pairs of grandparents, one pair is omitted. Gamaliel (talk) 14:36, 23 April 2020 (UTC)[reply]

Q1815122[edit]

We need a more reliable source about this person's mother. Currently English and French Wikipedia is saying differently.--GZWDer (talk) 23:07, 12 June 2020 (UTC)[reply]

Template:Wikidata/FamilyTree[edit]

The template adds color to 5 generations in each direction, can anyone figure out how to get the colors to repeat again for subsequent generations? We have charts for some people with 7 generations. --RAN (talk) 19:45, 6 September 2020 (UTC)[reply]

I figured it out and added in colors for three more generations. --RAN (talk) 00:59, 13 September 2022 (UTC)[reply]

Adopted children/parents[edit]

What is the proper way to document adopted children/parents? Thanks. Gamaliel (talk) 00:40, 14 November 2020 (UTC)[reply]

One way is to create "relative" in the parent's record with this child and to give him a qualifier "kinship to subject" with "adopted child".

And for the child record to create "relative" with this parent and to give him a qualifier "kinship to subject" with "adoptive parent". Alexandre.rozanov (talk) 18:10, 15 January 2024 (UTC)[reply]

Strange parents[edit]

Hi

Notified participants of WikiProject Genealogy

Today, I stumble upon Sir William de Stanley (Q75616810) who has 3 fathers and 3 mothers. Even, worse 2 of the fathers has for father the first father... Thankfully there is rank but still, this seems suboptimal.

Has anyone guidance of what to do in such case?

Cheers, VIGNERON (talk) 17:30, 10 May 2021 (UTC)[reply]

When mass imports are done I've found there are a lot of duplicates, so these probably are all potential merges. Gamaliel (talk) 18:10, 10 May 2021 (UTC)[reply]

Try to understand what the items are about and what the sources say. This might lead to merging items that are about the same person. It might also lead to the conclusion that one source is right while another isn't, then you deprecate the sources you believe to be wrong. If you can't tell which source is right with reasonable certainty you leave it with multiple claims but may set the most likely one to be correct as best rank. ChristianKl ❪✉❫ 18:22, 10 May 2021 (UTC)[reply]
That's why I always prefer to work with primary sources, instead of those genealogic compilations... I suspect that in order to settle that, someone has to dig into the primary sources and see what is there in what context.-- Darwin ^Ahoy! 18:38, 10 May 2021 (UTC)[reply]

There seem to be some incosistencies in the sources such as Geni, I'll try to resolve, this just shows that Wikidata is very good at the detecting possible conflicts. Germartin1 (talk) 01:44, 11 May 2021 (UTC)[reply]

Thanks all for your answers!

@Gamaliel: indeed, it's probably but merge is not obivous here and we need to clean the items first (otherwise, it's will block as someone would end up being their own father).

@ChristianKl: I tried but it's a messy situation, 4 men with almost the same name and no reliable sources. That's why I'm asking for help here ;)

@DarwIn: primary sources are maybe better than tertiary sources (debatable but let's assume) but here, we have too few information to even know where to find primary source (and given the date, even for noble people it's possible there is no more primary sources). And anyway, secondary sources are by far the best (as this situation demonstrate).

@Germartin1: yeah, I looked at Geni and I'm even more confused now (if it's even possible).

Below is a table to try to better view the situation.

Qid	dates	identifiers
Sir William de Stanley (Q75616810)	1368-?	genealogics.org person ID (P1819), Kindred Britain ID (P3051), The Peerage person ID (P4638), WikiTree person ID (P2949)
William de Stanley (Q75616710)	died 1360 (according to genealogics and kindred britain), 1398 (according to thepeerage and roglo)...	genealogics.org person ID (P1819), Geni.com profile ID (P2600) (twice), Kindred Britain ID (P3051), Roglo person ID (P7929), The Peerage person ID (P4638), WeRelate person ID (P4159), WikiTree person ID (P2949)
William de Stanleigh (Q96075770)	1337-1398	Kindred Britain ID (P3051)
William de Stanley, of Stourton (Q99075987)	1337-1398	genealogics.org person ID (P1819)

Clearly this is a mess and the two laters are the most problematic ones and should probably merge (together but then, with the first or the second? evidence points both ways) and maybe the second need tobe split... In order to clean these items, we really need some work and good and reliable sources.

