search

uwlib-cams / MARC2RDA

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

336 content type #146

Open CECSpecialistI opened 2 years ago

CECSpecialistI commented 2 years ago

https://github.com/uwlib-cams/MARC2RDA/blob/main/Working%20Documents/3XX.csv

CECSpecialistI commented 1 year ago

Decided during 2023-05-03 meeting we should examine values and re-evaluate mapping output to use RDA values whenever possible, and create map/alignment from RDA VES to id.loc VES and send to RSC to see if they will add to RDA Registry. @szapoun will post National Library of Greece alignments, @GordonDunsire will look at it

szapoun commented 1 year ago

NLG mapping for 336 336.xlsx

GordonDunsire commented 1 year ago

I have checked the NLG alignment and don't see any errors.

For the purposes of an RDA to MARC 21 mapping, RDA prefers to use SKOS to relate RDA concepts, which are already expressed in SKOS, to external concepts.

The LC concepts are also expressed in SKOS, albeit alongside a MADS representation.

In all cases, I think the correct alignment and mapping is SKOS "close match".

The LC vocabulary and concepts appear to be a near copy of the RDA vocabulary, although it has additional, non-RDA concepts.

However, LC does not provide definitions for any of the content types, so this may be a risky assumption.

[This should be queried with NDMSO. For example, there is no definition for m21's "other", so how is it to be applied? I think this is a quality control (aka "MARC Standards") issue.]

The RDA concept "performed movement" is missing from the LC vocabulary, and cannot be mapped.

RDACO2M21NT.txt

RDACO2M21NT is an n-triples version of the map (in text format for GitHub purposes). The NT format is suitable for submitting the map to the RSC Technical Working Group for approval and publication in the RDA Registry.

Note: skos:closeMatch is effectively a symmetric property, so the map can be flipped and used to reverse-engineer M21 data that uses the M21 vocabulary.

The NT file is extracted from an Excel spreadsheet; see the discussion on Maps from RDA value vocabularies to M21

pennylenger commented 4 weeks ago

Do we have a way to map $3 in the 336 field? It is expression level. We have discussed this but haven't reached a decision. As well as for 335 work level.

GordonDunsire commented 3 weeks ago

@pennylenger: subfield $3 indicates that the manifestation being described is either an aggregate that embodies multiple manifestations, or a multipart manifestation that has multiple sub-units/parts. I'm not sure that we have a reliable test for differentiating these cases, so a decision remains pending ...