2022 AskQC office hour member Q&A

February 8, 2022

Is code "a" in CrTp used for a single sheet that contains more than one map on it (one on each side of the sheet, for instance)?

It's for a single map on a single sheet. If there is more than one map on a sheet, that is typically recorded in the 300, like you would see "five maps on one sheet." However, if you didn't have that situation, of a single map on a sheet, you would still have to pick a code. It doesn't fall into any of the other types, so you end up with type "a" sort of by default, when you have that situation of multiple maps on one sheet.

CONSER guidelines say that 008/33 "should be" blank if there is no authenticated Key title. OCLC guidelines say that the code may be based on language of title proper (245) if there is no Key title. When authenticating a record for CONSER, are we supposed to remove a code from 008/33 if there is no Key title?

That instruction that we have in documentation has been there since the 1980s. I personally have always considered that an ISSN element that I would typically not bother with in ordinary cataloging. For CONSER, since there is a guideline that says that it should be blank, if you are authenticating an existing record where that is coded and you end up with an authenticated CONSER record where there is no Key title, follow the CONSER instruction and change the element to be blank so it's in step with CONSER practice.

When using a code like "k" (vocal score) or "c" (accompaniment reduced for piano) for Format of Music, should "b" (arrangement) also be used for Transposition and arrangement?

Both of those elements are optional. You're welcome to leave both of them as they are or code either or both. It's entirely up to you.

If illustrative content (008/18-21; 006/01-04) has a lot of importance to a library that will be starting to use WMS soon, would you recommend them to continue to enhance illustrative content in this shared environment?

Yes, the illustrative content is important to everybody using the records. So if your library is in the process of making sure you thoroughly code those elements, that is a great idea. Everybody who uses WorldCat will benefit from that coding.

Electronic version of a loose-leaf (updating) monograph: the SrTp for the print is coded "l", but the electronic version normally presents the resource with the pages interfiled and updated, so it is an updated resource as a single resource. Should we retain the "l" in the SrTp? Or, because the vendors only provide the "current" and fully updated resource, should these electronic resources be coded "w" instead? This is a common situation with updating law materials.

It seems to me that "l" is really intended for the print situation where you're receiving updates and you're filing those in. The situation where you have that same publication presented as an online resource, where basically they replace it with the updated version, seems like it's much more the situation where you would want to code it as "w," to say it's more like an updating website in that case. But that's just my take on it at this point. There must be other records out there that people have coded that tend toward a certain practice. There may be documentation that addresses this. But that's what comes to mind.

The definition for "loose-leaf" in this position, the code "l," is defined as a tangible resource. It's a base volume updated by separate pages, which are inserted or removed and substituted and that's very print-oriented. Generally, you wouldn't use the code "l" for an electronic resource. Unless there's another code in this position that's more appropriate, I would generally use "w" as website, because that's the remote resource equivalent of an updating loose-leaf.

When you say a code is not indexed, I assume you mean not indexed in the OCLC database. How are these codes used in our ILS or LSP systems? Does this usually correlate with the Mandatory codes?

Yes, we are discussing indexing in the WorldCat database, and our information is coming from the document Searching WorldCat Indexes, which is available on the OCLC website. Indexing is unique to each system, so if you want to know what is indexed in your own integrated library system or local system, you'll want to check with your system documentation.

Could you comment on fixed fields for Galleys, Advance Reader Copies, Uncorrected Proofs?

There really isn't anything different in the fixed fields for galleys, advanced reader copies, or uncorrected proofs from the published version. The differentiation for those would be in the variable fields, in the fields where you record the bibliographic information. You would want to put an edition statement in. It's really important to include an edition statement, either from the resource itself or to bring out that characteristic in a supplied edition statement in the 250.

I asked about Galleys, etc. because I have noticed that records our institution has previously cataloged with the supplied edition statements have been later merged with the final published edition of the book. Shouldn’t these be on separate records, since there may be variances from the final published edition?

Pre-publication versions of a resource are definitely separate and should be separate. Make sure that information is presented as an edition statement in field 250. We try to look for certain words and phrases in quoted notes, which has been a past practice for these sorts of materials, but we would urge you to start using field 250 for that information, regardless of whether it's on a cover, stamped somewhere, or just the information that you happen to know, that the item you have in hand is some kind of pre-publication version. Use the information that's on the resource as a field 250, or provide a cataloger-supplied edition statement for these pre-publication items. Even if it's in brackets, that should keep DDR from incorrectly merging a version with a field 250 with one without a field 250. If you have specific examples of incorrect merges, please let us know, give us the OCLC numbers and we can pull them apart and try to make sure that doesn't happen again. And we apologize for the extra work.

Is there a way to search WorldCat for items labeled as open access in the 856 $7? I do not think this subfield is indexed at this time.

In Connexion, this element is not yet indexed. It is something we would like to index in the future; it's on our list. I think it is accessible in some other WorldCat interfaces. The element itself is not indexed, but it is being used to show open access items in our discovery systems.

If anyone knows, how quickly do FAST headings get updated when the LCSH term changes?

FAST headings are not controlled in the same way that LCSH headings are controlled, so that they change pretty quickly when the related authority record changes. Instead, FAST headings, they're all backed up by authority records, but the FAST authority records have to be updated first. A number of those are tied to the corresponding authority record in LCSH and will get updated. There's also additional FAST authority records that are created out of combinations of headings and subdivisions that have been used in WorldCat. That will lag behind by about a month, in the monthly processing of FAST authority records, and it's only after the FAST authority records are updated that we would then proceed to change all of the headings. It could be spread out over a number of weeks. Certainly within two months, they would presumably all be updated, presumably. But it's not a really quick process. We're working toward the ability to control FAST headings, and in that situation, updating should be quicker than it is now.

February 17, 2022

What if frequency has been changed over time, how to code it in fixed code, single one single code is allowed?

In that case, you update the fixed field to be the current frequency. Or if the serial is dead, the last frequency that it had, and this corresponds with what you would also do down in the body of the record, where you would update field 310 to reflect the latest frequency and move the former frequency to field
321. So you can have multiple former frequencies, multiple 321s, but only one 310 and that corresponds to what's recorded in the fixed field.

Why does the fixed field get condensed into one field in our ILS?

It's very hard for us to say what's going on with your ILS. The fixed field is the 008 field, most generally. There are also elements from the Leader, which we discussed in part 1. So OCLC displays these all as separate elements, but behind the scenes in the MARC format, these are all part of the Leader or the 008 field. Your ILS is probably displaying something closer to what the original MARC looks like. You can see this in the OCLC cataloging interfaces in Connexion or Record Manager, you can see the field as one field rather than displayed as we had it on the slides.

I am having trouble with the spacing on the 006, do you have any suggestions.

In the Connexion Client interface, you can opt for something called "guided entry" to help with entering elements in the 006 field. I have no idea what you might be doing in your local system with that, but I know guided entry is how I make sense of the 006 field.

March 8, 2022

In provider neutral records, providers can be added in 856 $3. Can this subfield be added to the index in WorldCat?

This is currently not the situation, but may be something worth considering so that records can be retrieved that come from a particular provider. What you can do instead is search the domain name that is in $u to retrieve items that come from one provider versus another.

In BFAS 3.2.1, does "In Analytics" apply to accompanying material?

Accompanying materials and "In Analytics" are treated separately. Accompanying material can be understood on its own, while "In Analytics" are an actual component of the resource and essential to use the resource itself.

I would like to see in the documentation an example of a special issue of a serial cataloged separately. I'm curious why 490/830 is used instead of a 77X-78X.

An example would be a special issue of a serial, which would often have its own special title.

Regarding the Corduroy Bear example, is there an example showing the MARC fields needed to link the two records if two records are created for the book and the bear? I.e., what is necessary if not using the 300 $e?

Typically we think of that particular link only going in one direction, from the separate record for the bear back to the parent record. But it could be the case that you have another link that goes in the other direction. It is recommended that one add the 300 $e as well, due to how accompanying material is treated in matching and deduping.

For parts of multipart monographs/serials, if a parent record for the series as a whole exists, is it preferable to use that rather than create a specific record for the volume in hand?

You have the option of either using the record for the entire multi-part or serial, or to create records for each individual part. Those aren't considered duplicates. There are a lot of reasons why libraries will take one approach versus the other. For example, if you have an item that is part of a serial, but you do not plan on subscribing to that serial. You may also choose to catalog all the issues of a multi-part individually to create more in-depth subject access.

If an item has marginal comments by a person of historical interest, where does this get entered?

This would be considered local copy-specific information (LBD or LHR fields), or in lieu of that, use the note fields which are not retained in the WorldCat record. If the person in question has interest beyond the local institution, use the $5 for special collections or rare material. Field 562 could be used as well.

For technical reports, what is the difference between standard and non-standard report numbers?

A standard number would be put in field 027. Those are the STRN and ISRN numbers which are defined by ANSI/NISO. Link to instruction: https://www. oclc.org/bibformats/en/0xx/027.html. Field 088 is for non-standard numbers.

Would you recommend a library that has recently migrated to WMS to split their collection record that described together tangible and online resources of recordings of an event? Or could they stay as described now?

If the record describes the tangible item and just notes the existence of an online resource (as the single-record approach outlined in the cataloging of electronic resources), it is possible to leave it as-is. However, the recent trend has been to catalog these resources separately, as it makes them easier to manage in the long-term, but either approach is acceptable.

For detached copies and offprints, why use 580 instead of 787 $i?

Historically, 787 has not displayed in a lot of online systems, so the use of 580 (and 500) is a carry-over from what has been done in the past. Using 787
$i is also acceptable.

I've come across a lot of maps that have been removed from atlases. If I catalog them as detached copies, should I still use their page number in the 300 $a, rather than "300 $a 1 map"? If so, if the page number isn't on the item, how should that be recorded?

In this case, catalog it as a single map, making mention of its previous atlas inclusion in a note.

When did the policy about being able to add 710/degree-granting institutions to thesis records change? I remember when this was not allowed and the 710's were removed from the existing master records. We have been using various local fields since then.

This was a change resulting from the adoption of RDA. AACR2 typically did not have any relationship terms in access points because if an institution was on every record for a thesis, when one was looking for publications of an institution, one would retrieve every thesis ever submitted for that institution. Under RDA the use of relationship terms is routine and "degree granting institution" is one of those terms, so we can clearly differentiate between the different relationships.