Cheers, VIGNERON (talk) 11:54, 11 May 2021 (UTC)[reply]

@VIGNERON: "Primary sources" in historical terms (that is, contemporary or close in time to the subject). They are generally secondary in "wikipedist" terms - and reliable - including original parish records, which are generally signed by witnesses and reviewed and conferred by a visitor. In this case, if the source does not mention where they got the information, I suggest discarding them entirely, since it's very much unreliable.-- Darwin ^Ahoy! 19:33, 11 May 2021 (UTC)[reply]

@VIGNERON: From what I understand from [[1]], we have :

Sir William de Stanley (Q75616810), married to Alice Massey, died around 1360 or 1398.
William de Stanleigh (Q96075770)/William de Stanley, of Stourton (Q99075987), married to some Agnes, died in 1398, not mentioned by some sources.
Sir William de Stanley (Q75616810), married to Margery de Hooton, died around 1428.

Now how to solve this ? First, merge William de Stanleigh (Q96075770) and William de Stanley, of Stourton (Q99075987). Do the same for his wife.
Second, I would tend to follow the sources that include the second William. This means I would deprecate the links between the first William and the third. --Melderick (talk) 22:38, 12 May 2021 (UTC)[reply]

Medlands also include the second William : [[2]] --Melderick (talk) 22:51, 12 May 2021 (UTC)[reply]

A series of genealogical entries is up for deletion[edit]

Wikidata:Requests_for_deletions#Ancestors of Sven-Göran Eriksson – The preceding unsigned comment was added by Richard Arthur Norton (1958- ) (talk • contribs) at 03:45, 6 September 2021‎ (UTC).[reply]

FamilySearch[edit]

Notified participants of WikiProject Genealogy -- Also pinging @DrThneed, Ambrosia10: who I think both have used the Familysearch site for research a bit.

Question: Is there anybody here, or do we know anybody, that has contacts at FamilySearch, either in the management or the community ?

Looking at the statistics on identifiers on the project main page, I find it striking that even though FamilySearch is by a clear order of magnitude the biggest database listed, we currently only have about 12,000 items matched to it -- it's not even in the top ten. Could we do anything to increase this? (eg earlier this year, when I was doing some wikidata clean-up, I found it was a really nice site for making sense of and organising together UK 19th-century census records to clarify family relationships, as well as marriage records etc).

One nice thing on the FamilySearch site is the integration they have with Find-a-grave and with BillionGraves -- as BillionGraves write on the FS wiki:

BillionGraves is a certified partner of FamilySearch with read-and-write integrations into the FamilySearch Family Tree.

In practice I think this means that FamilySearch keeps up-to-date extracts from both the Find-a-grave and BillionGraves databases. The FS system monitors these, and if there are enough points of similarity between the information on a FamilySearch person page and the information on someone from one of their entries, FamilySearch will show a prompt under 'research help' on the person page (as it also does for potential matches for other sorts of record), suggesting that the reader compare to see whether it is indeed the same person -- and at the same time also link any related people that match, that are also referred to in the entries.

I was wondering whether we might ever be able to persuade them to implement something similar with Wikidata? Both sides I think would gain from being able to link to the other, and the auto-suggestion of a wikidata/wikipedia link (with all of the biographical resources we cross-reference) might sometimes be very useful to a family history researcher using FamilySearch, as well as for us in the other direction.

I would have thought that the structured nature of Wikidata ought to make it very easy to be able to extract genealogically useful event and relationship information out of a scheduled dump of all Q5s, together with an auto-text summary; and then to munge that into a format very similar to whatever they are using to summarise Find-a-grave and BillionGraves entries, that could then be fed into something very very similar to the machinery they are already using to offer and confirm matches for those sites. And when a user did confirm (or un-confirm) a match, a bot-account could automatically reflect that action here. We could then start to get to rather more than the 12,000 FS links for the 10 million items we currently have for humans.

I do find FamilySearch links can be incredibly useful, particularly when entries on the FS site have been cross-linked to all the relevant vital records registrations and censuses. I think systematic Wikidata links could be useful for FamilySearch researchers too (and through Wikidata all the specialist databases). Does it seem at all possible that we might be able to interest FamilySearch in the same kind of "valued partner" with "read-and-write integration" arrangement that they have with Find-a-Grave and BillionGraves?

