uwlib-cams / MARC2RDA

mapping between MARC21 and RDA-RDF
Creative Commons Zero v1.0 Universal
32 stars 2 forks source link

Mapping Spreadsheet for X00's #448

Open CECSpecialistI opened 5 months ago

CECSpecialistI commented 5 months ago

Create mapping spreadsheet for X00’s based on HeadingsFieldsPersonalNames and Personal names portion of Transforming Subject Data

@pennylenger will be responsible for creating the initial pass at this documentation, and will ask questions as needed. @CECSpecialistI will keep an eye out for issues that need attention from @GordonDunsire or Deborah, unless they get to them first :) @cspayne will use the spreadsheet to write transformation code for the relevant fields, using Personal names portion of Transforming Subject Data for additional context as needed.

pennylenger commented 4 months ago

@tmqdeborah Hi Deborah, I have some questions about HeadingsFieldsPersonalNames and Family name in terms of the format. How can I make them clear by mapping individually or combination of subfields in the google sheet. If you get time, could you please look at following questions? @GordonDunsire hi Gordon, I have added you subject part to the family name tab of Deborah's original google sheet. I didn't separate them into different rows for one heading, because I think they should be put together. Could you please take a look at it to see whether it works or how I should separate them to fit into Deborah's format?

For x00 $c-Titles and words associated with a name is used for occupation and other distinguishing characteristics, not just affiliation, so needs different mapping – it is a catch-all

For personal names, it has been mapped to authorized access point for person. Should we try to use some more specific element based on different conditions? If yes, how can we write the conditions?

For family names, it has been mapped to hereditary term of honour and profession or occupation. How should we write conditions to make the mapping right or are there any other appropriate elements?

For the personal names tab: Should we list a subfield or combination of different subfields and their mapped RDA elements clearly in each row to avoid confusion and redundancy? For example: for $q, $c, $d, the note says include values for other elements or designations in an access point, if required]. Since these subfields need to be mapped together with other subfields as authorized access point for person, not individually, just like $a,$b,$c,$d,$g,$j,$q,$u -> Authorized access point for person, I think that they should not be listed individually. It may be confusing.

List them individually in a row unless it can be mapped individually to a more specific element, for example: $q - Fuller form of name (NR)->Fuller form of name; or list them in a row with the the combination of different subfields if they can be mapped to other elements, for example: $a$b->Name of person $a$b is also mapped together to Authorized access point for person; and $a individually mapped to Authorized access point for person. Is it redundant as $a,$b,$c,$d,$g,$j,$q,$u has been mapped to Authorized access point for person.

For $q, the note also says to include values for other elements or designations in an access point, which I think is mapped as an AAP together with other subfields. The RDA element in this field is Fuller form of name.

$t and $k are put together, no RDA element in this row, in the notes it says Map as part of title, so is the element for $t$k title of work? And how to map as part of ?

$n is mapped to several elements according to different conditions What does it mean by “map as part a AP” ? Is the element used access point for person? Then what subfields are other parts of the AP? If subfields are listed, it will be clear.

For $p, is the element for $t$n or $t$p$n title of work? and they should be mapped together, no individual mapping for $p ?

For Gordon's documentation. We had a discussion on 5.15 meeting and below it some points.

For x00 first indicator = 3-Family name Family name AAP will be $a$c$d$g In a MARC access point, there would never be $u or $j LCSH and Name Authority File have different approaches to family name

And other people think that $y$z aren’t topical subdivisions, but they are subject subdivisions. And there are still some controversy over $v. Concept (exists(600, 610, 611. 648, 651 $x, $y, $z))

rdawo:P10256 . // has subject; topical subdivisions skos:prefLabel “stripEndStop(concatenateOA($x, $y, $z))” . skos:inScheme .
tmqdeborah commented 4 months ago

Hi @pennylenger, Will you be at the "Optional MARC2RDA Meeting" tomorrow? (https://washington.zoom.us/j/95925664112)

We will be discussing creating a separate Access Points table, which will (hopefully) clarify most of your questions about which subfields are used for access points.

The purpose of the "HeadingsFieldsPersonalNames" table was to get the discussion about the fields and subfields started. So, it combines the access points and the attributes that we can map from those fields and subfields, along with links to format guidance, explanations and questions.

I think we can make things much more understandable for coding if we deal with mapping access points and attributes separately.

Hope to see you in the meeting.

pennylenger commented 3 months ago

Deborah has created a new form Access Point Mapping Table.20240528

pennylenger commented 3 weeks ago

@AdamSchiff Is it possible for subfields of title portions to appear in the 100/110/111 although it is valid in MARC21? do we need to map the title portion subfields in 100/110/111?

Title portion $f - Date of a work (NR) $k - Form subheading (R) $l - Language of a work (NR) $n - Number of part/section of a work (R) $p - Name of part/section of a work (R) $t - Title of a work (NR)

tmqdeborah commented 3 weeks ago

Is it possible for subfields of title portions to appear in the 100/110/111 although it is valid in MARC21? do we need to map the title portion subfields in 100/110/111?