We have a collection of maps removed from National Geographic. Can I put them as a collection into one bibliographic record?

You can do that. Use bib level code C. If they have page numbers, record that in the 300 field and then add the 580 "detached from" and 787 for the National Geographic(s) that it came from.

Under what circumstances should we remove local information from OCLC records? Examples: copy numbering for a limited edition ("No. 45 of an edition of 100 copies"); "Signed by all contributors" (when that only applies to a local copy), etc. We're a WMS library and all of these local notes, added entries, etc. show up in our catalog and can be confusing to users. I find myself adding notes to our LBD (Local Bibliographic Data) records saying that, by the way, such-and-such doesn't apply to our copy.

OCLC attempts as best as possible to make distinctions between local information that has interest beyond the institution supplying it, and that which is only of interest to that institution. Some of this local information could possibly be removed, but if you have some fields marked $5, look at them closely and consider whether they have interest beyond the institution.

What are the best practices for cataloging mp3 files contained in tangible carriers, especially with reference to fixed fields: audio books on CD recorded as an mp3 file, for example. Has anyone established best practices for audio-enabled books?

Answered via email

To code the bibliographic record as Type (Leader/06) “i”, adding a textual 006 for the book aspect, a computer file 006 for the audio file aspect, and sound recording and computer file fields 007. Use field 020 for any associated ISBNs, field 024 for other standard numbers, field 028 for any publisher numbers.

RDA 3.4.1.3 and its Policy Statement allow the use of “a term in common usage (including a trade name, if applicable) to indicate the type of unit,” although the MLA Best Practices document frowns upon that, at least in the example cases of “CD” and “CD-ROM.” So using “VOX Book,” “Wonderbook,” or the trade name of your device in field 300 is permissible; a further explanation in field 500, possibly quoted from the resource, would be
advisable. Include any appropriate fields 344 and/or 347.

The 33X fields would be as follows:

336    spoken word $b spw $2 rdacontent

336    text $b txt $2 rdacontent

337    audio $b s $2 rdamedia

337    unmediated $b n $2 rdamedia

338    other $b sz $2 rdacarrier

338    volume $b nc $2 rdacarrier

For picture books, an additional 337 ($a still image $b sti $2 rdacontent) may be added.

Depending upon your device, you will want to include notes about the type of device (for instance, “Issued as a Wonderbook, a pre-loaded audiobook player permanently attached to a hardcover book”), any need for such accessories as a battery and/or USB charger, some details about the original print publication as needed. Generally, those are some of the main differences. In many other respects you will catalog such a device as you would many other standard sound recordings.

Although it is from 2010, is AACR2-based, and lacks many more recent MARC additions, you may find the OLAC/MLA Guide to Cataloging SlotMusic Based on AACR2 Chapters 6 and 9 useful. Much more recent (2020) and also possibly helpful is the OLAC Best Practices for Cataloging Objects Using RDA and MARC 21.

I'm creating records for some photomosaic maps from the Department of Agriculture. For similar items in WorldCat, most of them (for individual maps, usually on multiple sheets) have the CrTp coded as b (map series), but in the records there's no series information. Why are these maps considered a map series? I'm wary of using CrTp b if I don't understand it.

Thank you to Paige Andrew, Cartographic Resources Cataloging Librarian at Pennsylvania State University who provided the following answer via email.

When coding the Cartographic Type fixed field as "b" it covers true map series but also map sets. These do not technically meet the definition of a published series, however it is common in the map publishing world for a geographic area to have been mapped at a larger scale (more detail) on standard-sized sheets of paper so it may take many sheets of paper laid side-by-side to show the geographic area. Typically all of the sheets are at the same scale but not always the case. Map sets can also be based on the same area mapped but each sheet showing a different topic for that area, similar in nature to a general atlas of a country which has individual maps to show climate, soils, population, distribution of crop types and more; and the individual maps in this case are likely at different scales in order to show the topical information as clearly as possible.

So, in OCLC Bib Input Standards code "b" is defined as "A map series is a number of related but physically separate and bibliographically distinct cartographic units intended by the producers or issuing bodies to form a single group". I think many catalogers get stuck on the word "series" here, where they need to pay attention to the "related but physically separate..." aspect of the definition, which can occur either in the case of a true series or not, as explained above.

Another element that using code "b" effects is the date in the call number. Whenever this code is used there should NOT be a date of situation in the call number because in a map series or set most often the individual sheets are published over time and so using one date to represent all is misleading at best. The date is replaced by an abbreviated form of the scale of the group of maps, e.g., s100 represents the fact that all sheets in the series or set were created at a scale of 1:100,000. And if different sheets in the series/set are at different scales we use "svar" as a way to say "scales vary" across the group. Probably the best explanation, along with examples, of this phenomenon are provided in the Library of Congress Geography and Map Division Map Cataloging Manual; alternately in earlier editions of the LC G-schedule there is a table showing the layout of different kinds of map call numbers for the same place (New York City) depending on what the published situation is (map serial, map series at a definite scale, map series at differing/varying scales, individual map published in a given year, and so forth)...I have the 1976 edition (4th edition) open to page 207.

My only other thought to share -- besides the fact that this kind of stuff is unique to cartographic resources and therefore looks odd to a cataloger not familiar with the format -- is that maybe OCLC could adjust the code "b" label and definition to be explicit about map sets to be helpful, so maybe the label becomes "Map series/sets" and definition something like "A map series or set is a number of ..."

March 17, 2022

Detached Copy- is that when something is removed from your own library's copy?

That can be any kind of thing that was removed from any publication, whether that publication is in your library's collection or not. This could be an article or chapter that has been taken out of a serial or some other book. Sometimes it will not be clear what it has been removed from, and the cataloging rules account for that scenario.

Additional context added in chat:

Think of special collections where maybe a donated collection includes one article that was detached from the original journal (maybe the donor was the author and just wanted to keep a copy of the article). Not from the library's own copy of the journal but received and part of the collection.

What about two different books that are bound together?

This is not specifically addressed in chapter 3. Consider when cataloging an item like this: were the items bound together after the fact OR are they two works issued together by the publisher?

In the case of two works issued together by the publisher up front, this item should be cataloged as issued by the publisher and transcribe both of the titles into field 245. If it is the case that the items are bound together afterwards (as in a local binding), then create two records. Use field 501 to link those two records (particularly in cases of rare materials).

Does OCLC allow donor names in 700 fields as long as there is a subfield 5?

Yes, for rare and special materials as long as the donor is of interest beyond the local institution. This is not encouraged outside of rare and special collections.

How does one determine what is "beyond local interest" for donor names?

This is somewhat of a judgment call. Typically this would mean that the donor is an individual of historical significance and who is well-known beyond the local community.

We see many 7XX $e donors in WorldCat that should not be in the bib record because they are not beyond local interest. Can these be deleted?

If it is not beyond local interest, it can be deleted. If you are unsure, reach out to askqc@oclc.org.

Does OCLC offer a service that can automatically add LC or Dewey classification in a bib record, based on the first subject heading assigned by the cataloger?

No, OCLC does not offer any automated classification assignment. OCLC does have an experimental (non-production) classification service that is available for public use, but it is not automated (i.e., it must be activated outside of cataloging interfaces). http://classify.oclc.org/classify2/

Is OCLC working on an interface with the LC BIBFRAME Editor (Marva) or Sinopia, or both?

OCLC will support BIBFRAME and other standards going forward to meet the diverse needs of the library community.

It seems as though 856 notes that point to a Table of Contents should be coded as 856 42 and yet I see most of them as coded as 41, including records from the Library of Congress. Is there any way these can be fixed globally?

OCLC has the ability to perform a scan to look for table of contents as the text and the 856 field in combination with a second indicator value and change them from one thing to another as appropriate.

Additional answer provided after initial answer:

The second indicator should be 1 for a table of contents based on MARC documentation, so these are coded correctly. It should be 41 when it's a table of contents and $3 will indicate because it is a unit of a resource.

Are the $5 issues perhaps a result of the data ingest from sync projects?

Yes, this is probably correct. People have used $5 and their own local database records get sent to us. If they don't match, they are added. So if data of purely local interest is put into $5, then it will be added to WorldCat. This is a case where one may want to remove that access point if it is clearly local. If unsure, contact askqc@oclc.org.

Some self published books are occasionally reprinted with minor edits, sometimes a dozen times. Should we create a new record for each of these reprints?

Information present in the title page can be useful in these cases. If it is clear that some changes have been made, particularly if there is a new publication date, then separate records should be created. Language such as "revised printing" is a clear indicator that sufficient changes have been made to warrant creating a new record.

Why is a Book Club edition considered to be something that can be merged with other editions? In some instances it may be the same version, but I'm not so sure if that is always 100% the case.

If the fact that it is a book club edition is the only difference, then they are probably the same. If there is some other feature (such as paging) that is different, then that would indicate that the editions are different.

April 12, 2022

You say that edition statements may in fact designate different printings of the same edition. Could you define "edition" as separate from
edition statement? I don't see a definition in RDA.

There isn't necessarily a difference between an edition and an edition statement. The issue is that in some languages the word for "edition" (or words that
are like the word for "edition") are ambiguous and sometimes say "edition" as we think of in RDA or AACR2, but are actually statements of printings. This
often happens in European languages such as Portuguese, Spanish, Italian, and French. The statement may have a work that looks like "edition" or
ambiguously refers to "edition" but actually refers to the printing rather than the publication itself. These statements may also include the name of the
publisher and/or the number of books that have been printed.

RDA LCC-PCC practice: apply the option to transcribe ordinal numbers as numbers. Example: First becomes 1st.

This particular policy statement could not be located in RDA ToolKit. See instead: 1.8.1: "When recording numbers expressed as numerals or as words in
a transcribed element, transcribe them in the form in which they appear on the source of information. Apply the general guidelines on transcription (see
RDA 1.7), as applicable."

Will proper transcription of edition statement prevent bib records merging? With DDR, will the presence of a 250 field in a bib record prevent
merging with a bib record without a 250 field or with a different 250 field?

Yes, that is the intention. DDR was built to recognize the difference between the presence and the absence of an edition statement and to recognize the
differences between edition statements.

Would paperback vs. hardcover editions with different publication dates justify the creation of a new bib record?

Any differences between the paperback and hardcover editions, such as date of publication, number of pages, size, pagination, etc., would justify the
creation of a new record.

Would you advise using the 251 field along with 250 for uncorrected proofs, ARCs, etc. now? Or wait until RBMS has combined their vocabs?

