Ben Fri 3/18/2011 2:54 PM
Following CCM 16.4.1, a serial that changes its physical medium from print to CD-ROM needs a new record. Since CSR guidelines (cf. CONSER Standard Record Metadata Application Profile (MAP) 07/22/2009: 130 field) tell us no longer to add a uniform title to distinguish among manifestation-level differences between titles, should we include information in the linking field that indicates the subsequent (and/or former) title is a different format? And if so, how? In |n (notes) or |m (material-specific information)?
Or is this a case where we are better off just not adding information about the subsequent title? |
Ben Fri 3/18/2011 3:04 PM
Just to clarify my previous post, by "Or is this a case where we are better off just not adding information about the subsequent title?" I mean information about the fact that it is a change in format. I did not mean to suggest not adding a 785 at all. |
Kevin Fri 3/18/2011 5:04 PM
I think it's very useful to include some indication of the format change in the linking field (or the 580 field if the linking field is suppressed).
Perhaps it could be as simple as using subfield $i, e.g.
785 00 $i Continued on CD-ROM: $t title |
Kevin Fri 3/18/2011 5:04 PM
Renee, you are correct that 2nd indicator 8 would not be used in 785 (or in 780) because that value was defined for something else a long, long time ago before subfield $i was created. However, while subfield IS defined in MARC 21 for 780/785, there is no guidance about how to use it "properly". It just says you don't use 2nd indicator 8. In most systems, I have the feeling it's going to generate an OPAC display like:
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="d1e83ff5-2378-42b8-8e97-912a4119f107"><ac:plain-text-body><![CDATA[Continued by: Continued by the CD-ROM version: [title]
]]></ac:plain-text-body></ac:structured-macro>
which looks kind of weird. An intelligent system should know to suppress any defined display constant when subfield $i is present. But how many of us have "intelligent" systems? :/
Sorry I didn't think my original response through very well... |
Ben at Mon 3/21/2011 3:05 PM
Kevin,
" Continued by: Continued by ..." This is exactly the result we get in our system when using a 785:00:$i. Yes, it looks weird. But I'm going to hazard a bet that the intelligence of our users can compensate for the lack of intelligence of our present OPAC.
In any case given the weirdness of the situation I decided not to add it to the national record.
Thanks again for the fruitful discussion, |
Renee Sat 3/19/2011 6:27 PM
I'm glad that you brought this up Ben, because the more I work with titles that change format, the more I wonder whether there is a need to discuss a potential change in policy.
Yes, the new format requires a new record, but I wonder about the relationship between the two and if a more forward looking way of describing that relationship shouldn’t be considered ---particularly in light of all of the digitization efforts we see going on.
I feel certain that all of us have come across titles that have ceased in print, “continued” only online, and at some later date, find that the publisher has retrospectively digitized earlier issues of the title. Even though there is a loss of “one to one correspondence” between the two formats at the point the title continues online only, wouldn’t we still describe that relationship using a 776?
Here's something to think about:
Perhaps, right from the start, the relationship of the print to the other format should be described as just that: another format -or not at all- or 787. Clarify the situation for users by adding an appropriate complexity note and include 776 #g <date-> allowing for the potential addition of earlier issues digitized at a later time.
In addition to alleviating that amount of record cleanup required each time the scenario I initially described above plays out, it may also help lessen the amount of time a cataloger spends trying to sort out relationships. For instance, I’ve come across records where the cataloger has clearly been conflicted; that is to say, not only is there a 785 in the record, but also a 776 pointing to the same resource. And yet other records where the later format has a 780 pointing back to the print and an additional 780 pointing to the print version’s earlier title.
As I said, just some things to think about, |
Regina at Sun 3/20/2011 9:31 AM
Renee,
You make some very good points. The primary relationship is the 776 relationship, I think, and that is the one that the linking fields should reflect. The use of the same title and ISSN in more than one 7XX link causes problems for U.S. ISSN records in the international ISSN Register so we are instructing ISSN staff to prefer the 776 link in these cases. |
Ben at Mon 3/21/2011 11:46 AM
Renee, et al.,
Thank you for the responses. While I tend to agree with your points about the potential complications of treating a format change as a title change, presently (at least at MIT) our OPAC displays 78x fields in a way that makes them more "navigable" by users trying to find the current record than the 776 field does. Though 787 might be an option.
Having tried out Kevin's suggestion of using |i on our test server, and gotten a pretty good result, I am tempted to go this direction... Now I'm just wondering if I should just make this edit locally. Does anyone see a reason why I shouldn't add the |i to the national record? |
Everett at Mon 3/21/2011 12:00 PM
Ben and All :
Hi – I agree with Kevin's proposed solution below.
I like that it is short and concise. Kevin's proposal also furthers the elegant principle included both in RDA and in systems-based logic that, for as long as libraries are wedded to the MARC 21 formats as a Communications/Data Exchange standard, we are far better off including related data in the same MARC fields than in dispersing it among separate fields (i.e., 5XX and 7XX fields, etc.)
cheers, |
Renee Mon 3/21/2011 12:10 PM
Ben,
For linking fields using $i, the 2nd indicator would need to contain value 8 (No display constant generated) and for field 785, value 8 (in some systems) would generate the note: "Changed back to".