If it's not completely impossible, how can we go about trying to make an introduction for ourselves? Who ought we to be trying to get to talk to who? Jheald (talk) 20:55, 23 December 2021 (UTC)[reply]

Jheald FYI I did 2018 an activity with Family Search / Danielle Batson (guess this is her profile) and Swedish parish pages see T201404 maybe we can point on this discussion and ask the same contact person if there is an interest in doing something and how we progress? - Salgo60 (talk) 21:03, 23 December 2021 (UTC)[reply]

@Salgo60: Sounds like she might be an excellent person to tell us who within FamilySearch we should maybe try to be in touch with. (And, more informally, whether this is something FS might perhaps be receptive to). Jheald (talk) 21:08, 23 December 2021 (UTC)[reply]

@Jheald: wrote this "Wikidata integration" please dont hesitate to correct my english - Salgo60 (talk) 21:23, 23 December 2021 (UTC)[reply]

@Salgo60: Thanks, looks good. Above I was primarily thinking of whether FamilySearch <--> Wikidata matching might be possible within the FamilySearch platform, using FamilySearch's established matching interface, and making the process ongoing and something that the FamilySeach community could be involved in. But, as you suggest, if FS would trust us with a database dump, we could also look at matching on our side, as we do and have done for other databases (albeit that the FS database is so very very big). Jheald (talk) 21:57, 23 December 2021 (UTC)[reply]

@Jheald The future will tell ;-)

I have had some contact with Lesko987a who is doing the Wikidata <-> Wikitree connection and he has

templates for same as Wikidata
scans the text in WikiTree for links to Wikipedia

maybe FamilySearch has dumps/API for that?!?!?!

As Wikitree has "data doctors" checking 22 million profiles every week see status report maybe we could have the same with Family Search....

I also would like to start the process of an echo system of correcting profiles and send signals of errors between sites using Wikidata:Mismatch_Finder Salgo60 (talk) 22:11, 23 December 2021 (UTC)[reply]

For what it's worth, just for interest, current counts of cross-referenced IDs https://w.wiki/4bLv and of wikipedia links https://w.wiki/4bLu for items with FamilySearch IDs. Jheald (talk) 22:16, 23 December 2021 (UTC)[reply]

I think the FS people would especially be interested since we have images for people that they do not have in their database. We loaded the entire Bain Collection (Q67388890) from 1910 to 1930 to Commons and most are now linked to a person in Wikidata. --RAN (talk) 22:32, 23 December 2021 (UTC)[reply]

@Jheald, Richard Arthur Norton (1958- ): thanks hm SPARQL --> 4601 without a Wiki article

I guess its impossible to guess the future of "digital" genealogy. In Sweden we got Swedish Portrait Archive ID (P4819) last year that contains > 880 000 scanned portraits see https://xn--portrttarkiv-kcb.se/about and they have an excellent API so I have build an integration to Wikicommons and created Notebooks finding all people at a cemetery without a picture and see if we get a match... see GITHUB spa2Commons
this week I started an dialogue with people at Uppsala University that use ML for structure all Swedish parliament speeches the last 100 years, looks like Wikidata/sv:Wikipedia is the best database for who is in the parliament in Sweden, see GITHUB Wikidata_riksdagen-corpus, I try to make them interested in WIkibase and en:Graph_database and maybe we could do NER on locations and on subjects what they speak about in the Swedish Parliament ....

- Salgo60 (talk) 23:10, 23 December 2021 (UTC)[reply]