RBMS has just recently released a beta version of the combined vocabularies with a public forum scheduled for April 20, 2022.

In French, how should we transcribe the edition statement for legitimate editions (not printings)?

Where there is an explicit edition statement (not a printing statement), transcribe it as it appears.

Is it OK to report for merging records when one is reasonably confident that a paperback record equals the hardback, or if the records are for
printings, and not editions?

Yes, either case can be reported.

Are very many libraries adding Galley copies or ARCs?

Not many, but some libraries are doing this. We do see these types of edition statements in the WorldCat.

There have been requests from libraries to retain paperback vs. hardback, because patrons want one or the other. Particularly novels;
paperback is easier to carry.

An approach to this scenario is to create two item records in your local catalog indicating hardback or paperback. However, in the shared cataloging
environment, if everything is the same and there is no explicit edition statement, the same record would be used.

What is a printing? Isn't it just a manifestation?

A printing is a manufacture of an edition, not a manifestation itself. It is a set of copies which were produced, that along with other copies of that same
edition would be part of what is the manifestation of a work.

Another thing to keep in mind is that if a paperback edition is catalogued separately and has a different printing date than the hardback edition,
then I think that the printing date would be treated as a publication date.

There is a hierarchy of dates, and it depends on the resource at hand. If you have a copyright date or publication date, and a printing date, you would
disregard the printing date and use the copyright or publication date. So it is not always true that the printing date would be considered a publication date.
Refer to DtSt:  Type of Date/Publication Status for more information.

Would local paperback/hardback be searchable, if library uses Discovery?

Answer (via email) from the Discovery Product Team: As Local Holdings Record (LHRs) do not have a standard way to indicate paperback or hardback
information which can be used to display to users for searching/filtering, WorldCat Discovery cannot create a such a feature.

April 21, 2022

Doesn't an edition statement end with a terminal period? Slide 9 examples do not have periods.

Examples follow guidelines from OCLC Bibliographic Standards and Formats (BFAS), which makes terminal punctuation optional.

Wouldn't FOR low voice be considered grammatically linked?

That depends on the context. In the grammatically linked example (see presentation slide 20), low voice, or a medium of performance statement, would be
part of the 245. In the non-grammatically linked example (see presentation slide 21), it says "version for low voice." Since it contains the word version (whi
ch is a kind of edition), it is linked to the version, but not anything else in the 245 fields.

What if a paperback edition and a hardcover edition are more than two years apart, would we create a new record for paperback edition?

Yes, since there is another difference besides the binding (in this case the year of publication), that would justify having two separate records.

If there are two books, one paperback and one hardcover, and only the size is different, are they the same edition?

A size difference of more than two centimeters would justify having separate edition records. Any difference at all besides the binding (paperback or
hardcover) would justify separate records.

Is it ever appropriate to use a date in the 250 field?

Yes, when the date is significant and shows a difference in the content for two editions, the date should be in the 250 field (see presentation slide 17 for
example). If it is on the resource, brackets are not needed. If the difference is known due to external sources, or by looking through the resources, then
the date information should be in brackets to show that this is a supplied edition statement that is not on the piece, but is known to the cataloger. Further
note that even the time of day may be a significant difference in editions. For example, a recital given twice in the same day. This information should be
recorded in the 250 as well.

What does OCLC or RBMS suggest for different printing statements? In some cases (e.g., when there are many different printings) it would be
useful to have separate records for them.

Under rare books cataloging guidelines, having a printing statement is justifiable. Make sure that the records are coded to indicate the rare books standard
to prevent the record(s) from being merged. If cataloging a general publication, the printing statement could be added locally using a 590 field, LHR, or
LBD.

If the edition statement reads "2nd edition, February 2019" do you include the date?

If explicitly stated in the edition statement, then the date information could be included in the 250 field. See BFAS chapter 4 for more information.

Is there anything being done to improve macro functionality in Connexion Client 3.0?

Connexion functionality is outside the scope of coverage for AskQC Virtual Office Hours. For information about Connexion, please see the OCLC
Connexion documentation at or contact OCLC Support in your region.

When is it appropriate to use a detailed date in the fixed field?

This is typically seen in technical reports. See BFAS 3.3.5 for more information.

Earlier a PCC rulebook was mentioned, where can I find that?

https://www.loc.gov/aba/rda/lcps_access.html

So if you are cataloging an item in French or German where they use the term for edition, how do you tell if it is really a printing date or if it
really IS an edition?

Generally, printing statements of this kind will include the publishers name or the number of copies that were printed. See BFAS chapter 4 for guidelines
on using the 250 field.

The examples you showed illustrated two separate 250 fields. Is there ever an appropriate time to include at least two edition statements in one
250?

Previously, field 250 was not repeatable, so there are many WorldCat records with multiple edition statements in the 250. Currently, 250 is repeatable so if
the statements are not directly related to each other, or could be thought of as intellectually distinct, it is preferable to include separate entries in the 250
field.

Are subfield i's in 700 fields optional?

Yes, they are optional. See BFAS input standards for each subfield.

Are subfield b's in 246 fields optional or required?

246 b is optional. See BFAS input standards.

How often is deduping run?

Under normal conditions, deduping runs continuously.

May 10, 2022

Regarding library location links and map: we catalog all materials for our countywide system under one symbol. The address associated with our institution is an administrative office, not open to the public; that is what displays if someone uses the map. We do have our catalog linked so users can click through to see which branch has an item available. Do you have any tips for our situation: best practice in the Registry when the cataloging institution's physical location isn't a place a patron would visit?

We recommend that the institution use their main address but use the "Also Called" name field to indicate that they are an "Administrative Office". Please contact OCLC Support in your region for assistance regarding questions about addresses.

What does the search use if there is no ISBN in the record?

If there is no ISBN in the record, it will take you the base URL to the search box at your main library catalog. If there's an OCLC number in the record it will take the patron to that specific item. This may be a trial-and-error kind of process for resources not held by your library. If you use the test link in the service configuration and it goes to an item that is not in your library’s collection, it will return a message stating “This search does not result in any entries” that the library patron would see. The library patron could then use the search box to locate the item.

How is Online Catalog Direct to OCLC Number (under Online Catalog) different from searching an OCLC number under "Accession Number" in a WorldCat Discovery catalog?

It is the same search, however it is the way that the search is performed and what the starting point is as to how the patron gets to the item. If you are searching WorldCat Discovery if the library has configured the direct to number link in the WorldCat Registry, then it is the number search that is being performed in the library’s catalog.

If a library has chosen the OCLC # search option, does it search only the main OCLC # from a WorldCat record, or is it also searching for all the record numbers found in an 019 field? For example, a library has a record in their local catalog that has an OCLC # that has since been merged in WorldCat with other records.

The direct to OCLC number search will only use the primary number when searching the catalog.

Follow-up question 1: Would it not also depend if you've flagged the "Add ocn" to the number?

It will depend on how the number is indexed in your catalog. For example, some are indexed within prefixes such as leading zeros. We recommend gathering an understanding in how this works when setting up the Online Catalog Direct to OCLC Number feature.

Follow-up question 2: As to whether it would get a hit in the 001 (which may have the ocn) versus the 019 which will not?

Similar to above, it depends on how your local catalog is setup.

We keep our union listing for serials up to date, but still get ILL requests for articles from issues we don't own. Has there been a change in how this information is supposed to be entered? Seems like people are not looking at the range of issues we hold, just whether or not we have the title.

This is not related to the WorldCat Registry but relates to how patrons interact with catalog information while using InterLibraryLoan.

WorldCat Registry, can you provide a bit of background on the Authorized Users tab?

The authorized users tab shows the people who are authorized to manage a record. Those are the authorizations that were mentioned in the presentation about WorldCat service authorizations and WorldShare authorizations. Once someone is authorized for the WorldCat Registry, they may add authorization to others. More information about authorization is available on the help page: https://help-nl.oclc.org/WorldCat/WorldCat_registry
/OCLC_Service_Configuration_WorldCat_registry/130Authorized_Users?sl=en

Can you talk about the Relationships tab and the types of connections? Also, I see doing a search that our library is also listed in the registry under a previous identity. Is there a way to remove that?

Yes, we are always looking to identify and remedy duplicate records. Main and branch relation types, but recently added academics, department, consortia relation types. Keeping this information updated. If you are authorized to manage the record, you may update the information. Please contact OCLC Support in your region with inquiries regarding duplicates or needing assistance updating a record.

Will the merge participant program ever expand to the registry?

Probably not, as the quantity is not the same as the bibliographic duplicates. If you identify duplicates, please send them to registries@oclc.org.

If we find duplicate bibliographic records in Connexion, should we report them? I’ve come across more than usual lately, that have the same OCLC#, so I assumed that they would be de-duped when the database is re-compiled at night.

If the OCLC numbers are the same on multiple records it is likely an indexing issue. Please send this to BibChange@oclc.org to remedy.

Can you point me instructions for diacritic entry into connexion? It’s been a while since I needed the information.

It depends on which interface is being used. For Connexion the keyboard shortcut is ctrl+e. Information about this may be found on the following Help page: https://help-nl.oclc.org/Metadata_Services/Connexion/Connexion_client_3_0/Connexion_client_basics/Connexion_client_interface_windows/Enter_Diacritics_and_Special_Characters

For diacritic entry in Record Manager, please refer to this documentation: https://help-nl.oclc.org/Metadata_Services/WorldShare_Record_Manager/Bibliographic_records/MARC_21_view/Insert_a_diacritic

We are using wms and fixing some LHR attached to wrong bib. When I move a LHR to another Bib. and change call no. at the same time, I can't find the new call no. using call no. browse. It happens more often lately. I think I need a reindex of call no. Anything I can do? or have to contact support for reindexing periodically?

Yes, please contact support@oclc.org regarding this question.

May 19, 2022

Where can I submit requests for registry changes/corrections?

registries@oclc.org

How does the WorldCat Registry tie in with Field 040 in WorldCat records?

The WorldCat registry record contains OCLC symbols and a numeric registry ID which is unique to an institution. The 040 field indicates that library that created the record, and $d indicates all the institutions that have modified it. The 040 ties into the registry in that if you wanted to get more information about the institution (for example, they put in a record and you have a similar item) then you could search the registry by ID to obtain more information.

I see a lot of picture books with 650s coded as LCSH but with $v Children's picture books. Shouldn't they be coded local or something?