Title portion subfields should NOT appear in 100/110/111 fields, but they do, very occasionally, in error. I found 209 records in the LC file of 5.6million records) with $t (far fewer also with other title portion subfields). I would consider these to be errors and would map the Name portion from the 100/110/111 fields and ignore those Title portion subfields in mapping (see my Access Points table).

Examples are attached. 110t.txt

Otherwise, you could treat those Title portion subfields as though they are 240 subfields.

AdamSchiff commented 3 weeks ago

Some of those $t's are typos and should be $b (Great Britain. Parliament. House of Lords). They've been corrected in OCLC. If LC had an authority vendor, they'd probably have been fixed in their database too.

Maybe we need to report these errors to them.

Adam

Adam L. Schiff Principal Cataloger University of Washington Libraries (206) 543-8409 @.***


From: Deborah Fritz @.> Sent: Wednesday, September 11, 2024 12:49 AM To: uwlib-cams/MARC2RDA @.> Cc: Adam L Schiff @.>; Mention @.> Subject: Re: [uwlib-cams/MARC2RDA] Mapping Spreadsheet for X00's (Issue #448)

Is it possible for subfields of title portions to appear in the 100/110/111 although it is valid in MARC21? do we need to map the title portion subfields in 100/110/111?

Title portion subfields should NOT appear in 100/110/111 fields, but they do, very occasionally, in error. I found 209 records in the LC file of 5.6million records) with $t (far fewer also with other title portion subfields). I would consider these to be errors and would map the Name portion from the 100/110/111 fields and ignore those Title portion subfields in mapping (see my Access Points tablehttps://urldefense.com/v3/__https://docs.google.com/spreadsheets/d/1CXBDgFvXzFiRytEz8G-9dkax8oyEA3q_/edit?gid=610540737*gid=610540737*5D__;IyU!!K-Hz7m0Vt54!kjJDhrfDpiGr1XMKasmrAfLi40qYNrdFBhICzhZ1unj55uHhDlOL0wMaZYCeIIPO2ssgQK0wczsb8FyoFFzOit4$).

Examples are attached. 110t.txthttps://urldefense.com/v3/__https://github.com/user-attachments/files/16959872/110t.txt__;!!K-Hz7m0Vt54!kjJDhrfDpiGr1XMKasmrAfLi40qYNrdFBhICzhZ1unj55uHhDlOL0wMaZYCeIIPO2ssgQK0wczsb8FyoFiLbDRY$

Otherwise, you could treat those Title portion subfields as though they are 240 subfields.

— Reply to this email directly, view it on GitHubhttps://urldefense.com/v3/__https://github.com/uwlib-cams/MARC2RDA/issues/448*issuecomment-2342908414__;Iw!!K-Hz7m0Vt54!kjJDhrfDpiGr1XMKasmrAfLi40qYNrdFBhICzhZ1unj55uHhDlOL0wMaZYCeIIPO2ssgQK0wczsb8FyoZM96eRI$, or unsubscribehttps://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/ADFBVB3OVS4JXSCPYHSTDBTZV7YXFAVCNFSM6AAAAABGVD7T56VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNBSHEYDQNBRGQ__;!!K-Hz7m0Vt54!kjJDhrfDpiGr1XMKasmrAfLi40qYNrdFBhICzhZ1unj55uHhDlOL0wMaZYCeIIPO2ssgQK0wczsb8Fyoz981TMk$. You are receiving this because you were mentioned.Message ID: @.***>

cspayne commented 3 weeks ago

Hi @pennylenger Would it be best for me to work from the individual mapping sheets for 100, 600, and 700, or should I work from the HeadingsFieldsPersonalNames sheet? I know that there is a separate sheet now for access points, but I'm not sure where to look for attributes. If they are the same across the X00s, it may be easiest to combine them into one sheet, but just let me know! Thanks!

pennylenger commented 2 weeks ago

@cspayne Hi Cypress, I can put the attributes in one table, which will be more clear. I'll first post the attributes mapping the team discussed earlier here for @tmqdeborah @AdamSchiff @GordonDunsire @CECSpecialistI to review and see if these mapping are correct now.

X00 - Personal Names-General Information (Attribute Mapping)

Name portion $a - Personal name (NR) $b - Numeration (NR)

$c - Titles and words associated with a name ®

$d - Dates associated with a name (NR)

$e - Relator term (R) $j - Attribution qualifier Not mapped

$q - Fuller form of name (NR)

$u - Affiliation (NR)

$4 - Relationship (R)

Title portion $f - Date of a work (NR)

$h - Medium (NR) [600/700/800]

$i - Relationship information [700] (R)

$k - Form subheading ®

$l - Language of a work (NR)

$m - Medium of performance for music (R) [600/700/800]

$n - Number of part/section of a work ®

$o - Arranged statement for music (NR) [600/700/800]

$p - Name of part/section of a work ®

$r - Key for music (NR) [600/700/800]

$s - Version (R) [600/700/800]

$t - Title of a work (NR)

$v - Volume/sequential designation (NR) [800]

$x - International Standard Serial Number (NR) [700/800]

pennylenger commented 2 weeks ago

Hi Cypress @cspayne, I make this sheet Heading Fields Attribute Mapping. I hope it is helpful. Some mappings may need revision. Please let me know if you have questions.

cspayne commented 2 weeks ago

This is great! Thank you Penny!