Recently conversation erupted on RDA-L that was initiated with a question about a subjective statement in a MARC Bibliographic tag 300 field (specifically the subfield b). Basically someone had input “illustrations (some coloured, all beautiful), maps ;” in said subfield. The subjective bit is the “all beautiful” and the question was regarding if RDA allowed for subjective comments such as this within the MARC record. A side discussion arose regarding the spelling of color/colour. Much back and forth occurred and you can read about it yourself on RDA-L, OCLC-Cat, AutoCat, etc.
Which is why in an ideal world, if we care about whether the illustrations are colored or not (and I suspect the time is LONG gone when our patrons or we actually DO), there would be a data element in the record which marked, in a machine interpretable way, whether there are illustrations (checkmark HERE), and whether they are colored/coloured (checkmark THERE). Which could then be translated to the appropriate spelling or even language for the given audience. [more to read in the original post]
Exactly! The data element should be in a machine interpretable way! The programming should be able to output as we required. This is what Karen Coyle and others have been saying for some time now. The majority of cataloging should be data entry-like.
WHAT???? I am saying the basics should be basic data entry. It will still require knowledge to input the data as well as determining call numbers, subject headings, etc. But the basics should be check here, type an arabic number there, etc – you know, like filling out a form online. This would be part of parsing out all the data into different fields or subfields or data areas or whatever-you-wish-to-call-it.
Why do this? It promotes consistency. It promotes easier sharing. It promotes tons of options for searching the data.
One of the reasons for RDA is to separate AACR2 from MARC Bibliographic. It is said to move ahead in the semantic web world, we need to separate the rules from the format or input mode. Yes, we do. AACR2 and MARC were ‘married’ in the 1970s. It is time for the D-I-V-O-R-C-E. I do wish we’d moved away from MARC first before killing AACR2 or, ideally, done it at the same time. Yes, I think we’d still have tons of issues and discussion and perhaps arguments but trying to force RDA into MARC Bibliographic is like trying to play baseball with a badminton racket.
It’s been, oh, about 4 years since I first heard of RDA. I read, I reviewed, I tried to comprehend. The ideas of FRBR came naturally to me, learning the terminology took a bit longer but I understood the concepts. RDA, not so much.
I disremember what I read initially but it has taken me four years to really understand. The idea of RDA is really to change the vocabulary and the way we think of cataloging. That is, instead of looking at it from an item perspective and starting with format as we do with AACR2, with RDA we look at the individual bits of data (name, title, etc.) and link them together. It is sort of cataloging by reverse engineering. I wish I could have seen more clearly 4 years ago but it took hearing from Shana (comments on Why RDA then in why RDA, revisited) as well as recent discussion on RDA-L, NGC4LIB, etc. which explicitly stated this – thanks everyone!
So, I understand RDA – at least, I think I do. The actual changes (such as rule of three, GMD, etc.) are not tremendous and truly, as things are, will not make a huge impact (other than work load perhaps). However, the idea is the future – when MARC21 is gone and we use a new input; when the ILS begin to program to the new rules/ideals; when other industries might also pick up RDA and use it as a standard – this is when the real change occurs.
The very difficult thing right now is trying to force new concepts into the old molds. Pushing and shoving RDA into MARC21. Trying to get the ILS to change, drastically change, the way they program their cataloging modules. Having other communities (Publishers, Museums, etc.) adopt and understand the RDA rules.
It seems to me, to make this whole thing work – EVERYONE has to buy in and make the change. The format too – MARC21 has to go. Will that happen? I don’t know. The world’s largest bibliographic database is wedded to MARC21 – can that be changed? How much will it cost? How much will it cost us all?
Two blogs posts I read today, plus a comment on one of my posts made me think again. Oh, and the CRCC RDA informal test results – they also peaked my interest.
So – let’s start with Diane Hillmann’s Irresistible Apology of the day. This is part of Diane’s report from her attendance at ALA Mid-Winter this past week. I quote:
I still think that it’s hard to justify the time and expense of the testing that has just concluded, which tests RDA only as used in a MARC environment, not RDA itself.
OK, what? I thought there were test records made using other ‘containers’ than just MARC. So I checked at the Library of Congress RDA test records site, and find the test records are indeed almost all of MARC. Hmm. This changes my opinion. I had thought the test was also testing in other ‘containers’ as the formats have been termed. If RDA is meant to separate the rules from the format … should not the test incorporate as many of these formats as possible? Otherwise what is the point? How is that a valid test for the intent of RDA? Can anyone give me more examples than the 11 MODS records on the Library of Congress RDA test records site? Was a similar test done in Dublin Core? How about RDA work with EADS? What about MARC in XML? Or ONIX? I have to agree with Diane’s statement:
The result of this from the point of the community has been useful insofar as it has provided an avenue for some initial training and participation, but not so useful from the point of view of really providing any understanding of RDA implementation.