Children's picture books is not a legitimate subdivision in LCSH. So children's picture books should be coded locally, corrected to a legitimate LCSH subdivision, or reported to bibchange@oclc.org.

Attendee noted in the chat: That subdivision is used consistently by New Zealand instead of Juvenile fiction/lit.

OCLC response: If you notice patterns of a particular incorrect subdivision being used consistently in certain libraries or in geographic areas. Please report it so that we can make corrections to multiple records.

It appears we have a registry account, but cannot get into OCLC Service Config; where do I obtain that number?

Check the documentation for account access, or contact registries@oclc.org for assistance.

When I'm using this site https://www.oclc.org/bibformats/en/3xx/300.html for cataloging help sometimes it will say things like see RDA 3.4.6.3. but it isn't a hyperlink and I'm not sure where to find what it is talking about.

We do not use hyperlinks to RDA because RDA requires a subscription that not everyone may have. If you do have an RDA account (to the original toolkit, in the case of these numbers), you can input that number in search on the RDA site, or use the left navigation to reach the resource.

A follow-up question, my login works in the new RDA Toolkit, but then it won't let me into the original one, does my institution need to pay for both or should my login work for both?

The understanding is that it works for both. There is a mechanism by which you can get access to both the original and the official or "new" toolkit. You have to re-enter your username and password after you click the link to the new toolkit. Link to the toolkit: https://www.rdatoolkit.org/index.php/node/262 and to the original: https://original.rdatoolkit.org/. If you continue to experience issues with access, contact: rdatoolkit@ala.org.

Can I ask general questions about the RDA toolkit? If our institution is subscribed to a certain number of users, is that total users, or is it concurrent users?

Concurrent users. Note that OCLC is not the provider of the toolkit, so this answer is unofficial.

June 14, 2022

Is it best to use a combination of the print Dewey and WebDewey together?

What's best is what works for you. Probably the strongest argument for WebDewey is that you’re always going to be up to date. We showed how you can cite specific dates in your MARC field to show that. In WebDewey you can also see, I usually say we update once a week, it’s not every day, but there are some weeks that every day there is some sort of change, we have a new built number or we tweak something. Compare that to print, there are things that are easier to scan in print. It’s certainly easier to look up and down on the same page than to click through different pages. In my work (which is not exactly the same as a classifier's work) I am using Dewey day in and day out and probably using WebDewey about 90% of the time or more. I do have some print volumes around and
sometimes it’s just easier to grab one of those and take a look. So, there is something to be said for that combination, but if I had to pick I would definitely still pick WebDewey.

Does Web Dewey integrate with Connexion Brower or local ILS such as WMS?

Sort of. Connexion Browser has a tab labeled “Dewey Services” that takes you to WebDewey. I think it’s supposed to automatically log you on with the same credentials, but I seem to recall some technical (or legal?) reason we couldn’t do this auto-sign on. At any rate, the integration is limited to this.

With ILSs/WMS, I would have no idea of any connection on the other end. In WebDewey, you can configure your local OPAC URLs such that when you’re looking at a particular number in WebDewey, you can click a “Link to OPAC” button and search that number in your OPAC.

Is there an easy way to view the geographic areas table in WebDewey?

That depends on exactly what you mean. There will be a link to that table, Table 2, at the bottom of any Dewey page and clicking on that link will take you to the beginning of that table, top level of it, however you want to think of it [Overview of WebDewey interface]. Otherwise that Table is something you can browse and search just like any other part of the classification. If you're looking for more of a one stop shop that lists lots of different numbers and Table 2, not necessarily in the same hierarchy, that might be the sort of thing you’d rather to browse through print versions for. Another thing, if you do use both WebDewey and print I would always double check WebDewey after finding something in print just to make sure it’s still up to date. If you’re using the print on demand for this year that is maybe not such a problem, but in older additions there are some very old Dewey numbers around in WorldCat and so on. So, one of the benefits of WebDewey is being able to check for a number’s recency.

A free Dewey Cuttering software was mentioned earlier. Do you need to subscribe to Web Dewey to access this software?

It is freely available https://help-nl.oclc.org/Metadata_Servi...Cutter_Program.

June 23, 2022

How do you "harvest" the LCSH for your mapping to class numbers?

It really depends. We have access, of course, to the full LCSH database. There was a period where we tried to map every new LCSH, and that sometimes would end up in some very specific numbers that maybe never got used besides the one title that I'll also authorize the LCSH. Today, it's something that when we're working on new developments we try to look at relevant LCSH in the area. But from a technical standpoint, it's very easy for us to use LCSH so, if you are ever using WebDewey and you're aware of some LCSH that you think could help you're welcome to send an email to do dewey@oclc.org. And we can add them pretty easily.

How can I find a list of new and revised numbers ? Is there a notification of some kind?

There is a built in feed of new changes to Dewey. In most of the WebDewey screen shots we showed there was a button at the top that said Updates with a specific number next to it. If you're in, live WebDewey, and you click on one of those buttons, you're taken to this feed that by default shows all changes from the past 30 days with
additions highlighted in green and subtractions struck out in red. But you can also change that to do different timeframes, you can do it to your last login, specific date ranges, you can go back to, I think about 2013. I think changes older than that will not show up in the feed. You can also look at specific duty areas so say you're just responsible for certain subject areas in your institution, say you only care about changes in science, you could tell it, for example, to show you what has changed in the 500s and the 600s in the last 2 months. There's a lot of mixing and matching you can do there. If it ever gets out of hand, you can reset those preferences to just show all changes from the last 30 days. And that's going to be the best most comprehensive way to see that kind of notification. If you use our print on demand version, there's new data for that every year. It's very much like a capture of WebDewey data, although WebDewey has more built numbers and so on that just would not fit in print. So, if you're using print on demand, the volume 1 foreword has some descriptions of some of the bigger changes from year to year.

We have a recurring problem with multi-volume records for graphic novel series being replaced with a single volume record. I don't know if it's because of bad merges or because a library changed the record to single volume for their own use but replaced the master record which was then merged.

I would say that when you come across something like this, if you could send us the OCLC Number to bibchange@oclc.org and we can investigate to see if it's a cataloging issue causing automatic merging to happen, or if a library is accidentally replacing the record, or if it's perhaps a CIP record that has been updated and no longer reflects what it initially did. But the best thing is to send us some OCLC record numbers, and we can investigate that for you.

OCLC has good Dewey and specifically WebDewey training materials, including the WebDewey user manual.

Dewey training courses is a page of free training resources. You can take those on your own. We kind of intended them for a trainer to teach groups with them but they are very much the sort of thing that you can walk through on your own as well.

Sometimes checking a number, I find in OCLC, it doesn't appear at all in WebDewey, why is this? Does it mean it isn't a real number? Are they using tables so they don't come up in a general search? Can you refer me to training on table use?

For this specific question about numbers in that are not in WebDewey there's a very real possibility that it's a built number that the classifier used tables for and it just won't appear in WebDewey. One thing I'll say with that is when you build a number in WebDewey you can save it to your account. And then you can further contribute it. And when you contribute it, it goes to me, and whoever else is on my team will review those and sometimes put them on WebDewey for everyone to see. So, that might help cut off some problems like that in the future. Otherwise, when you use number building there are so many possible Dewey numbers you can build. It's not quite infinite, but it's so large. And I wish I could offer something where you could plug in absolutely any number and be told yes, this is a valid Dewey number or not. But I would recommend some of the techniques that Brian showed in the presentation of checking those history notes. Maybe browsing on your numbers to see where that number would fit in and maybe get some sense of whether it is a valid number or not.

Could someone talk about issues with VIAF here, or is this not the forum? How does one get multiple persons in the same cluster separated and conversely the same entity on different clusters?

We can pull apart clusters that need to be pulled apart and we can also merge clusters that need to be merged. If you wanted to send. Names that you have, um, identities that you have for those to bibchange@oclc.org we can
take a look into that for you. VIAF has a link at the bottom of every page, "Send us a comment", that sets you up to email bibchange@oclc.org too. There will be a delay in getting VIAF clusters resolved due to the number of requests we receive.

Number building can be sometimes daunting. I search WorldCat and truncate, to see if double zeros are required for standard subdivisions, etc.

Thank you for that.

Has there been any thought given to standardizing animal names? I never know if I need to search the singular or the plural. If searching for penguin it retrieves the state in Australia and no animal number.

That's a good question. We have a system, but I know it's not always the most intuitive. Generally, when you get to a species name, it's going to be singular and larger groups will be plural. I'm going to use a hypothetical to say something like Squirrels versus, brown squirrel when you get to that specific level that we're going to use singular form. I see the example chat and yes, if you search the plural penguins since there are multiple species that's going to show up. But, yes, you do get some funny results, for example there must be a place in Australia named Penguin that if you search the singular can show up in the results. with. That's something I think of a lot because before there was WebDewey the idea of keyword searching didn't really come up. If you're looking at a print index or something, having both penguin and penguins maybe wouldn't be necessary. But I'm often looking when it comes to choosing index forms or mapping LCSH and thinking about keyword searching as well.

August 9, 2022

Can you elaborate more on the effect of a certain field not transferring in a merge? If OCLC merges some records, this means this field disappears?

There are certain fields that will automatically transfer from the delete record(s) to the retain record during the merge process, whether a manual or automated merge. If a field is not one that will automatically transfer and it is not already present in the retain record, then it would not be in the retain record after the merge.

Could you define PCC?

Program for Cooperative Cataloging (PCC) has several programs: BIBCO, NACO, SACO, and CONSER. BIBCO deals with bibliographic records and can be identified by code 'pcc' in field 042. NACO deals with creating name authority records. CONSER deals with creating serial records, that are essentially BIBCO records, but for serials. SACO deals proposing new LC subjects, genres, and other controlled vocabulary by the Library of Congress. with. For more information, see PCC: https://www.loc.gov/aba/pcc/.

Can you explain the indicators in Slide 22?

The first indicator for the 247 field, specifies whether a title added access point is made. The second indicator is called a note controller and specifies whether or not a note is made from data in this field. It essentially turns on or off a note that a system could display. If your system is able to read this indicator and display a former title note, then you would use code '0'. If your system is not able to display a former title note, you would use code '1' as the second indicator.

For those fields applicable to Serials/Continuing Resources, would editing by PCC institutions be restricted to CONSER participants or open to any PCC institutions?

