Everett and all,
NLM has been very supportive of this proposal for two reasons. First of all, historically we have been coding 600, 610 and 630 fields in our records with indicator “2” (MeSH), but in fact we only use MeSH terms for the 650 fields. Our names and title subjects are established as found in the NAF. However, we still follow NLM practices, in that we would not add subfields $x, $v or $z to a 600, 610 or 630, whereas libraries that follow LC practice may do so. So it might be useful to NYU to identify these fields, if only to see if they might need modification in your own files.
Also, we believe that as we move more fully into a linked data environment we are going to want to know the provenance of various pieces of data to help us decide if we want to incorporate it into our local databases. Libraries will probably trust LCSH assigned by LC or Harvard more than they would a small public library or MeSH assigned by NLM over a small hospital library. By beginning to add the $5 now to subjects, we make our records more linked data friendly.
Head, Cataloging and Metadata Management Section
National Library of Medicine
8600 Rockville Pike
Bethesda, MD 20894
Dear Reinhold and All :
Hi -- I'm sorry, but I think I am having trouble with this paper and the examples cited as well.
As Sherman indicated in his initial message, "The use of $5 is used to mark certain portions of the record as having particular relevance to one institution and not necessarily to other institutions."
Within the example in the Discussion Paper and in Reinhold's additional example, it appears to me as though each of these 6XX fields are applicable to the "of-ness" and/or "about-ness" of the described resource. Therefore, wouldn't we want these 6XX subject tracings to remain as part of the shared resource description? That is, how or why would any of these 6XX tracings be of "particular relevance to one institution but not necessarily to others?"
Perhaps I'm focusing too much upon the historical usage of $5 by libraries to designate specific provenance information, often for a particular Bibliographic Item. In the Discussion portion of 2017-DP05, there appears to be an intent or interest in using subfield $5 within the 6XX block of fields almost as a "tagging" mechanism for two distinct purposes: 1) policy level information (i.e., "Organization X has cataloged this field according to its own rules and regulations"), and 2) to designate institutional provenance information ("Organization X has added this field to this MARC record.")
Is that correct? If so, perhaps a bit more explanation and some additional examples would be helpful. For myself, I'm having trouble envisioning how the need to record either of these two data attributes would effect the "of-ness" or "about-ness" (i.e., the 6XX fields) of the shared or universal resource description.
Finally, there is another aspect of this paper that concerns me. For the last few years, New York University (NYU) has been running some validation and Bib-enrichment/enhancement
trials upon Import and Export of records into our Ex Libris Aleph ILMS. Among the recent "clean-up" sweeps within our internal database, was the removal of hundreds, more likely thousands, of non-NYU Bib descriptive fields and 7XX access points.
These descriptive elements and access points that are NOT contained within NYU's collections were identified by scanning the database for Bib fields containing subfield $5s that did NOT include one of New York University's MARC Org Codes (or one of our approved affiliates, projects, programs, etc.).
One of my concerns with 2017-DP05 is that if we venture down this road of designating fields with subfield $5 even in cases where the data in those descriptive fields and access points applies universally to the shared resource description, those fields and access points may then be inadvertently stripped from the resource description for all the wrong reasons.
For instance, in NYU's recent Bib sweep searching out records with fields containing non-NYU subfield $5 MARC Org Codes, most or all of the 6XX fields in Reinhold's two examples would have been stripped from the descriptions because they contain $5 DE-101 or $5 DE-91. This "Bib enhancement" sweep would have then left NYU's Aleph Bib resource description in this case with virtually no Subject Access for Users. 😞😧!
Now certainly, NYU can re-write the coding and scripts we have been using to scan our Bib database to identify non-NYU descriptive and access data. However, before we need to do so it would be helpful to have a clearer understanding of the reasons why. And of the added value for defining subfield $5 (Institution to which field applies) information to the 6XX block of MARC Bib fields.
thanks very much for any further clarification. Apologies for the ramble (smile!)
best to all,
On Mon, Dec 19, 2016 at 4:45 AM, Heuvelmann, Reinhold <[log in to unmask]> wrote:
thank you very much for your comments on the discussion paper, and for your agreement to continuing the discussion on the open listserv.
I can provide an example, available at
which is based on a real record online at
Here we have 650 fields with different contents in the subfields $5 (to be defined), among them two subject headings provided by the German National Library (DE-101) according to its rules, and three subject headings provided by the “Technische Universität München, Universitätsbibliothek” (DE-91) according to different rules (or at least different interpretations of the same rules).
The fields 655 with the form heading “Zeitschrift” (“serial”) are all the same, each of them is the second part of a sequence of headings. These sequences of headings are transported in the local field 689, mostly redundantly to the other 6XX fields, but controlling the grouping and sequencing by the two indicator positions.
By the way: The URI for the German National Library is
and the URI for the Technische Universität München, Universitätsbibliothek is
It would not be easy to find an example where important 6XX fields are missing institution information in subfield $5, because this metadata provenance information is mostly mandatory here, at least in the context of the Union Catalog for Serials (Zeitschriftendatenbank, ZDB).
I hope this helps.
German National Library
Office for Data Formats
60322 Frankfurt am Main
Telephone: +49 69 1525-1709
*** Reading. Listening. Understanding. German National Library ***
The use of $5 is used to mark certain portions of the record as having particular relevance to one institution and not necessarily to other institutions. The example in the Discussion Paper has all of the 6XXs (except 653) marked with $5 DE-101 and this agency is the cataloging agency. In effect, the whole record was presumably prepared by DE-101. I am not sure why the 6XXs that come from the originating cataloging agency would need to be marked with $5.
Would it be possible to provide an example where only some of the 6XXs are institution-specific? Something along the lines of a record with some but not all 5XXs marked with $5.
VRA liaison to MARC Advisory Committee