2018-02-15 Metadata Management Meeting Notes

(no update today)

Targeted search and advanced search feature discussion & UX initial mockups

• https://discuss.folio.org/t/inventory-app-search-and-filter-user-feedback-on-current-version-alpha/1562
• https://discuss.folio.org/t/inventory-app-display-of-instance-data-in-detailed-view-user-feedback-on-current-version-alpha/1563

Discuss requirements for Searching and filters:

https://docs.google.com/spreadsheets/d/1ubVlAvZ8PKgKtQ7sOqpmqR3T2xidP0uYJdP9UzRJsbk/edit#gid=952741439

Reviewing the spreadsheet with various features. See additional notes in the spreadsheet itself.

1. General comment: if library does not want some of the data elements included in a search, allow for preferences to include/exclude various data elements in each search (index). Think of these choices as defaults, not fixed.
2. Perhaps highlight search term in the results, especially if search is an alternate title, and not found quickly.
3. Bib ID number: We want to be able to search on bib ID number. Not clear yet if we need a shorter, more eye-readable version of the ID number as well.
4. Bib ID number: Is it useful to be able to browse by a partial match on bib ID, or only search ID number on exact match? For now, only search on exact match.
5. Metadata source: is the record MARC or Dublin Core or created in the inventory? Not needed (yet)
6. Title: Yes, and include title, alternative title, uniform titles, related titles, and series title
7. Series title: only search series title
8. Identifier: maybe only have general identifier in the basic search, and individual identifiers in the advanced search? Maybe basic search is by all, and then filter by specific identifiers. Be sure to include ISMN as well.
9. Identifier Note: Searches for numbers need to be normalizing their numbers (e.g. removing punctuation or spaces in ISBNs). Also applies to things like GPO numbers or Technical Report ID numbers. Questions about how to normalize the incoming data and how to make sure teh searches are forgiving of entering variously-formatted numbers.
10. Contributor: any people names or organization names
11. Personal name, Corporate Name, Meeting name - what order should they come up? For now, Personal, then Corporate, then Meeting. Ultimately, allow setting as personal preferences or personal default.
12. Subject search: is it needed for basic, or only advanced? It wouldn't hurt to include in basic, but doesn't seem super-important.
13. Note: We will have a search all data elements box, as well as being able to limit to specific tyeps data elements.
14. Classification: definitely needed, search by the beginning of the number, not exact search. Important to normalize spaces and punctuation (especially for SUDOC numbers). Would like correct sort order (which is not as easy as it sounds) and ideally being able to search by a range. Classification searches are always in the context of the classification type.

From Filip Jakobsen to Everyone: 03:16 PM
https://discuss.folio.org/t/robot-1234-sequential-number-generator/1450

From Duke People: Jacquie, Dennis, Sarah to Everyone: 03:20 PM
It is our opinion that no one wants to browse by system number/ system key.

From A&M Group to Everyone: 03:28 PM
Shouldn't 'title' search pick up on both? Also, why would 'related title' also not be picked up in the title search? And what about uniform titles? Another question. Ditto: why can't the local institution set their customized indexing?

From Patty Wanninger to Everyone: 03:32 PM
Sorry, I thought I was muted. ;-o

From A&M Group to Everyone: 03:35 PM
Yes/!!!!!! TAMU: All of them. Could ID search hit on all of them and then you filter by type>

From Duke People: Jacquie, Dennis, Sarah to Everyone: 03:36 PM
That sounds better, except they may be held in multiple fields, right? Also, don't forget the ISMN.

From A&M Group to Everyone: 03:36 PM
TAMU: ones most frequently used rise to the top?

From Wayne Schneider to Everyone: 03:37 PM
@Duke — in the inventory, all identifiers are part of the “identifiers” property of the instance…they may come from multiple fields in MARC, but they end up in the same place. Each identifier in the list then has a type (ISBN, ISMN, etc.)

From Duke People: Jacquie, Dennis, Sarah to Everyone: 03:39 PM
OK, that sounds good Wayne. I am concerned about the definition of "standard" numbers in the list, since some are not. Also, identifiers are held in notes fields, etc., so the mapping will be interesting.

From A&M Group to Everyone: 03:39 PM
TAMU: And we've seen systems where isbn/issn is combined in the same search.

From Duke People: Jacquie, Dennis, Sarah to Everyone: 03:50 PM
Won't some staff always want to use Advanced search?

From Laura Wright to Everyone: 03:51 PM
I agree with Duke above.

From A&M Group to Everyone: 03:52 PM
Could you also do the same search in the codex? That's what we think: but yhen I'd be doing that in the catalog module or oclc.

From Duke People: Jacquie, Dennis, Sarah to Everyone: 03:58 PM
I think it is a weird way to search for processing library resources. And if used as keyword, it could provide overwhelming results.

From A&M Group to Everyone: 04:00 PM
NO! I need classification for doing holdings work TAMU: what we meant is that it's not the same as subject. We need classification especially for serials that have the same class and I want to get all of the records.

From Duke People: Jacquie, Dennis, Sarah to Everyone: 04:02 PM
LC sort is a fixed problem. Others have tackled and coded fixes.

From A&M Group to Everyone: 04:04 PM
bye! Thanks!

From Duke People: Jacquie, Dennis, Sarah to Everyone: 04:04 PM
Thanks Charlotte!!