A CONSER authorization or CONSER role is needed to edit a CONSER record. BIBCO or PCC authorization or role cannot edit a CONSER record. Those with a NACO authorization or role can edit BIBCO, or monographic, PCC records, but not CONSER records.

Why are edition statements (field 250) no indexed and searchable in OCLC? It would be especially helpful to be able to search this field when working with older materials that yield a large number of search results.

We agree, that would be very helpful. Unfortunately, we do not know why is it not indexed, but that decision could be revisited at some point.

For the 250, if the $b mentioned the author himself, should we keep it or delete? Also, can you explain more about the 246 field, should we keep it for the subtitle (245 $b)?

It will depend on how the author is presented with the edition statement. Is he or she responsible for the revision of the work, as well as the content - it depends on the item in hand and how the information is presented. You could potentially use field 246 for the title and the subtitle. Generally, you would not have the same title in 246 subfield $a as you would in the 245 subfield $a. For instance, a cover title that had a title and subtitle and the subtitle is not on the title page, then you would have a field 246 14 with the title and the subtitle. But, if the title in the 246 subfield $a is the same as the title in the 245 subfield $a for that cover title, then you would only record the subtitle in the subfield $a of the 246 field, instead of subfield $b, with a note saying "Subtitle on cover".

Can you delete field 263 once I have the item in hand to catalogue?

Yes, field 263 can be removed when upgrading from CIP to full level, with the item is in hand. If it is an LC CIP record, you are not permitted to change the encoding level 8 to blank, unless you have the appropriate permissions.

Would field 243 be used now that uniform titles for aggregates are based on the aggregate title? So, for instance, would we do the following, or am I misunderstanding the use of this field?

100 1  King, Stephen, $d 1947- $e author.
240 10 Night shift
243 10 Short stories. $k Selections
245 10 Night shift : $b stories / $c Stephen King

OCLC does prefer use of field 240 versus field 243.

Can you elaborate on field 246 versus field 740? I have seen field 740 only rarely. What is the reason for using one or the other?

When you have multiple titles in field 245 for separate works, for instance 2 works (work A and work B), by different authors but published in the same resource, field 246 would represent the varying titles for work A that appear on the resource. Field 740 would be used for the title of work B that appear on the resource, not for the varying titles of that work.

Is there a reason field 250 isn't sufficient for describing field 251? "Draft" and "Author's original" seem like edition statements to me.

Field 251 is a fairly new field, and we are still exploring exactly how it would be used. Based on the proposal and from MARC documentation, it's mainly intended for controlled vocabulary related to the edition statement, versus the editions statement in field 250, which is transcribed from the resource exactly as it appears. Field 251 is a different kind of edition statement, in that it's not intended to be transcribed so much as discerned from the resource.
Proposal that created field 251: https://www.loc.gov/marc/mac/2018/2018-04.html

Could you talk a bit about when you would use field 258?

An example would be for cataloging stamps. Looking at WorldCat statistics, this field has not been used that often. So, the cases where it would be used are very, very rare.

Is there a difference between "uniform title" and "preferred title"?

Preferred title is the title found on the resource that is being cataloged. A uniform title is supplied by the cataloger and may or may not match the preferred title, and is the title that pulls together all of the different works or expressions of a resource.

About the 264 _1 $c, sometimes the date is in brackets, sometimes not; may have 264 _4 $c with (copyright symbol) date. I'm trying to figure out the pattern of use.

The 264 _1 subfield $c is for the publication date, which may sometime be inferred from the copyright date because the resource doesn't have an explicit publication date, it just says copyright 2020. In which case, the 2020 date would get recorded in the 264 _1 subfield $c in brackets. An additional field 264 _4 subfield $c would also be added for the date preceded by the copyright symbol, not in brackets, since it is coming directly from the resource.

For field 222, who creates the authorized form for the key title?

The ISSN national centers are the source for the key titles. They assign both the ISSN and the key title. If you are entering a new record, whether CONSER or not, the ISSN and key title may be added if they are clearly labeled on the resource. If creating a microform or photocopy reproduction, you may transfer the ISSN and key title from the original to the reproduction record.

Could you give a couple of examples of the difference between continuing resources and serials?

Continuing resources are resources that intend to continue over time. Serials have discrete parts, such as volume or issue, and they are intended to be added over time and there is no real end of publication planned for them. Serials fit within the umbrella of continuing resources.

When should the $b be coded in the 250 field instead of the 245 field?

That will depend on how the information is presented on the resource. Is the title followed by the statement of responsibility and then a separate edition statement that is freestanding? Does the edition statement include its own statement of responsibility other than the title statement of responsibility? It just depends on how the information is presented on the resource.

Would you please include an example of 2xx best practice for bound-with in the notes?

The only 2xx involvement in a "bound-with" situation that comes readily to mind is that only part of a non-collective title may be recorded in field 245 (with possible related 240s). In BFAS field 501 (https://www.oclc.org/bibformats/en/5xx/501.html), there is an example of AACR1 cataloging of a sound recording where each side of the disc is a different, separately cataloged musical work. Although such an example isn't literally "bound-with", the cataloging practice was identical.

501 With: Peer Gynt (Suite) no. 1-2 / Edvard Grieg -- Till Eulenspiegels lustige Streiche / Richard Strauss

Is it correct that in RDA we don't put a devised title in brackets?

Devised title - see RDA 6.2.2.6.2 & 2.3.2.11. RDA 2.3.2.10, last sentence "Make a note to indicate the source of a title proper."

If the cataloging agency translates the title in field 242, can the translated title be used as the title statement in field 245?

No, the purpose of field 242 is to allow the cataloging agency to translate the title that is in the 245 field to facilitate access for their users. It is not found on the resource and is not in the language of the resource, so it should not be added to the 245 field.

Regarding field 264 publication year and copyright year, if the same and both are on the resource, then does one use code 't' and repeat the years in the 008 field?

Yes, but the cataloger also has the choice, if the copyright year and the publication year are the same, to leave out the copyright year because that is optional information. So, you could have a single date in that case. Either way is correct.

If the cataloging agency translates the title in field 242, can the translated title be used as the title statement in field 245?

No, the purpose of field 242 is to allow the cataloging agency to translate the title that is in the 245 field to facilitate access for their users. It is not found on the resource and is not in the language of the resource, so it should not be added to the 245 field.

If you leave out the copyright year in the 008 field, that would mean also leaving it out in field 264?

Yes.

August 18, 2022

What does "transfer in a merge" mean?

There are certain fields that will automatically transfer from the delete record(s) to the retain record during the merge process, whether a manual or automated merge. If a field is not one that will automatically transfer and it is not already present in the retain record, then it would not be in the retain record after the merge.

The majority of the items that I catalog are for special collections. Most of the time, there is no title for these items, and I create a title or the library provides a title. Should all of these titles be in brackets in the 245 field?

That would depend on the cataloging guidelines that are being used. You can use brackets, but under current cataloging, RDA has an allowance to simply notate it in the notes field that it is a cataloger or supplied title.

Would we use both field 250 and field 251 together?

This is a fairly new field, and we don't believe there is anything preventing use of both fields in the same record. Field 251 is intended for controlled vocabulary related to the edition statement, versus the edition statement in field 250, which is transcribed from the resource exactly as it appears.

Could you please repeat when it is appropriate to use field 246 versus field 740?

Generally speaking, field 246 would be used for other varying forms of title that are found on the resource. In the case where you have a resource that has 2 separate works, either by the same author or different authors, published together with no collective title, you would record both titles in the 245 field. Instead of tracing that second title in the 246 field, you would trace the it in a 7xx field - either a 700 with subfield $t or a 740 field.

"May not be edited or added to"--can it be deleted when no longer applicable or no longer correct?

The action to delete a field would fall under the same as the action to edit a field. Meaning if it can be edited, then it can be deleted, and on the reverse, if it cannot be edited, then it cannot be deleted. Also, if data is no longer applicable in a record, consider the item you have may be different and requires a new record.

Can/should the 263 field be deleted if an institution is upgrading an ECIP record?

While you are not able to change the encoding level in the CIP record, you should delete the 263 field, if you are able to, when upgrading the record from pre-publication to full because the item is in hand. The 263 field should not be deleted for serial records. The US ISSN Center uses that field to manage their internal files. BFAS points out that catalogers with National Level enhance authorization or the Cataloging BIBCO role may change the encoding level and remove field 263.

Could you repeat the practice for field 243 versus field 240?

A bibliographic record ought to have just one preferred title in 24x in combination with 1xx. Bibliographic Formats and Standards (BFAS) for field 243 says to "Prefer use of field 240 in place of field 243 when entering or replacing data in WorldCat records" and goes on to say "Do not enter field 243 and field 130 or field 240 in the same record". So, field 240 and field 243 should not be in the same record, and any 243 field should be changed to a 240 field.

What is an example of a 243 field?

Prefer use of field 240 in place of field 243 when entering or replacing data in WorldCat records.

With the new fields, will the macros in Connexion be updated? And how will it be deployed?

We will be working on updating the macro that is in the Connexion client that is used to derive electronic records from print records as it currently supplies the 655 Electronic books field which will be deleted from bibliographic records. When it is ready, we will post it on the OCLC website and put out announcements that it is ready for download. When the next version of the Connexion client comes out, it will be incorporated.

Besides the 655 Electronic books cleanup, are there other projects that your team is working on that you can talk about?

We are currently working on the change of encoding levels, i.e., M to blank, 7, or 3, in addition to the other alphabetic codes that we've been working on for a while now. We also have a lot of little projects that we've come across, such as the language of cataloging (040 $b) is coded incorrectly. If you notice a particular pattern of error or problem, please send those to bibchange@oclc.org so that staff may look into it and make the necessary corrections.

Are multiple 245 fields allowed when providing the title in different alphabets/character sets? If so, how would one link them in Record Manager?

Field 245 is technically not a repeatable field. Non-Latin character sets are stored in system supplied 880 fields. In both Connexion client and Record Manager, two 245 fields would be added and linked together and these do display as two 245 fields, however in the background the non-Latin field is a 880 field. By simply putting non-Latin characters in a 245 field, the system will automatically make it an 880 field behind the scenes, even though it still looks like it's a 245 field. Any 245 fields with no non-Latin characters are an actual 245 field and can be linked to a 245 field that contains one or more non-Latin characters.

If a LCCN has already been assigned to a title, but the LCCN reappears in a reprint of the same title, should the LCCN be supplied if creating original cataloging for the reprint?

