Zoom Logo

Bibliographic Conceptual Models (Core IG Week) - Shared screen with speaker view
Julie Reese, Core Staff, she/her/hers
25:39
Hi, everyone - I'm a Core staff member. Just a quick note that our Code of Conduct can be found at http://www.ala.org/core/about/statement-of-conduct . Please contact me through private chat or at jreese@ala.org if you experience or see any violations of the Code.
Julie Reese, Core Staff, she/her/hers
30:15
If you’re not already a member of this interest group, you can join it at https://alacore.org/go/to/mcbcmig
Jesse Lambertson
31:55
Please feel free to post any questions you have for Jim in the chat at any time. We will be asking these questions toward the end. thank you
Stephen Hearn
32:44
Were 7XX added entry fields checked for controlled forms of a publisher name on some records?
Kevin Randall
34:19
Autocomplete for *transcribed* data is a very bad idea. It is currently being used in Ex Libris Alma, and it is a huge pain.
Susan Moore
34:46
Agree with Kevin on his point.
Jim Hahn
36:42
https://upenn.box.com/s/5gx69ru6qm81khpe3sb1ealaa2p0sjl5
Beth Guay
36:48
+1 more agree with Kevin
Thomas Dousa
37:08
Could tell us a bit more about the Fp growth algorithm and how it works?
Stephen Hearn
39:03
Can this sort of analysis give us a best choice for transcribed data when transcriptions vary, as they do under older rules? especially for transcribed fields like 26X?
Beth Picknally Camden
40:46
Is it of value to continue to transcribe publication information? Creating this as linked data can be much more useful to users.
Kevin Randall
41:57
@Beth: The transcribed data and normalized/authorized data serve different purposes. Sometimes one is useful, sometimes the other. And that can't always be predicted.
Amanda Sprochi
42:37
I think we need to remember that one function of transcription is to help a cataloger identify if the piece they have in hand is the resource described by the record you are looking at. We might examine what needs to be transcribed exactly (or in the predictable pattern using rules on capitalization etc) so they piece in hand can be matched to the correct record. With modern publishing this is less of an issue, but it's crucial for rare and special items.
Anchalee Panigabutra-Roberts
43:15
+1 for rare materials
Kevin Randall
43:39
Who knows what is going to be rare or special in the future?
Paloma Graciani Picardo
46:33
+1 Amanda, Joy, Kevin
Adam Schiff
47:03
Wikidata allows you to do both: for place of publication you record a linked data value but can qualify it with the property "stated as" to record what it actually says on a manifestation. E.g. Place of publication = New York City [linked data identifier] with qualifier Stated as = New York.
Kevin Randall
47:10
An autocomplete-like function that would suggest *access points* as *added elements* when inputting a transcribed element may be quite useful.
Laura Daniels
47:28
+1 Kevin
Beth Guay
47:34
Transcription ensures trust in the data
Kevin Randall
48:27
+1 Adam
Anchalee Panigabutra-Roberts
49:01
I guess we need both collocation and description to I.d. the materials. Balancing the two needs - accommodating both needs would be ideal.
Adam Schiff
50:04
We could do the same in MARC by splitting up the 26X fields.
Beth Picknally Camden
50:10
The PCC Sinopia templates currently have both.
Adam Schiff
50:22
264 $a New York $0 URI for authority for NY
Stephen Hearn
50:31
How does this analysis deal with very abbreviated transcriptions in 260, e.g. "the Association" when an association name is given in full in a controlled field?
Adam Schiff
50:35
Of course in MARC we'd need to add $0/$1 to 26X fields
Thomas Dousa
53:01
Is there any way to automatically deal with publishers’ names when they occur in different languages (as sometimes happens with European presses publishing in English as well as the local vernacular)?
Bela Gupta
54:37
How do you deal with different language transcription for the same place. The transcription in MARC would be different but it is the same place. Same goes for publisher. 260 field depends on as found in source. So it could be confusing for data mining and control.
Adam Schiff
55:21
That was former practice in AACR2 when the name of the publisher was in the title
Adam Schiff
55:35
or in the statement of responsibility
Anchalee Panigabutra-Roberts
56:42
+1 Adam for the reminder.
Adam Schiff
59:51
Example: 245 00 American Society of XXX membership directory. 260 $a New York : $b The Society, $c 1996.
Jesse Lambertson
01:00:11
thank you Adam, that is helpful
Paloma Graciani Picardo
01:00:16
@Adam - thanks for the example
Jim Hahn
01:06:05
https://www.kaggle.com/sajidcse/market-basket-analysis
Thomas Dousa
01:06:09
Could you speak a bit more about the advantages of the set-theoretical model in a linked data context?
Thomas Dousa
01:08:40
Great answer!
Julie Reese, Core Staff, she/her/hers
01:08:48
Consider attending the Core Forum, October 7-9, 2021 in Baltimore. Registration is open. https:/coreforum.org
Julie Reese, Core Staff, she/her/hers
01:09:29
If you’re not already a member of this interest group, you can join it at https://alacore.org/go/to/mcbcmig
Bela Gupta
01:09:37
Great presentation Jim. Good luck with your data work!
Jim Hahn
01:09:51
Thanks Bela, good to hear from you! :)
Jesse Lambertson
01:10:59
thank you so much Jim!!
Jim Hahn
01:11:10
Thank you Tom! :)
Jim Hahn
01:11:14
Thanks Jesse !
Julie Reese, Core Staff, she/her/hers
01:11:24
Thanks you! I'm happy to help.
Tiziana Possemato
01:11:39
Thanks Jim, your presentation was great and really interesting!
Anchalee Panigabutra-Roberts
01:11:41
Many thanks, Jim and colleagues for the discussion too.
Paloma Graciani Picardo
01:11:58
Great presentation, Jim! And thanks everybody!
Jesse Lambertson
01:12:00
cheers all, thank you everyone