ChristianKl (talk) 15:11, 24 June 2017 (UTC) Melderick (talk) 12:22, 25 July 2017 (UTC) Richard Arthur Norton Jklamo (talk) 20:21, 14 October 2017 (UTC) Sam Wilson Gap9551 (talk) 18:41, 5 November 2017 (UTC) Jrm03063 (talk) 15:46, 22 May 2018 (UTC) Salgo60 (talk) 18:10, 18 June 2018 (UTC) Egbe Eugene (talk) Eugene233 (talk) 03:40, 19 June 2018 (UTC) Dcflyer (talk) 07:45, 9 September 2018 (UTC) Gamaliel (talk) 13:01, 12 July 2019 (UTC) Pablo Busatto (talk) 11:51, 24 August 2019 (UTC) Theklan (talk) 19:25, 20 December 2019 (UTC) SM5POR (talk) 20:17, 29 May 2020 (UTC) Pmt (talk) 23:22, 27 June 2020 (UTC) CarlJohanSveningsson (talk) 12:13, 30 July 2020 (UTC) Ayack (talk) 14:39, 12 October 2020 (UTC) EthanRobertLee (talk) 19:17, 20 December 2020 (UTC) -- Darwin ^Ahoy! 18:20, 25 December 2020 (UTC) Germartin1 (talk) 03:13, 30 December 2020 (UTC) Skim (talk) 00:13, 10 January 2021 (UTC) El Dubs (talk) 21:55, 29 April 2021 (UTC) CAFLibrarian (talk) 16:36, 30 September 2021 (UTC) Jheald (talk) 18:50, 23 December 2021 (UTC)[reply]
Notified participants of WikiProject Genealogy If you use Familysearch, you just received a notification there, that some images and stories that are stored there, have been accidently deleted. Familysearch is offering a recovery tool to be able to restore the images from any backup you kept on your computer or other genealogy online program. That is why I always load images to multiple places, Findagrave, Ancestry, Geni, Wikitree, and Flickr. --RAN (talk) 18:00, 30 August 2023 (UTC)[reply]
@Richard Arthur Norton (1958- ) Thank you very much for the heads up! Darwin ^Ahoy! 18:46, 30 August 2023 (UTC)[reply]
The explanation is here: https://www.familysearch.org/en/help/helpcenter/memories-restoration-dashboard They lost 108 biographies or images for my account. --RAN (talk) 23:03, 30 August 2023 (UTC)[reply]
Thank you for the heads-up. I seem to have been spared any lost content, and luckily I too store across Genealogical services and locally so it would be at most an annoyance if it occurred.

Here's hoping for a future where wikidata can be a home for a global tree, and Commons a home for images/documents. ElDubs (talk) 23:09, 30 August 2023 (UTC)[reply]
@ElDubs: Isn't Commons already that home? :-) Certainly, I've tried to only add photos to FamilySearch where I've also uploaded them to Commons (or the InternetArchive, for post-1955 ones). Interesting that WikiTree has also just announced a data breach. Sam Wilson 00:36, 31 August 2023 (UTC)[reply]
There is a small group of people that nominate entries for deletion based on the concept of English Wikipedia, which is some sort of fame. At one time someone nominated every image I uploaded to Commons as retaliation for opposing them in a debate at English Wikipedia. Just recently some images of people I uploaded with Wikidata entries were nominated for deletion for the 6th time, but this time, none were deleted. Again this was retaliation for opposing someone in a debate for deletion. RAN (talk) 01:47, 31 August 2023 (UTC)[reply]
Precisely. This very page a couple sections down lists a number of items for deletion simply for "no notability". A number of Wikipedia editors seem to try and apply those rules on other wikis. It brings up the old debate of Deletionism and inclusionism in Wikipedia.

I think before Commons and wikidata can become a home for genealogists, these wikis need a rule/policy that is clear so these nominations for deletion can just be speedily closed.

I had a discussion that got a response from someone at WMF who explained wikidata currently does have structural limitations that holds back the indiscriminate adding of information, and that they are working on a fix so that wikidata doesn't reach capacity. I think that this would be a barrier for genealogists. We bring with us a lot of information. There are ~106,000,000 data items on wikidata. That's (no offence) peanuts for the genealogical community. FamilySearch alone has over 1.3 billion searchable people. We could very quickly become one of the largest contributers of information. And despite what a lot of people like to say about the quality of genealogical research, a significant amount of it is solid information.

I do hope wikimedia will be ready for us one day. ElDubs (talk) 21:56, 31 August 2023 (UTC)[reply]

Non-contemporaneous siblings[edit]

Notified participants of WikiProject Genealogy

As part of the project-without-end to identify item duplicates in The Peerage person ID (P4638) people, a query I return to and gnaw at from time to time is tinyurl.com/2p8jwkve -- apparent pairs of children of a parent, that share the same given name (P735). The query ignores any pairs connected by sibling (P3373), giving a way to mark and exclude false positives.

However (perhaps it's just me?) but it seems a bit strange to mark people as siblings, if their lives did not overlap at all -- eg these examples: https://w.wiki/4bMC I feel it would be usefully informative, and reduce dissonance (at least mine), if we marked when this is the case, ie with an appropriate qualifier on the sibling (P3373) statement.

But which qualifier, and what value? I see that sourcing circumstances (P1480) = posthumous (Q60503972) is in limited use, for award received (P166), military rank (P410) and (one case of) spouse (P26). (Query: https://w.wiki/4bMU) Would that make sense here? ie

sibling (P3373) = X / sourcing circumstances (P1480) = posthumous (Q60503972).

