Felix will be out June 20-27 & July 4 is a U.S. holiday. Do we want to meet June 20/27? if so, a convener is needed. Please reach out to Felix if you would be willing to convene.
UAT has been extended. Christine & Ryan led a review of the test cases, starting with searching and sorting diacritics.
Standard for sorting with leading diacritical marks or special characters that are not associated with the first filing character: When they are not counted as nonfiling characters, do they need to be disregarded in sorting?
Sorting & Searching with diacritics/special characters in MARC Authorities app, Inventory
Authorities: Currently, diacritics are ignored in the sort – this is what is expected by US libraries, it is acceptable in other places as well (though not always exactly correct)
Authorities: Diacritics are also ignored in search – this is also what is expected/desired
Inventory: Titles beginning with characters such as “ – should these also be ignored? NO – titles beginning, e.g., with “ are expected to sort before any other titles; that is, we expect the sort to be consistent with the index title/non-filing indicators
this expectation is not shared, some users do expect the initial quotation marks to be ignored in sorting
however, the current behavior is what happens in other systems as well
if we did remove the initial non-alpha/numeric characters from sorting, we’d want them removed from the index title as well
Felix wonders if this could be manipulated to meet varying expectations using the MARC-Instance mapping
Lynne: I think what we expect as SMEs is different from what our colleagues (outside of cataloging/metadata) expect to happen when they use the Inventory app – most people would not understand why something with an initial non alpha/numeric character was at the very top
since there is not consensus, this may need to be customizable
Christine will also look at how this is handled in other apps and will report back
Sara notes that Orders does not have default title sort, so the comparison may be difficult
Contributor sorting in Inventory: very similar to heading sorting in MARC Authorities
Sara brings up spelling differences like encyclopedia vs encyclopaedia – how are characters like the diphthong indexed? Christine will investigate this
Data Import test cases:
OCLC number (MARC 035) Normalization
ocn/ocm prefixes and leading zeros will be removed, de-duplication will occur
what about “on” prefix? will be brought back to the dev team and added to the normalization
Christie: at what point does this transformation happen? and does it apply to all 035s?
occurs during Data Import processing, on all (but only) 035s with (OCoLC) prefix
Why strip leading zeros?
typically not used, not necessarily and not used in searching, most of us are removing them in various ways already
“We remove leading zeroes because systems handle them differently when the (OCoLC) prefix is not present.” (Christie)
libraries who were not stripping them out before will end up with some OCLC numbers with and some without leading zeroes
clean-up script will be needed in those cases so that matching can occur
clean-up script should also help with duplicates created via DI over past iterations
Ryan will take this back to dev team also & keep MM posted
Jeanette: some institutions have OCLC numbers with -m, -i (for institution records) and the normalization process is treating these incorrectly; would prefer those with dash not get caught in normalization
Navigating between preceding/succeeding and parent/child for Instances with source = MARC