User talk:Dogfennydd

Welcome to Wikidata, Dogfennydd!

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! ミラP 17:29, 19 August 2020 (UTC)

P2949

Thanks for this edit. Can you also make the fix I requested for Chinese language here? Q699605102 (talk) 19:38, 30 November 2021 (UTC)

Not being a Chinese speaker, I’m afraid I can’t. Perhaps try here? Dogfennydd (talk) 20:40, 30 November 2021 (UTC)

Q56176803

Libraries have separate listings for the works of an author and translations into various languages. The translations listings must be kept separate from the primary listings, or there will be no way to connect the listings between libraries. --EncycloPetey (talk) 00:18, 13 December 2021 (UTC)

@EncycloPetey:: OK, but the National Library of Wales identifier linked does not refer to translated works of Euripides, but identifies Euripides himself (on the linked page it states “Type of entity: Person”). The NLW identifiers are either for documents (in which case they would link to a specific document hierarchy in the NLW archives, which isn’t the case here) or for authority records (for people, families, or corporate bodies). Those authority records typically have several identifiers for each entity (person/family/body), with a specific role (e.g. translations, archives, death, family, photographs, etc). Entities who have multiple identifiers linked to them have had all but one of them deprecated (see, e.g., Royal Cambrian Academy of Art (Q7373882)) but I suspect that a better solution would be to link them all to the primary entity with a qualifier object has role (P3831) (e.g., in this instance, Euripides (Q48305) > National Library of Wales Authority ID (P2966) euripides-translations-into-english > object has role (P3831) translation (Q7553) perhaps?). Perhaps @Jason.nlw: can comment? Dogfennydd (talk) 09:37, 13 December 2021 (UTC)

Either this is like a MARC record, where the identifier is listed as "person" because it applies generally to no specific work" or there is an error in the database at their end. "translations into English" cannot logically refer to a person, because only works can be translated, not persons. The "person" label is thus either an artefact of the way the data was coded or is an error. --EncycloPetey (talk) 17:34, 13 December 2021 (UTC)

In any case, the correct identifier to match with the Euripides data item would be an identifier for the person himself, not for English translations. A Welsh library can have works in the original Greek, or translations into English, or translations into Welsh, or any other language. These translations into languages identifiers should be paired with the same labels from other libraries, not with the individual. --EncycloPetey (talk) 17:36, 13 December 2021 (UTC)

@Dogfennydd@EncycloPetey Hello both. National Library of Wales identifiers are currently a bit of a mess. In theory these are all authority id's for people, families or organisations and are tagged as such in the NLW database. However as you point out, many people have multiple id's, some of which are worded in a way that suggests they relate to specific works or aspects of a persons work or life. Ive been talking to archive staff and it seems like the authority files have just been poorly managed, so what i am currently doing is trying to clean up the authority records already on Wikidata (It wasnt the National Library who uploaded these by the way) so that we can use Wikidata to help clean up the authority records at our end. This will likely involve a lot of merges, which we will then update on Wikidata. I know this is not ideal form a Wikidata perspective but given that the id's have already been added to Wikidata by others i feel like this is the best way to proceed. I'm planning on raising this internally at NLW again in the new year in the hope that we can start making improvments. In theory this will mean that only one NLW Authority id will be needed per item. I hope that makes sense!? Happy to discuss further anytime Jason.nlw (talk) 15:37, 14 December 2021 (UTC)

Yes, that makes sense. Thanks for explaining. --EncycloPetey (talk) 19:40, 14 December 2021 (UTC)

Diolch/thanks both. Dogfennydd (talk) 20:06, 16 December 2021 (UTC)

Identifier shared with on WikitreeId

Hello @Dogfennydd.
As much as I agree that an identifier should not be shared with someone else, there are cases [1] where it happens.
For example, when historians have changed their mind (or just disagree) about 2 occurences of a name in primary sources being 1 or 2 persons.
Wikidata don't have to be aligned with Wikitree, especially as we are also linked to other Genealogical databases like ThePeerage, Rodovid, Roglo, Genealogics and WeRelate.
Personaly, when in doubt, I try to follow the same identification as in Medieval Lands project (fmg.ac).
So sadly, it happens to have identifiers shared. And marking these cases with identifier shared with (P4070) is a good way to know that they have been analyzed and are different from the cases where we have 2 items for the same person. Melderick (talk) 14:07, 11 June 2023 (UTC)

Hi @Melderick:, thanks for the message. I agree of course that Wikidata and Wikitree need not be aligned at any particular point in time, but the ideal end case envisaged by both projects is that they should be aligned (one identifier per person in Wikidata; one identifier per person in Wikitree). I also agree that tagging with identifier shared with (P4070) is a good way of highlighting that the idealised end case has not been achieved for these individuals (and flagging where further research would be beneficial).

The reason I reverted your edit is that it would stop shared identifiers from appearing in the constraint violations report, implying that Wikitree intended the identifier to refer to two different entities (it doesn't, it's an artefact based on conflicting prioritisation of sources).

There are other human (Q5) identifiers where identifier shared with (P4070) should not raise a constraint violation; the example in my mind is Dictionary of Welsh Biography ID (P1648), which mostly has one identifier for one person, but occasionally has an entry for multiple persons (see this edit of your example — and thanks for giving me the pattern for fixing the constraints for this identifier). Dogfennydd (talk) 22:02, 12 June 2023 (UTC)

Well, I do agree that it would stop shared identifiers from appearing in the constraint violations report and that's exactly the reason why I did it. Current report is way too big and it should help us (wikidata) to focus on potential problems without having to remove manually items already analyzed.

Anyway, I am not going to fight over it if you prefer to have big reports. Melderick (talk) 22:27, 12 June 2023 (UTC)