And is this a 'sourcing circumstance' when applied to siblings?

Or: would a new purpose-make pair of values make useful sense, eg

sibling (P3373) = X / sourcing circumstances (P1480) = "born after death of subject" / "died before birth of subject"

(Also again: is sourcing circumstances (P1480) the right choice for the qualifier, or something else?).

Or: do people think all/any of this would just be unnecessary elaboration, and not actually useful? Jheald (talk) 23:22, 23 December 2021 (UTC)[reply]

I think it was discussed on project chat a while back. Technically, beyond the mother and the spouse, no relatives need to be contemporaries. --- Jura 02:27, 24 December 2021 (UTC)[reply]

Gravemap seems to be abandoned (so maybe mention that one the page)[edit]

website is down and github project hasn't seen any recent commits Thibaultmol (talk) 16:08, 18 July 2022 (UTC)[reply]

How notable do you have to be to get an entry at Wikidata[edit]

There is a debate at Wikidata:Administrators' noticeboard about how notable do you have to be to get an entry at Wikidata. These are dead people, so there is no self-promotion going on. RAN (talk) 18:16, 10 August 2022 (UTC)[reply]

Well, if those dead people are related to you or member of your particular special interest group, that’s about as problematic for Wikidata as self-promotion … --Emu (talk) 18:31, 10 August 2022 (UTC)[reply]

You are extending the definition of "self" to anyone "related to you" and anyone you are interest in, which is absurd. The alternative is to ask people to contribute their time to create entries for people they are not interested in. I am going to have to repeat: Dead people don't get involved in self-promotion, because they are dead. --RAN (talk) 22:38, 10 August 2022 (UTC)[reply]

People who engage in genealogy generally want to contribute accurate data, while people who engage in self promotion don't care about accuracy. Self-promotion brings inherent problems that don't exist in the same way for genealogy. ChristianKl ❪✉❫ 21:53, 10 August 2022 (UTC)[reply]

Query for line of descent between two people[edit]

I am trying to find the line of descent between two people (in this specific case, Eleanor of Aquitaine and Winston Churchill). However, the query I made results in a timeout. (Sadly, I lost the code, but the main idea was to find the people who are descendants of Eleanor and ancestors of Churchill at the same time - due to the exponential nature of this query, it is not surprising that it fails, though.) Could there be a more efficient way to find this using Wikidata?
Thanks! Krmarci (talk) 20:50, 28 November 2022 (UTC)[reply]

@Krmarci: If you want to find the path between the two people, you can use Geni. You put a push pin in one person then search for the second person, and then click "find path". Ping me if you need more help. --RAN (talk) 19:15, 1 May 2023 (UTC)[reply]
This query works well, though this relies on BlazeGraph-only feature that will stop working once Wikidata Query Service moved to other backend. GZWDer (talk) 16:56, 31 August 2023 (UTC)[reply]

Imports: married name versus maiden names[edit]

We just had a large import from GENI that brought in women under their married names, while previous imports from other websites have been using their maiden names. What is the standard that we want for Wikidata? If we don't have a standard, we will not be able to match to existing entries when we import large databases like Findagrave in the future. RAN (talk) 12:34, 1 May 2023 (UTC)[reply]

I would think Wikidata would follow the same practice as Wikipedia in this instance, which is using the name most associated with that person. For most people, that means using their married name.

For the purposes of genealogy and matching people, we can use birth name (P1477). ElDubs (talk) 00:28, 2 May 2023 (UTC)[reply]

Part of the problem with married name is that in the Wikidata Familytree we have what appears to be incestuous marriages, where people with the same surname are marrying. We long ago imported all the famous people, where could do a Google search, and quantitively determine their most common name. We are now importing ordinary people, where we have to come up with a standard if we want to match new imports to existing entries, to avoid duplicates. We ended up with over 1,000 duplicates from The Peerage because we imported people as "Sir John Doe" and "Prince John Doe", where we already had "John Doe". --RAN (talk) 21:01, 2 May 2023 (UTC)[reply]

Alternative mothers and fathers[edit]

