Display of non-Roman titles in results lists

Reported to me by David Sulz, this issue is about which MARC title fields are displayed for non-Roman script items in results lists (we think that 880 may be a field that needs to be displayed). Currently, we see the transliterated title, the translated title, but not the original title. David would like the original title to display as well, as this would aid in discovery and disambiguation.

Here is David's email:

"Our high-level search results should show original language characters for title as well as transcription (and translation if available).

The first example from Harvard library below shows an ideal example, I think. Original character title with translation and transcription with translation.

This is important for several reasons and groups.

if someone copy/pastes a CJK title into a search, they may not be able to match the transliteration we currently show with the item they are looking for. (e.g. search for 宮沢賢治の世界 )
when CJK searchers enter words in CJK, they use English letters on their keyboard
even native Japanese speakers, for example, may use the transliteration to figure out which reading some characters might have.
for non-native CJK patrons, having a transliteration and translation makes their searches much more efficient because they can hone in on more relevant items more quickly without the effort to translate all the search results.

Example search in Harvard library http://hollis.harvard.edu/primo_library/libweb/action/search.do?ct=facet&fctN=facet_rtype&fctV=books&rfnGrp=1&rfnGrpCounter=1&pcAvailabiltyMode=true&query=any%2Ccontains%2Cjapanese%20dictionaries&indx=1&fn=search&vl(51615747UI0)=any&vl(1UI0)=contains&dscnt=0&search_scope=everything&scp.scps=scope%3A(HVD_FGDC)%2Cscope%3A(HVD)%2Cscope%3A(HVD_VIA)%2Cprimo_central_multiple_fe&displayMode=full&mode=Basic&vid=HVD&onCampus=false&ct=search&institution=HVD&bulkSize=30&highlight=true&tab=everything&displayField=all&vl(freeText0)=japanese%20dictionaries&dstmp=1483737983419&fromDL

Another Harvard example: Original + transliteration at least. Translation when available.

Catalogue record for the first one: Incidentally, they also have tab for virtual shelf browsing.

http://hollis.harvard.edu/primo_library/libweb/action/display.do?tabs=detailsTab&ct=display&fn=search&doc=HVD_ALEPH010107179&indx=17&recIds=HVD_ALEPH010107179&recIdxs=16&elementId=16&renderMode=poppedOut&displayMode=full&frbrVersion=&frbg=&&vl(51615747UI0)=any&vl(1UI0)=contains&dscnt=0&scp.scps=scope%3A%28HVD_FGDC%29%2Cscope%3A%28HVD%29%2Cscope%3A%28HVD_VIA%29%2Cprimo_central_multiple_fe&tb=t&mode=Basic&vid=HVD&tab=everything&srt=rank&vl(394521272UI1)=all_items&vl(freeText0)=%E5%AE%AE%E6%B2%A2%E8%B3%A2%E6%B2%BB&dum=true&dstmp=1483736608520

The Library of Congress also has transliterations in their search results: "

Any thoughts on this from a cataloguing perspective @theLinkResolver? @redlibrarian are there any significant issues with achieving this? @ualbertalib/discovery-testing are there other cases (e.g. languages) where this is also an issue (@linds37)? I immediately think that available real estate and wrapping are potential issues.

It would be ideal to include both vernacular and transliteration in the display whenever possible. I agree that it is confusing to search in the vernacular, then get presented with a list of book titles in transliteration only, while the article titles do appear in the vernacular. For example, if you search for the Russian word книга, you get a list of 28 transliterated titles on the books & more list, while the articles display gives 1000s of titles in Cyrillic. When you go to a book record, you can see the Cyrillic title in the "original" field, eg: https://www.library.ualberta.ca/catalog/7041267. Only 28 of our catalogue records actually have the Cyrillic word книга in them (it means "book", so that's not very many at all). As you know, this is always going to be a confusing issue for library users, because historically most of the records are transliterated, not in the vernacular. When searching in our discovery tools, transliteration must be used for a comprehensive search. If you search for the transliterated word kniga, then you get 1200 results on the books & more list. The ideal would be to be able to choose an option that does automatic transliteration between scripts in a specific language : )

I believe a translation of the title will only be included if the text itself actually includes a translation or if the author titled it with a parallel title -- a cataloguer should correct me if I'm wrong. I don't think English parallel titles should be included if no translation is actually available. This would be confusing, giving the impression that a title may exist in English translation when it doesn't. Some journals do that, and only a translated abstract of the article is actually available. It's annoying.

ualbertalib / discovery

Display of non-Roman titles in results lists #1095