What is the “Language” field for and why am I asked to enter it for every entity, including authors and publishers?
2 Likes
That’s the language of the entered name, as far as I understand (but that needs to be made more clear). BookBrainz uses aliases as names, rather than having a primary name, IIRC, so it’s the equivalent of the alias languages in MusicBrainz.
5 Likes
Ah, OK I see now. When I first added a creator I thought it meant the language that they wrote in. Having the first option be [Multiple languages] is confusing (IIRC this isn’t an option for MB aliases). Also, I think it would be better to make this field optional for submission.
1 Like
This is an old thread, however I think it would be helpful to obtain further clarification on this topic as there seems to be a bit of confusion.
Is the Language field on the Author editor page used to nominate the person’s primary language, or is it used to nominate every language that the respective person’s books might have been published in?
Is it really necessary to have this field at all?
1 Like
As I understood it, the language in this field describes where the actual writing of the author is used. It’s the same field you will find when adding an alias. Most latin written names are used in “Multiple languages”.
That’s interesting because I think it relates to the person’s primary language. If you are right then any name that is written in Latin alphabet is by definition a “Multiple languages”.
Hopefully, someone actually knows what it means.
I don’t think that this is correct, because the field correlates to the field in the alias section, If you were right, the language field wouldn’t make any sense.
I think we need someone else’s opinion.
If you are correct then anything that is written in Latin alphabet will be immediately classified as “Multiple languages”.
Why not just have the names of languages that use non-Latin alphabet scripts, e.g. Japanese, Chinese, Korean, Russian Cyrillic, Serbian Cyrillic, etc.? Another exception would be a German name that uses an umlaut.
You might not have heard of famous writer, Viljams Šekspīrs! 
Or sci-fi pioneer Julio Verne a.k.a. Žils Verns 
Adapting names from other Latin languages isn’t actually that rare (Latvian does it pretty much always, and other languages do it sometimes, seemingly at random!)
1 Like
I still wonder if it is necessary to tag the language.
Does a Latvian need “Viljams Šekspīrs” and “Žils Verns” to be tagged before they recognise their own language? For everyone else this is irrelevant.
Jules Verne is common to various languages; the tag “Multiple languages” is vague and doesn’t promote any understanding. If anything it raises questions.
I just think it is a waste of time.
I don’t want to flog a dead horse, but the more I think about this subject the use of language tags is not only a waste of time, but it could result in errors.
When I saw Julio Verne my first reaction was to classify it as Spanish. But on reflection it could also be Catalan, Galician, and Basque. I’m not a language scholar, so I would have to research it before I could either use the tag “Spanish” or “Multiple languages” with any certainty.
In future, I will simply use “Multiple languages” when contributing an author, unless the name is written in a language specific script. And for what it is worth as a descriptor “Multiple languages” could be renamed more accurately as “who knows?”.
Just for the record (according to Wikipedia): Catalan = Jules Verne; Galician = Jules Verne or Xulio Verne; Basque=?
1 Like
Why not just use Spanish? If it should also be the main name in Basque, then that can also be set to Basque later (or whatever) See the top section of Jules Verne - Wikidata (with all the labels) - I assume that’s the general idea this is going for.
I still contend it is a waste of time adding a language descriptor.
If the alias Julio Verne is listed on the Jules Verne BB author page then that is sufficient. The name of the language is irrelevant, as it is self-evident to anyone familiar with that language.
From the aspect of data entry it is far simpler to type in “Julio Verne”, than worrying about which language group it belongs to. I can also assure you that most contributors have misinterpreted the intended role of language.
I can’t add anymore than that, so we might have to agree to disagree.
One of the main supporting arguments for this language field on aliases is for programmatic access.
For example, the BookBrainz API that is currently in test allows anyone to query the database. If I wanted to do something with that data it would be useful to know what language each alias is in, for example to choose a display language: https://api.test.bookbrainz.org/1/author/5dbda212-4780-466d-ad23-973755e51f5b/aliases
Incidentally we could choose to do the same for the BB website itself and show an alias name (for example on hover over over a name) based on the user’s browser’s language settings.
2 Likes
I can see there is some logic in your explanation, but the system seems slightly flawed when there is the option “Multiple languages” which could mean anything.
If the majority of author names in the DB are tagged this way, IMO the search results are going to be compromised. I’m not into programming, so maybe there is a use for it.
2 Likes
You are quite correct, the [multiple languages] option should not really be present for aliases, for the reasons you put forward.
Instead, it should be reserved for Work languages (meaning a Work could have multiple primary languages). That being said, perhaps that would also be best served by adding the individual languages, and [multiple languages] should be removed altogether…
1 Like
I just want to make it clear that I am referring to the language descriptor function for Author aliases.
The “Multiple languages” descriptor is meaningless, but in the example of Jules Verne which is common to umpteen languages, without this tag you would have to add multiple language tags (possibly 100+ languages that use “Jules Vernes”): https://www.wikidata.org/wiki/Q33977
As I have already said, if there is an alias such as Julio Verne (which seems to be uniquely Spanish), then this will be readily apparent to any Spanish speaker, with or without a descriptor. For everyone else knowing that it is Spanish is irrelevant. For me, adding a language descriptor is an exercise in futility, but my perspective is from data entry and retrieval.
Work:
Unless I have been using the language field in the Work incorrectly, my understanding is that each language title has its own unique Work. In the case of a translated title that uses the same name as the original, it is standard practice to create a unique Work and add some form of disambiguation. For that reason the tag “Multiple languages” seems to be redundant for Works.
A specific work could in itself be multilingual Not that common but they certainly exist. Although in that case I’d expect if we can recognize the languages we should just set them.
3 Likes
It’s not apparent to software.
But otherwise I also find the need to add a language confusing. I think it is useful, but it should not be mandatory. In many cases a name is a name and just used as is in other languages. For different scripts it needs transliteration, but then again this transliteration can apply to multiple languages.
Making the language field optional would IMHO make editing easier and still preserve the option to set a language in cases like above for Jules Verne for alternative spellings in some languages.
3 Likes
That is a good point, but it is possible to list the actual languages individually on the Work.
I’m not a programmer and there might be a reason that I just can’t relate to. I’m happy to add the language, but in most cases the tag that I have been using is “Multiple languages”, which is the equivalent to “miscellaneous”.