Notified participants of WikiProject Genealogy

What are the rules for alternative mothers and fathers in Wikidata? In most of the collaborative genealogical projects (like Rodovid, GENI etc) the rule is that only one father and one mother are allowed in the genealogical tree. These should be the most probable father and mother. Other alternative fathers and mothers (with lower probability) should be mentioned and linked only in the "Note" or "Comment" areas, but not added to the main genealogical tree. Is it the same for Wikidata? If yes, in which area one can put the information and the link for alternative parent? If not, what are the actual rules? Alexandre.rozanov (talk) 15:24, 16 January 2024 (UTC)[reply]

One can probably deal with alternative mothers, fathers and children using the Ranking mechanism in Wikidata. For example, it means to introduce multiple fathers, but to give to the most probable father "preferred rank" and to all others "deprecated rank". If possible to give "reason for preferred rank" with a reason for preferred rank (P7452) qualifier and "reason for depreciated rank" with with a reason for deprecated rank (P2241) qualifier. One can also state that the father is disputed and by whom: the qualifier statement disputed by (P1310) . Is there actually a consensus of using this Ranking mechanism for alternative fathers, mothers and child in the Wikidata Genealogy community? If yes, could somebody give some examples? --Alexandre.rozanov (talk) 13:30, 18 January 2024 (UTC)[reply]

One more way to deal with alternative mothers, fathers and children in Wikidata is to introduce alternative person as "relative" NAME "kinship to subject" "may be father" (or "may be mother" or "may be child"). Is there any rule in which case to use "may be ..." relative and in which case to use Ranking?--Alexandre.rozanov (talk) 16:31, 18 January 2024 (UTC)[reply]

Wikimedia disambiguation pages for people with the same name[edit]

See the discussion at Wikidata:Project_chat#Wikimedia_disambiguation_pages. Currently we already have a disambiguation page in Wikidata, but the disambiguation is done at Wikipedia. Wikidata has more entries than Wikipedia to disambiguate, Wikipedia only includes "famous people". I want to allow the disambiguation here at Wikidata too. It is much easier than listing "different from" in 5 entries, the permutations become too difficult to manage. See for instance Governor Hughes (Q5589569). RAN (talk) 20:13, 22 February 2024 (UTC)[reply]

Referencing BMD registration numbers with P248[edit]

I'm working on a little tool to help add references to people who have birth, marriage, or death records in Western Australian Registry of Births, Deaths, and Marriages (Q42333722). Thus far, it's adding three properties for the references: stated in (P248), publication date (P577), and section, verse, paragraph, or clause (P958). Is the last of these correct for the "registration number"? It will also have registration district (P5564) once I've finished populating another ~70 items for those. Sam Wilson 02:41, 21 March 2024 (UTC)[reply]

Wikidata talk:WikiProject Genealogy

Wikimania 2016[edit]

Properties - don't simplify too much; see Familypedia's as a starter[edit]

Wikidata (genealogy) <-> WeRelate[edit]

Template talk:Marriage at Wikipedia[edit]

References to civil registers?[edit]

Patronyms[edit]

What are the best modelled items for your areas of interest?[edit]

Identifier property clarifying it does not imply notability - for catalog references?[edit]

Wikidata Notability[edit]

Charlemagne[edit]

property for deletion: Familypedia person ID (P4193)[edit]

Help:Import BLKÖ from wikisource#Genealogical tables[edit]

Ancestors viewer[edit]

Q1815122[edit]

Template:Wikidata/FamilyTree[edit]

Adopted children/parents[edit]

Strange parents[edit]

A series of genealogical entries is up for deletion[edit]

FamilySearch[edit]

Non-contemporaneous siblings[edit]

Gravemap seems to be abandoned (so maybe mention that one the page)[edit]

How notable do you have to be to get an entry at Wikidata[edit]

Q15268919[edit]

Q111333077[edit]

Q111336806[edit]

Q111333170[edit]

Q111336807[edit]

Q107290947[edit]

Q107291515[edit]

Q107291489[edit]

Q107291485[edit]

Q107291182[edit]

Q107291556[edit]

Q107291500[edit]

Q107291945[edit]

Q107297872[edit]

Q107297869[edit]

Q107297904[edit]

Q107218828[edit]

Q107291969[edit]

Q107297988[edit]

Q107340578[edit]

Q107340552[edit]

Q107297915[edit]

Q108130039[edit]

Q112366095[edit]

Query for line of descent between two people[edit]

Imports: married name versus maiden names[edit]

Alternative mothers and fathers[edit]

Wikimedia disambiguation pages for people with the same name[edit]

Referencing BMD registration numbers with P248[edit]

Navigation menu

Search