Now look at What’s the point’s RDA and OPACs. Again, I quote:
I think RDA is looking into the future and predicting what we will all want and trying to make provisions for it. We (some of us, including me) criticise RDA because it neither sticks with the standards we’ve already got, nor offers anything our present OPACs can make use of in any kind of a helpful way.
I have to agree again, currently RDA is not being used by any Integrated Library System out there – not to my knowledge. From everything I read and see, the vendors are waiting to see what we want. The latest cataloging modules I’ve seen are all still written so that the cataloger (or clerk or whatever) has to have an intimate knowledge of MARC, not AACR2 and not RDA but MARC. I have the same question as What’s the point”:
What do we want, really really want – something that used to work, something that works now or something that might work in the future?
Finally, James Weinheimer commented on my post Why RDA, revisited.
I still maintain that the RDA folks must demonstrate the business model that will show precisely how things will change for the better: what will libraries get from the changes? I haven’t seen anything at all convincing yet and after all, we’ll still be stuck in the “horse-and-buggy days” of transferring MARC records in ISO2709 format! Why not change that first?
I agree. I think replacing MARC has real potential for proving a difference and demonstrating the benefits of change. When we force MARC to try to do what RDA is intending … I think failure occurs. MARC requires quite a bit of handling to making the desired linking/FRBR-like concepts (that RDA is based upon) occur. And please note, I have been of the camp stating MARC is fine but I finally saw the light after a Karen Coyle talk.
After they demonstrate the real advantages of RDA in concrete terms that all can understand, everybody could begin to discuss it, do some research and ask the various groups: reference librarians, selectors, and yes: even the users themselves. Then we may be able to figure out if it is worth the expense and disruption during one of the most difficult moments in the history of libraries–that I can remember, anyway.
You’re right of course James. The actual physical benefit is not clear. I am now at least understanding RDA more – that it is a restructuring and re-wording of AARC2 to try to engage and communicate with Systems and others (that is, not just ‘catalogers’). That RDA is meant to be more flexible than AACR2 as new formats come around is clear. But what is not clear is the benefit to the libraries right now – and let’s be honest, now is what matters to budget.
The results from the CRCC informal test seem to support that view. There are positives but, perhaps it is my reading, it seems the negatives are stronger. If nothing else, the idea seems to be RDA is not ready – not yet anyway.
Shana (see comments on my why RDA post) wrote an erudite and thoughtful post on why we must move to RDA. She said,
In a sentence: RDA is fundamentally different in it’s approach to describing materials/resources/things/[insert pithy term here].
then went on to explain what she meant. Basically, the big switch for RDA is to reformulate or restructure the rules into a format-neutral structure. That is, instead of referring to “chapter X of AACR2″ when you are dealing with a book or sound recording, you would deal instead with the actual elements involved in the description. Whaaaatttt? RDA is looking more at the individual things that build up the record, such as looking up the title element or the author (main/added entry) element or whatever rather than each chapter devoted to a format such as Monograph or Sound Recording.
Shana and I continued this conversation via email, if I get this wrong please blame me. I asked if my interpretation above was a fair description. She said it was and we discussed the idea that RDA is the description of what you put in the “container” (such as MARC or Dublin Core, etc.). The idea, Shana explained, is to separate out the rules from the container. That would be like [my example], a stop sign is a stop sign regardless of if you are driving a car, a motorcycle or a bicycle. It still means STOP.
Going back to RDA, we have some relatively minor changes such as the rule of three, abbrev. and such. The big switch is to change the format and the language of cataloging. J. McRee (Mac) Elrod has written a fantastic guide, Major RDA Changes From AACR2 By MARC Field, about the specific changes.
I also read Jonathan Rochkind’s post RE: Straight Jacket? on the RDA-L list serv. This makes sense to me. Jonathan addresses the way “MARC serves as our ‘data vocabulary’ and even our rules for entry in many cases come either from MARC itself or are formulated in terms of marc fields“. Exactly. Jonathan points out that the switch to RDA is an attempt to separate out the data vocabulary and entry guidelines. Basically, pulling the two apart so we can restructure.
And, I listened to the ALCTS session “FRBR as a Foundation for RDA” by Robert Maxwell. I do understand FRBR (or so I think) and this was a jolly good reminder of the ideals of FRBR.
OK, so I see that RDA is a re-structuring of AACR2 to incorporate the terminology of FRBR. This I have understood. I see there are very minor (to me anyway) changes to the actual rules of AACR2 within RDA.
I can understand the need for the new vocabulary. I can see Systems and Programmers and Catalogers all need to speak the same lingo (and have ranted on this before).
So. We are restructuring the language of our rules (from AACR2 to RDA), but we are not significantly changing the rules – or so it said. Is this a fair statement? The rules themselves are not leaping away from what we know today but the language we use to describe the rules has jumped? That is, we are perhaps switching from using Mandarin to Swahili?