If chosen to include the LCCN, it should be recorded in a subfield $z of the 010 field for the reprint.

When will the next version of Connexion client come out?

Version 3.1 of the client was just released, and so we don't know when the next version will come out.

Does OCLC plan to remove the 375 field from name authority records?

The 375 field denotes gender of personal names, and PCC made a decision to no longer use that field. We have been in contact with the Library of Congress because the Name Authority File is LC's file, and OCLC has a copy that is synced every 24 hours with that file. So, we are not able to just remove things on our own on a large scale like this. We have been talking with them for the past 3 months, and we do plan to remove them but we need to coordinate that with the Library of Congress and the other NACO nodes who house copies of the files. We are waiting on word from the Library of Congress to when we can begin that work.

In records we have recently derived rom existing WorldCat records, I coded ELvl as blank, but when I go back later and pull up the records in Connexion client, they have reverted back to M and the subjects I controlled have become uncontrolled. These records are not being batch loaded, they are contributed live. I sent this to support@oclc.org, should I have sent it to bibchange@oclc.org?

It depends on how the records are being contributed to WorldCat. If they are batchloaded, that would be why the encoding level is changed to M and the uncontrolling happened. If the records were created live in an active session in Connexion client or Record Manager, the encoding level would not randomly change. However, if created in a live session and then your institution batchloads all your files and one of the Data Sync options is set to Replace Own Record, that means that the incoming record will replace your record in WorldCat if your institution is the only one with holdings attached. So, we assume these were added via a batchload or ingest process. Yes, please send these types of issues/questions to bibchange@oclc.org.

Is there any way for OCLC to clean up subject headings that have the subfield Comic books, strips, etc. followed by anything (usually Juvenile fiction/literature or Fiction)? The following subfield should either be removed or the heading not tagged as LCSH.

We can definitely make note and research the issue and make corrections as necessary.

I have come across records with non-Latin and Latin titles that are not linked correctly. How can we fix that? Should we sent it to bibchange, or is there a way for us to fix it ourselves?

There are instructions in both Connexion client and Record Manager there is an option to unlink and link fields. For more information, see the instructions for linking/unlinking non-Latin fields in Connexion and Record Manager.

Do you want people to report as errors random juvenile 520 notes when we find them in records for entirely different works?

Yes, please do report these to bibchange@oclc.org so that they can be corrected.

I've recently noticed a lot more headings in records in Connexion that are not controlled than I used to. Some are even the type that should get controlled by an automated process (names with dates, subjects, etc.). Why is that?

We are still investigating what exactly has caused that or why that happens or happened. It seemed to be a part of the retrospective that we completed earlier this year, and so we are looking into why it happened and trying to find ways to prevent that from happening as much. For headings that have dates and should be controllable, they should get recontrolled when the records go through the controlling process again, once we've figured out why they got uncontrolled in the first place.

September 13, 2022

Sometimes I run across records with local data in a 500 with a $5, or a local collection name in a 710/730 with a $5. Should we leave those in the WorldCat record?

If they are truly local and have no other interest to any other institution, they can be removed.

How does a cataloger get a record deleted from WorldCat?

Contact bibchange@oclc.org to request a record deletion. A record can only be deleted by request of the library that created the record, and if it has no holdings on it (or if the only holdings are those of the creating library). If you are a record manager user you have the capability to delete it yourself if it has no holdings.

Regarding editing wars, is there a way to track editing changes similar to Wikipedia?

Contact bibchange@oclc.org to report editing wars. OCLC staff are able to track changes using our journal history tool.

What about subject and genre access points in languages that differ from the language of the cataloging institution?

Please leave those in the bibliographic record. For your local version of the record, you can filter them out or delete them locally before you add your holdings to the record.

What's the logic of not allowing non-PCC catalogers to edit the 500 in PCC records, when many other 5XX fields can be edited?

The 5XX, or 500 filed itself, can contain so many different types of things, that OCLC felt it was best to leave it un-editable. However, this policy can be reconsidered. Note that many other 5XX fields can be added or edited.

If we have a book in Hebrew and add a 520 with summary from back cover, is it ok to add a second 520 with an English translation?

Ja

I may have misunderstood this, but if my library has the only holdings on a record created by my library, I am not supposed to replace it?

You may replace it, but you should not change the nature of the record to represent another bibliographic entity.

There was a discussion on PCCLIST a while back about restrictions on adding subject headings to PCC records that already had subjects in the same scheme. The libraries who raised this question were trying to add DEI-related headings to the records. Might this be a reason to consider easing those restrictions?

The idea behind a BIBCO record is that every heading (including subject headings) is backed up by an authority record. Even when those headings that would be potentially added with the good intentions of increasing access in DEI, the principle still remains that those headings must be backed up by authority records. If you are not a PCC institution and there are headings that need to be added, but the heading scheme is already represented in the bibliographic record, please report them to bibchange@oclc.org.

If I have a book in hand, and I see the LCCN number on the title page verso, may I add it in the 010 field and replace?

If it is not a BIBCO record, it can be added.

May we delete institution-specific access restriction notes? For example: "Available to School of Music students and faculty only."

Yes, those can be removed. However, please report to bibchange@oclc.org so we can investigate if those notes have been applied to a large quantity of records.

Are there a handful of more prevalent errors that you wish more people would edit to correct?

There is an extensive list of common errors. Misspellings of subject headings, mismatched 2nd indicators in 856 fields, misspellings of titles and the 245 are examples.

I tried to add two 500 fields to OCLC # 932171599 but was deemed "unauthorized." Should I ask bibchange to do this or add the notes as 545 or other fields?

If the information you are adding is more specific to one of the 5XX fields as opposed to the 500 field, then do add it there. Otherwise, please contact bibch ange@oclc.org.

When a book changes to be a multi-volume, is it better to create a new record or change the existing record with open publication date?

OCLC prefers that the record is left as a single volume and a new record is created for the multi-volume. BFAS guidelines do allow for the existence of records for both single and multi-volumes.

Sometimes, there are many institution-specific 856 fields referring to ebooks from vendors (often as proxy prefixes). What is the recommendation of OCLC?

OCLC would prefer not to have institution specific URL in the 856 field. However, do not confuse them with provider specific URLs, which are appropriate in the bibliographic records.

So, to be clear, 856 fields with institution-specific proxy prefixes should be deleted from the master record, correct?

In general, yes. However, OCLC prefers the last 856 in the record be retained if possible. If the URL does not work, it should be moved to $z to remain as a reference.

Are electronic book bib records supposed to be non-vendor specific?

They should not be vendor-specific if cataloged under provider neutral guidelines. However, many records do come from vendors that are often incomplete or pre-publication. In those cases, once the item is in hand (or access to the electronic resource is gained) please upgrade the record to make it provider neutral.

What do you think about adding 050 number, versus changing an existing 050 with a better one?

This is a matter of judgement. Generally, OCLC prefers that 050 not be changed, unless it is explicitly incorrect, such as when a classification has become outdated. There are instances where multiple 050 fields are appropriate. See BFAS for more information: https://www.oclc.org/bibformats/en/0xx/050.html.

I have seen institution-specific 583 "committed to retain" notes in records. Should they appear (be retained) in OCLC records?

This type of information is often seen in consortia where the data is shared and is therefore not necessarily institution specific. Best practice is to leave the field as-is.

What about "commitment to retain" notes by national libraries?

Usually consortia-related notes are appropriate to remain in the record. Please see: https://www.oclc.org/en/services/sha...uidelines.html for more information.

What should we do if a URL found in a record is no longer functioning and we are able to supply another? Do we edit the existing broken link, or just add a new working link?

Broken links may be edited. If uncertain, contact bibchange@oclc.org for investigation. Please provide the new URL if available when submitting.

If a library has used field 050 for Library and Archives Canada classifications (PS8000, FC) rather than field 055, should we delete these fields or change them to 055? Often there is already a field 055 in the record.

If the classification is in the wrong field, then do edit if possible.

Where is the best place to find online help for editing records?

Refer to Bibliographic Formats and Standards https://www.oclc.org/bibformats as well as previous AskQC recordings.

All Virtual AskQC presentation can be found here: https://help-nl.oclc.org/WorldCat/Metadata_Quality/AskQC this includes the recording plus Presentation slides and Member Q&As.

Editing-specific presentations:

January 2020: Best practices for editing WorldCat bibliographic records 

February 2020 :: Best practices for enriching WorldCat bibliographic records
 

Once a request is submitted for a change to OCLC how long does it take for the change to happen?

Timing depends on staffing and volume. A couple of days to one week maximum is typical. Reporting duplicates typically takes longer as they are sent to a backlog, and turnaround time varies based on format (since, for example, there are a higher number of duplicate reports for books than other format types).

Is tracing 490/830 getting obsolete? I see it less.

One reason that it may be seen less is that Library of Congress has stopped doing series authority work, so it is up to other libraries to decide whether a series name and/or title will be traced. The practice is more infrequent now, but unlikely to go away entirely.

This came up on one of the mailing lists recently. If we see LCSH on a record that do not follow the Subject Headings Manual, e.g., "American literature--21st century" added to individual novels, should we leave them in or remove them? The SHM specifically says not to do this, but I think a lot of public libraries add it for their patrons.

Those are incorrect and may be removed.

If we see headings (such as names) that need to be edited in a lot of bibs, is that OK to forward? Do you have global update capabilities?

Yes, please forward and the records can be edited in bulk by OCLC.

Is it better to use the Report Error function in OCLC Connexion or email bibchange@oclc.org if we see a record needing changes that we cannot make?

Either approach is acceptable. Please use whichever method is most convenient to you.

If we edit records while doing authority work, it's OK now to edit them even when our holdings aren't on them?

Yes, any incorrect data can and should be updated, even if your institution's holdings are not on them.

I see many 655 entries with $2lcgft, only the term really isn't LCGFT. Should these be deleted, or changed to $2 local?

Those may be removed.

Are we free to remove 655 (any indicators) that say "Electronic books" and even more so, "Electronic journals?"

Yes, you may remove these at any time. However, OCLC has a project starting soon to remove those headings from the records in bulk.

The problem of duplicates seems to be getting worse, especially for e-books. Is OCLC doing anything to address this?

