search

internetarchive / openlibrary-librarians

Coordination between the OpenLibrary.org Librarian community
18 stars 3 forks source link

Remove slurs from language categories #80

Open stopregionblocking opened 8 months ago

stopregionblocking commented 8 months ago

OL currently uses a slur for arctic indigenous peoples to refer to an entire category of languages. This should be removed/addressed.

Evidence / Screenshot (if possible)

no

Relevant url?

Steps to Reproduce

Details

https://feheleyfinearts.com/whats-the-difference-between-inuit-and-eskimo

Proposal & Constraints

Since this is being done to refer to a language group, it should be replaced with either a proper name & code for the entire group or names & codes for its constituent languages. Related issues include internetarchive/openlibrary#8127, internetarchive/openlibrary#8139 (seems to suggest that the material should be split into constituent languages), & internetarchive/openlibrary#8140.

Related files

Stakeholders

tfmorris commented 8 months ago

This is a duplicate/subset of internetarchive/openlibrary#8139, which says

Note there are also 217 records with the obsolete code esk for "Eskimo Languages" but these will need to be recataloged with the correct individual languages because there is no current equivalent. In many cases, it may be possible to infer the correct language from the subjects which are assigned (e.g. Inuktitut language)

"Slurs" seems like pretty strong phrasing (does this affect more than just esk?). The Alaska Native Language Center calls it "colonial". Of course, it should still be updated.

The good news is that, since Open Library is crowd sourced, you can do it yourself! I did a couple of examples to show how it's done:

Just be careful of things which were miscataloged to start with like:

You can get a list of the remaining 209 records using this url: https://openlibrary.org/search?q=language%3Aesk If you're a native speaker (or linguist), you should be able to knock these out in no time. Unfortunately, I'm neither.

hornc commented 4 months ago

Is there a better term to describe the range of languages that are currently grouped together under the deprecated esk code? Updating that display text might be a good first step until the proper fix, which seems necessarily quite manual, can be completed.

I'm hesitant to destroy information by simply removing the esk code and leaving the items totally un-categorised. (not that I think anyone is necessarily proposing that).

From the link above 'Inuit and Yupik languages' is suggested -- I don't know whether that is complete, or an appropriate step .

ypk (Pacific Gulf ) Yupik covers Aleut, Sugpiak and others. https://www.loc.gov/marc/languages/language_name.html#p

via https://www.loc.gov/marc/languages/language_name.html#i Inuit USE Inupiaq (ipk) Inuktitut (iku) Kalâtdlisut (kal)

There is also Aleut, ale

edit: There might be other levels of mis-categorisation in these groupings: https://openlibrary.org/books/OL25412755M/The_Acts_of_the_Apostles currently has language code ypk, but from the subject descriptions it looks to me like iku = Inuktitut would be more accurate/specific?

further edit Other candidates for a better encompassing term for the wider language group (until things can be properly assigned):

(from https://en.wikipedia.org/wiki/Talk:Eskaleut_languages)

I think Inuit-Yupik-Unangan is the most encompassing and probably best represents the range of material that might currently be grouped under the esk code.

Is this a step forward: change the text description of esk to 'Inuit-Yupik-Unangan languages'?

hornc commented 4 months ago

I've been bold and edited the language description: https://openlibrary.org/languages/esk

It's not a complete solution; the specific languages need to be coded in based on the material. Since I'm on a roll with language codes I'll take a look at seeing if there are obvious groups which can be moved together, but I think this is predominantly a manual librarian task to update the cataloging.

tfmorris commented 4 months ago

Thanks! Will autocomplete match on the code esk? If not, you might need to include something like "(Eskimo)" or "("formerly Eskimo") to help users with search. Political correctness is good, but usability is important too.

tfmorris commented 4 months ago

p.s. There are fewer than 200 volumes categorized with the old code now: https://openlibrary.org/search?q=language%3Aesk

Using subject facets may be helpful in updating the records (assuming the subjects are assigned accurately) e.g. https://openlibrary.org/search?q=language%3Aesk&subject_facet=Inuktitut+language https://openlibrary.org/search?q=language%3Aesk&subject_facet=Inuktitut+%28Langue%29 https://openlibrary.org/search?q=language%3Aesk&subject_facet=Kal%C3%A2tdlisut+dialect

A good task for an amateur linguist or ethnologist!

mekarpeles commented 3 months ago

I'm all for this being worked on by librarians, it would be great for us to both use the right language preferred by different groups, as well as spending time thinking about how we enable these items to remain findable by people who may be used to antiquated language (e.g. aliases) -- great discuss for librarian channel. Moving the librarians repo