Hmm. I guess I’m back to costs and access to the rule then. If I agree (and I am still teetering on the fence folks) with the switch from AACR2 to RDA, I wonder about the costs involved. I do not just mean the costs of the “toolkit”. I do understand ALA’s need to recoup the costs of production and maintenance of this toolkit. I can’t imagine the gnashing of teeth that occurred to develop the pricing structure (whether I agree with the final sum is moot, it is incredibly difficult to place a price on a brand new and controversial item).
Costs include training and retraining. Conversion of old records. Conversion or change of ILS. And more that I cannot fathom at the moment.
Conversion of old records? One of the things being stated is that AACR2 can co-exist with RDA. That makes no sense to me. If they can co-exist and work fine, then why switch in the first place? Shouldn’t at least the major elements be converted to the new? How will the GMD ‘exist’ with the 336, 337, and 338? Wouldn’t the inconsistency drive the patron nuts – or just me? And wouldn’t the search results be greatly affected?
Conversion or change of ILS? Won’t there have to be changes made to the existing ILS to accommodate the needs of RDA? And will those be part of the current service plan or an update or cost the library $$$?
So – after all this, what do I think? I am now leaning on the RDA side a bit. I can see the need to change the vocabulary. The restructuring of the rules to take the format out of it makes good sense to me. Pain in the posterior to readjust how I look things up? Why yes! I do still think that perhaps changing MARC first or in conjunction with the rules would have been more beneficial (and an even bigger pain in the posterior). I still worry about the costs involved. How can this be brought down to manageable for the majority, especially in the economy of today versus when AACR to AACR2 happened?
Oh – and what exactly are the ILS doing? I know VTLS has been on board since the beginning (and have very much enjoyed their presentations) but what of the others? And what about Systems people? Are they involved in the new vocabulary? Do they understand it better than the old? And how does this make the linking possible that is desired? Oh, that’s right, it’s the first step. What’s the second?
On the OCLC-Cat list for November 2010 there have been several posts of interest. Let’s look at some, shall we?
First, let’s look at Wojciech Siemaszkiewicz’s ”November 2010 Memorandum Against RDA Test” from November 2, 2010. In this post, Wojciech instructs OCLC:
Immediately suspend coding the test RDA records as acceptable records and recode them as substandard records with a code “RDA” (no PCC, LC, etc. coding should be allowed on these records). The encoding level for these records should be “K”, which usually triggers a full review of the record by highly trained technical assistants or professional catalogers. The LC records should be coded as level “7”. The RDA test records should be treated the same way as records coded with Spanish, French, German, etc. codes. This would allow catalogers to create parallel records for 040 English records according to existing and widely accepted AACR2 rules. Under no circumstances should RDA testers be allowed to create conflicting NAF or SAF records in LCNAF or LCSAF. This has already created a great deal of confusion and has been universally rejected by catalogers involved in the discussion.
We instruct agencies responsible for the RDA test to instruct its testers to follow above mentioned rules as a way to avoid workflow complications and growing confusion in libraries around the world.
We understand that the RDA test is just a test and in no way is an indicative to a future cataloging procedures and rules that would replace universally accepted AACR2 rules
Subsequent posts include a November 3, 2010 “Petition to support Wojciech’s memorandum” created by cataloging managers at Indiana University, Bloomington. The link to the petition is http://bit.ly/noRDAtest.
The discussions on both threads have been fascinating. This culminates in a November 16, 2010 post “Update on the Petition to Support Memorandum on OCLC’s RDA Testing”. Another very interesting read.
On November 17, 2010 a post appeared cross-posted to several lists “US RDA Test and OCLC”. This does not directly address the Memorandum or Petition postings but certainly it seems to be aimed at them.
What I find interesting, in reading all of this, is that the idea is no duplications should be allowed in the OCLC database. I understand and applaud this. However (you knew that was coming, right?), why is it considered a duplication when we already have “parallel” records?
Parallel records? What’s that?? Check out OCLC Technical Bulletin 250: Parallel Records. Starting back in October 2003 OCLC began to allow “parallel records within WorldCat by language of cataloging”. Huh? OK, let me break this down for you. If I, in my little library in anywhere, USA, catalog the latest James Patterson using my AACR2 (or whatever) rules then add it to the OCLC database as the first bibliographic record ever to describe the book then it exists and others can use it, right? OK, now a few days (or weeks) later a cataloger in Spain catalogs that very same James Patterson book and also adds it to the OCLC database. Duplicate? NO!! Parallel record!
Identifies the language for those portions of the record, which according to cataloging rules, appear in the language of the cataloging agency (e.g., notes). Subfield ‡b is in records created by libraries for which English is not the language of the cataloging agency.
Huh. So basically this is to accommodate the different rules and language of different countries. Why then can we not have AACR2 records and RDA records?
Perhaps I’m missing something but that seems to be the same concept…or am I being blonde again?