OCLC tries to merge as much as possible. Please report duplicates that would benefit changing OCLC's Duplicate Detection and Resolution (DDR) algorithm, which is continually updated. E-books are particularly tricky because descriptions can vary widely, yet still represent the same resource. The DDR program attempts to be as specific as possible while also "doing no harm" (i.e., it is designed to err on the side of caution when determining duplicates).

September 22, 2022

If a record is cataloged in one language, but language of cataloging does not match, do we submit an error report rather than fixing it?

Please report to bibchange@oclc.org. OCLC will investigate to determine the nature and extent of the error.

I have wondered about correcting romanization if you believe someone has followed incorrect romanization rules for a non-Latin script.

Please do correct such records if you possess the appropriate language and script skills.

Can you explain (again) the constraints on linking 6xx fields to non-Latin script equivalents?

It is not a proper practice to link non-Latin fields to subject headings because there is no non-Latin equivalent to LCSH (i.e., there is no one-to-one correspondence).

What about paperback reprints with and without different pub date?

All other fields being equal (including pagination, publisher, size, etc.), then these are not legitimate different records. Please report duplicates to bibchange@oclc.org.

Why are we limited in how many times we can replace a record? Is it just for internet bots?

There are no limitations on replacements. However, depending on what has been changed and/or added, the record may be changed in such a way that it may not be replaced again.

What about names used in 600s? This is related to the linking to non-Latin scripts. Can you explain the constraints on linking 6XX fields?

The remarks about non-Latin script fields for subject headings apply chiefly to “Library of Congress topical subject terms and non-geographic subdivisions,” as stated in the presentation. Names in Latin script 6XX fields may be linked to parallel non-Latin fields in the same way that they may be linked in fields 1XX and/or 7XX. The same cautions regarding the maintenance of relationships between Latin and non-Latin parallel fields also apply in the common cases of (Arabic) dates and Latin script qualifiers and subdivisions that will often be in the same field as non-Latin text.

     Examples (correct):

     600 14 , d 1611-1680 x Criticism and interpretation.

     600 10 Li, Yu, d 1611-1680? x Criticism and interpretation. [#925513697]



     651 4 (Jiangsu Sheng, China) x History.

     651 0 Nanjing (Jiangsu Sheng, China) x History. [#917131808}

What we were trying to say about not linking Latin script subject headings to non-Latin fields was to discourage an LC subject heading (for instance) being linked to a simple non-Latin script translation of that heading.

     Example (incorrect):

     650 4 [Chinese script word for “economics”]. 650 0 

     Economics.

I just want to confirm, what's the official take on adding information on FolletBound and PawPrints rebinding and other aftermarket library binding editions to existing records? They'll have the original ISBN number on the copyright page inside the book, and then have a separate ISBN on the outside cover.

All other fields being equal (including pagination, size, date, etc.), use an existing record and add the additional ISBN(s).

Occasionally, a glyph intrudes into a field (for instance the black diamond with a question mark incident), and then a parallel, linked field without the glyph is created, resulting in an 066 for the presence of symbol coding. Is it permissible to delete the erroneous instance of the paired field and then delete the 066?

Yes, that may be corrected (removing the incorrect character) and the 066 will disappear.

Could we see an example of editing a record in OCLC Connexion? if you make the change then is it Action then Replace Record, or Replace and Update Holdings?

If you already have holdings on the record, then you would just have to do a replace. If you want to add holdings at the same time, then you would do a replace and and update holding.

Could you clarify the use of 655 vs 690 for local subject headings when creating original records?

655 is for genre/form terms (including local ones). 690 is used for local subject headings.

I do see that language of cataloging in 040 is eng but some 6xx in French is it okay or what should I do in this case?!

Language of Subject headings is not a factor in determining language of cataloging. A record cataloged in any language can have subject headings in any language.

OCLC has asked us to delete the FAST headings if we change or delete incorrect LC subject headings or LC genre terms. But if we try to delete some Répertoire de vedettes-matière or Medical Subject Headings that were added in batch processes based on the incorrect headings, we might get validation errors. Is it possible to remove the validation error from the profile so we can judiciously and quickly delete Répertoire de vedettes-matière or Medical Subject Headings?

Because they are controlled, it's necessary to be careful when deleting them or otherwise trying to work with them. As for the validation/replace error, generally if working in Connexion, it would be an exception to the guidelines to do all changes in the same transaction. For deleting invalid MeSH/AAT
/RVM headings that correspond to a deleted LCSH, you should delete them, replace the record, and then add or delete LC headings.

You could use the 655 with a 2nd indicator of 7 and a $2 local, correct?

Yes, that is an option to add local genre headings to a record.

Can you add ISBN for a Paperback reprint to a hardback record if the pub date is different? Or if it is the same?

If the publication date of the paperback is the same, then the ISBN can go on the record. If the publication date is different, then it should have its own record.

Is it possible for the validation error to be removed when trying to export an authority record with different language symbols?

All records can be exported. The Connexion settings need to be default which means "no validation." For more information see: https://help-nl.oclc.org
/Metadata_Services/Connexion/Connexion_client/Authorities/Export_authority_records.

A point of nuance regarding paperback printings on records for hardcover manifestations -- be alert to differences in printing date vs publication date. I would think that printing date differences do not require a separate record. Those pesky "Paperback edition" statements might be a different matter.

That is correct. Refer to BFAS on when to input a new record: https://www.oclc.org/bibformats/en/input.

Is there a way for OCLC to review situations where records have been merged so that the record now represents a different bibliographic entity? Example: we have created original bib records that were merged into other bib records even though the publishers or publication dates were not the same. It is possible that someone modified our original bib record so that it no longer represented the same bibliographic entity?

Yes, those should be reported to bibchange@oclc.org for investigation. If necessary the original record can be restored.

Yes, those should be reported to bibchange@oclc.org for investigation. If necessary the original record can be restored.

These are available on the NACO website.

General NACO Information: https://www.loc.gov/aba/pcc/naco/index.html

NACO Training: https://www.loc.gov/aba/pcc/naco/training/

RDA in NACO Training: https://www.loc.gov/catworkshop/cour...RDA/index.html

New Zealand always uses Children's fiction instead of Juvenile fiction but codes the heading as LCSH. Can these be corrected automatically or do we just have to uncontrol them all, fix them, then replace the record?

Thank you for reporting this. This issue can be targeted and automatically corrected by OCLC.

Can you change non-authorized name headings with authorized ones based on your own intuition and using the authority file?

Yes, if the record does not have a controls name and the form is not authorized, you may update it to the authorized form and control it, then replace the record.

I've requested a record merge because I created a record in error. It's been months and no change. What should I do next?

I've requested a record merge because I created a record in error. It's been months and no change. What should I do next?

This may be a trivial question, but I'm curious: in 300 $b I've been noticing that some libraries are using the phrase "black and white" to describe illustrations, maps, etc. that are not in color. Is this a new standard to follow?

This is not a standard practice. Relevant attendee comment: There's an alternative to RDA 7.17.1.3 that allows for recording "black and white" for color content.

October 11, 2022

Is there research available about FAST being easier to train and use?

I am not sure if there is research, but we have put together a bibliography for FAST and we haven't got that up on the website yet. We hope to, soon. There are a lot of people who have used FAST and studied its use locally and have said, based on their experience, that it is easier to train and use. Granted, it’s not actual research but we certainly have people that have shared their experiences along those lines. I believe in the British Library’s document FAST Forward they describe their experience and the reason they've switched to using FAST and so there might be more information there.

How quickly after a change in LCSH are the corresponding FAST headings changed?

It takes almost a month for the changes to be published and made available to libraries and FAST. The reason is that we still do quite a bit of manual review for the changes to headings to make sure that FAST is able to accommodate the changes. We have automated software that makes the changes, but then we do a manual review to make sure that our software has created headings that make sense. So, it takes about a month for those changes to be released.

How to handle geographic locations that are not available in searchFAST yet? Such as State-City but the city is not yet indexed ... cities that were available in LCSH

Currently, we scan WorldCat periodically to look for geographic headings that occur frequently in bibliographic records and add those headings to WorldCat again, the geographic, because they are so different from settings they require manual review before we add the headings and so they're not always available in FAST. Right now, if
you have a special request, we take the request online, or via email to FAST@oclc.org, and you can send your request in. If it turns out that the volume is such that we need to come up with a different solution for adding those headings we will figure something out, but we don't allow them to be imported the way we do personal names and corporate names, because the construction is much more complex than constructing a personal name or corporate name.

How often are FAST added back into records that are edited?

I assume by “edited” you're asking if you change the LCSH headings in a record. In that case, our algorithm or our software does not revisit headings that currently have FAST headings in them. So, if the change is a one-to-one change with an LCSH heading, because the heading change, then our software each month effects that same
change to the FAST heading. But if you are adding an LCSH heading or removing an LCSH heading the best way to have the FAST headings updated for that record is to remove the FAST headings, all FAST headings from that record. And then our software identifies bibliographic records that have no FAST headings and rebuilds the FAST
heading array for that heading. But again, if the change is just LC changed a topical heading or a geographic heading, and it's a one-to-one change, then FAST will make that change usually within 30 days.

Where are the documents that list out all the FAST terms for a particular facet?

I don't know that there's a document that lists out all of the FAST terms but the FAST Authority File is downloadable and I believe you can download specific facets. And so the other things that you can do is, you can use our tool searchFAST and searching on terms, you can limit by facet. So if you want to look at, for example, named event headings, only you can use search FAST. There's a drop-down list to limit to only the named event headings. And you would, then any search terms would return only those named event headings.

Are new FAST headings being created for FAST or do these still depend on creating an LCSH first?

We're still based on LCSH headings but what we have done is created a FAST Funnel where libraries can submit to the FAST Funnel proposals for headings as those are approved. Then we add those terms to the past Authority File.

Where is the FAST Funnel?

Fast Funnel is part of the PCC SACO Program, which is the Subject Authority Cooperative Program that is organized by the Library of Congress. The administrators for the FAST Funnel right now are at the British Library. But if you go to the SACO Funnels page you'll see the FAST Funnel listed there and you'll have the links and the email addresses of the people that administered the FAST Funnel and instructions that will lead you to instructions for how to submit a new heading request or a changed heading request.

Do you foresee that changing or will FAST always depend on LCSH?

It's hard to say, right now, we don't have the infrastructure in place that LC does to vet new headings and to see how they fit into the structure that is already in place so that there aren't conflicts or differences, the same concept expressed in 2 different ways. So, it certainly is a possibility that FAST could accept and have new headings on its own at some point in the future, but we don't have any current plans in place to make that happen.