So, I'm not sure putting it in the national level record would be advisable at this time. |
Renee Mon 3/21/2011 12:34 PM
can't find the rule at the moment, but the same holds true for cataloging in general; that is, only one linking relationship is to be established between 2 given resources-- the relationship that is the prominent one. |
Steve at Mon 3/21/2011 1:08 PM
That was never a "rule" but there was a CONSER practice documented in the CEG (we abandoned the practice at some point to allow both 776 & 780/785 links to show the simultaneous relationship of 776 & 780/785...I think this is still documented in CCM Module 31). Personally, I find it disappointing that we have to abandon the practice in the wake of ISSN Network issues, but on the other hand, I think it makes sense from a FRBR perspective to no longer code this situation as 780/785.
Typically (and I put big emphasis on the fact that this situation is far from universal), we consider a major change in serial title to be a new work (or at least a new description is created and in RDA all the rules about authorized access point, etc. kick in). In the librarian's ideal world (PIE-J, Project Transfer), the serial title changes across all the formats at the same time, so we can pretty clearly define which expressions and manifestations are associated with each serial work. I know that 780/785 is relating AACR2/OCLC records (which predominantly describe manifestations) but generally one can look at 780/785 as an indicator of the existance of a new/different serial work.
Or put another way, when print ceases in favor of online, a new/different serial work is not expressed. The print manifestation ceases, the online manifestation continues...the work itself does not cease. Obviously there is a 776 relationship between the online and print manifestations. But if we generally figure that 780/785 is more reflective of a change of work (which it really is when there are no record/control numbers in the linking field so you're not pointing to a specific manifestation), then it's not necessary to use 780/785 in the case where print ceases in favor of online and there's no change in the serial title.
Just my two cents. |
James at Mon 3/21/2011 1:27 PM
May I add my two cents to Steven's cogent two-cent remarks?
I have never been very comfortable adding the 785 field in such cases.
I think the 776 is enough (along with the cessation note). |
Renee Mon 3/21/2011 1:57 PM
Thank you, Steve, for refreshing my memory re where the linking field information is documented-- now I can go to lunch without driving myself mad thinking about it. Apparently, I also missed an update somewhere along the way. |
Everett at Mon 3/21/2011 4:52 PM
John :
Hi – thanks for the quick statistical count.
I suppose the follow-up question regards whether this change would be seen as valuable in terms of the CONSER database, record consistency & maintenance, etc.
I know just thinking about it quickly, I find the idea of being able to encode relationship information in 780 & 785 fields via second indicator 8 and subfield $i as I am able to with 776 and other Serial Linking fields, to be an attractive proposal.
As noted earlier, keeping related data in a single MARC 21 tag is certainly more logical and straightforward from a systems-logic perspective. I believe this is one of the directions that the new RDA Cataloging Code is trying to move us in as well.
cheers, |
Mon 3/21/2011 5:14 PM
Robert,
Good suggestion. Thanks! |
Regina Tue 3/22/2011 8:44 AM
Steve, I always appreciate your two cents!
I want to address your disappointment at having to "abandon the practice in the wake of ISSN Network issues." Perhaps I did not elaborate enough on my comments to indicate that I thought there were other good reasons for aligning CONSER practice with the ISSN Network approach. I try not to advocate for ISSN Network practices when I believe the ISSN Network practice is not in the best interest of CONSER. In those cases I try (not saying I necessarily succeed) to advocate for making the change in the ISSN Network practice because I believe it is in everyone's best interest that ISSN and bib records align to the extent possible. In this case, I think there are good reasons, e.g., the maintenance issue that was mentioned, and the nature of the relationship (not really the same relationship as a major change of title as you have explained, Steve) that call for at least an examination of what should be our best practice in this case. I think (but maybe Robert Bremer can better inform us) that some uses of linking relationships, e.g., xISSN, Linking ISSN, FRBRized displays, may be confounded by the situation of two simultaneous relationships like 785 and 776. Just my 2 cents! |
Steve at Tue 3/22/2011 11:04 AM
Trust me Regina...this is a disappointment I can easily get over  I (along with everyone in CONSER I assume) appreciate the fact that you do the best job you can of balancing the needs of the two programs which can sometimes be at odds. I've mentioned why I think it's in our best interest long-term to tag this relationship as a 776 (and not use 780/785)...I think the additional examples you've given provide more fuel for the fire.
Thanks much. |
Les at Thu 3/24/2011 8:20 AM
I'd like a small group of volunteers to propose a practice for linking records that reflect a format change, identify the places in CEG and CCM that need to be updated, provide wording, and examples. Depending on the scope of the proposal, maybe there are just a few places in the documentation that need to be changed.
We can modify and/or approve the proposal at the operations meeting.
Volunteers? Please let me know by Wednesday March 30. |