With FAST named events like "647 Spanish Civil War$c (Spain :$d 1936-1939)" is the idea that you would also add an additional FAST heading for the geographic heading alone? Are you also supposed to add one for the time period alone?

Right now what our algorithm does is it would create the 647 field for the Spanish Civil War. It would create a 648 field 1936-1939. It would create a 651 field for Spain, and it would create a 645 field for history. Based on the original form of that heading, which I believe does contain the subfield $x history and the heading itself.

But if they were adding FAST headings to the record, though, they could add those separately or not if they wanted to, it's all depending on their preference?

Right, if you're cataloging and you want to bring out Spain or a city in Spain, you could add that as a geographic. What we've tried to do is represent each of the terms in the original cataloging. But you're welcome to choose whichever facets that you want to add to your records.

When a book is about Canadian dogs and United States cats (sorry for the silly example), how do you manage to not have the record show up when searching for US dogs or Canadian cats?

And that's one of the downsides of having a faceted vocabulary. What you're trading is the recall for precision. We're giving up precision to have greater recall. So, there isn't a way. And that's one area where the LCSH works better for these headings than FAST does.

Status of the FAST Converter tool?

Current URL https://FAST.oclc.org/lcsh2FAST/

Apologies if I missed this, but if named events include the location and dates(s), why don't conference headings include that data?

Named events do not include the location and date. Those are actual separate assets. So, the location would then be put in a 651 field, and you can find the dates in 648 and conference headings are the same way. They do not include the location or the dates as those are considered separate facets.

But we do have in the subfield $c the geographic location of a named event, and in the subfield $d we have the date of the named event. And the reason we do that is because there's usually not recurring events, for a named event. The reason we don't do it for conferences is because when we did our initial research we found that the
number of Conference headings used as subject headings were small enough that to indicate just the conference name itself was usually sufficient. And then you could add the date as a separate field to identify the conference heading. We are looking into whether we do need to add more information and this is based on changing LC their
formats of how a conference setting is constructed because back when we started FAST the dates were part of the subfield $a and often the geographics were part of the subfield $a, and then as LC has started at pulling that data out and putting it in its own subdivisions now we're losing information that was originally part of the subfield $a. And so, we have been looking into what information we need to add to differentiate the various meetings used as … and keep in mind that we're talking about meetings that are used as 151 headings in WorldCat. So, we do know that for some very frequently written about meetings are 111 headings, 611 headings, that we do need to revise or revisit that question. But for now we encourage you to add the geographic location as a separate facet and the 688 chronology is a separate facet to help identify the specific meetings. identify the specific meetings.

Why is "History" considered a genre/form heading in FAST, when it is not a genre/form heading/subdivision in LCSH?

This is again one of those conversations we had at the beginning of a FAST and whether, I don't know if it's more that history is a form genre, or whether the topic is history itself. And so, history is a subfield $a in a 650 heading, but the work itself when history is included is of a historical aspect I think you would say, and so FAST chose to convert subfield $x history to the form/genre heading.

Have you found that FAST lends itself better to some subjects than others?

I think that's really hard to answer, because we are not people that are assigning FAST. We are utilizing, here at OCLC, the terms, and other people assigned to then convert into the FAST headings. So, I think that question would need to be posed to people that are assigning FAST originally.

What is the status of updating J level records in OCLC?

There are currently a little bit fewer than 100,000 Encoding Level J records in WorldCat. We have a project underway to convert all of the alphabetic encoding levels to the MARC Standard in America Encoding levels. So, ELvl: J is one of those that we need to tackle and that we haven't tackled yet, but we'll get to it.

How do multi-level geographic heading URIs work? Is that sort of precoordinated, or would the URI be associated with the most specific geographic entity?

If I understand your question each heading has its own and so if you choose a multi-level geographic heading, for example the one that Shanna used for South Africa, there was a 3 level heading, then it has its own to be associated with that record or that heading. If you go to the search FAST page and look up any geographic headings in the FAST record themselves you can see, it's in the 024 field. The URL for the heading itself, or for the record.

Is there any plan to include Chronological index in the assignFAST API (https://www.oclc.org/developer/api/o...n-FAST.en.html)?

The chronological index is really only in searchFAST, and we've established these headings to help people know the time span or the actual years for most frequently used chronology headings. And so, they are there for more for a reference person than for example for controlling a heading. And so, I'm not sure of how it would be of help to have it in the assignFAST API, but if you could share that information, maybe send an email to FAST@OCLC.org and give us a little more information about that we'd certainly consider it.

I had another thought about the chronology headings just so everyone understands a chronology heading in FAST is made up of just the years. We don't have the descriptor terms, like, early modern or Elizabethan or things like that. So your years can be as simple as 1986 you could just create a 648 for 1986 or you could do for 1986-1987,
the only requirements are that the 1st year be earlier than the 2nd year.

Isn't "20th century" actually 1901-2000?

We also had that debate at the beginning of FAST and we decided that most people, and I think we were thinking in terms of the public, consider the twentieth century to be from 1900 to 1999, we've had discussions about whether we should say plus or minus a year, things like that. But FAST went with 1900 to 1999 to define twentieth century.

October 20, 2022

Many ILSs cannot suppress headings at that granular a level - our system cannot suppress headings based on the value of $2.

You may want to talk to your system people and make that request for an enhancement.

Is there/will there be an automated way to add FAST headings within Connexion to original/newly created bibliographic records so we can bring them along right away when we export to our system?

That's certainly on our wish list. We don't have any actual development going on to make that happen right now.

Most likely it won't be possible in Connexion at least not in the foreseeable future. But something may be available in Record Manager, soon.

When editing a bib record with LCSH and FAST ... if I edit the LCSH should I also edit the FAST or leave as is or delete all FAST headings to trigger the repopulation of FAST in the record?

If you're going to be editing the headings that are already in the record, we suggest that you go ahead and delete the existing FAST headings that are there because that way, since the record will no longer have any FAST headings, but it has LCSH, the automated process will pick it up and add the FAST headings back to the record.

My memory from the June session was we were told that if we made changes to the LCSH (perhaps first-time CONSER authentication) that we could delete the existing FAST headings as they would be regenerated automatically at some point. But what about any libraries that might only be submitting FAST directly? We wouldn't want to delete these if revising LCSH?

You don't want to delete FAST that have been assigned directly from a library. But if there are LC Subject Headings in the record, you are not going to be able to tell that. If there are FAST headings in the record only then yes, they have been assigned by a library. If you are adding LC Subject Headings to a record that has only FAST headings and you think the FAST headings are incorrect, you could certainly delete them and then the automated FAST process would come along and add new FAST headings or re-establish FAST headings based on the Library of Congress Subject Headings you have added to the record.

If I create a new OCLC record and have a mix of LCSH and FAST, what happens?

What would happen there is when you've created the record and you have both LCSH and FAST headings in it our automated program for generating FAST will not touch your record. Our automated program only touches records that have no FAST headings in them, but have a Library of Congress Subject Headings in them. And we suggest using the FAST Converter to help in choosing FAST headings if you are planning to assign them yourself.

Can you give an update on how new FAST headings may be created for concepts not present in LCSH?

We don't have a process in place right now to do that. We will create headings for any type of names that are in subject headings but that wouldn't be in Library of Congress Subject Headings, but we don't do that for topicals right now. There is something called the FAST Funnel which just got started this year. That's part of the SACO Program with is part of the Program for Cooperative Cataloging or PCC. The FAST Funnel allows people to suggest new Library of Congress Subject Headings, topical or otherwise, that could be Then submitted to the Library of Congress on behalf of FAST to be added to Library of Congress Subject Headings, or for revisions to Library of Congress Subject Headings that would then filter into FAST when the Library of Congress Subject Heading gets established. So, that's one possibility for how new headings can be established. Also, we do mine WorldCat to look for LCSH headings to add, based on literary warrant.

If things are unclear with when to delete and when not to delete, would it be ok to just contact bibchange as we would for other questions?

You're welcome to do that. We'd be happy to look into it and see if we're able to figure out what one should or should not be deleted.

How often is the FAST database refreshed to pick up new LCSH?

FAST is synchronized with LCSH each month.

We're seeing quite a lot of DEI-related changes made to LC Subject Headings. When such changes are made, can you walk us through how the corresponding changes are then made to FAST headings in the affected bibliographic records?

When the changes are made, if it is a one-to-one change, those changes are then made, when the FAST database is updated on that monthly cycle. So, even if the changes are made to the LC headings one month through the Controlling and the updating controlling process, it may take another month or so, depending on timing to get that change made in the FAST headings. I know that when the heading change was made at the end of last year beginning of this year to illegal aliens that was split into 2 headings. And that required manual intervention on the part of our staff here at OCLC or at LC in order to make the changes to FAST.

The other changes that are a one-to-one change, those are part of our natural update process. And so we get the records from LC at the middle of the month now and then by the middle of the next month we start making applying the changes to WorldCat Records.

Is there an effort or work to "translate" FAST to other languages, let’s say Spanish?

I don't know about Spanish, but I know that the University of Laval has done some work translating into French. I know there have been other libraries. We had one library contact us that was working to translate some of the FAST concepts into, I think it was Greek. But we haven't heard from that library in several years, so I'm not sure where that project went. But I think they did a lot of work on geographics, converting those to French. I am not sure on the details or specifically what was translated, I just know that there's an RVM FAST and it contains 381,782 authorities. But that's not something that OCLC does. That's usually something that other libraries do.

How different or similar is FAST to what was called PRECIS?

Yes I remember PRECIS but I've never used it so comparing it to FAST it would be very difficult to say how different or similar it is. But just remember that FAST is based on Library of Congress Subject Headings. So, that's where the origin of FAST is in terms of where the terms come from. And I want to emphasize based on, it's not exactly the
same. Some terms are translated differently because when you take them out of the Library of Congress Subject Headings order they don't make sense unless the wording is changed a little bit.

From user: PRECIS was faceted terms within a single field, I believe.

LCGFT shows in my catalog as a subject, should I expect the same with FAST?

Without knowing the specifics of your catalog my best guess is yes, because the FAST headings are coded as 6XX fields in the MARC records. And so, I would assume they would show up as subject headings in your catalogs.

Do you have any sense of how many records in WorldCat have FAST but no LCSH?

Records with only FAST and no LCSH and no LCGFT = 634,700 [as of this recording]