February 2009 Archives

The following is the report that I submitted to ALCTS, CCS, etc., on my activities as ALA Representative to the JSC between June 2008 and January 2009:

Content development for RDA: Resource Description and Access is winding down, and the product development is gearing up.  In addition, now that we have a complete text for RDA, implementation planning can begin.

Since my last report in June, 2009:

In August, RDA was presented at a very successful IFLA satellite conference at the IFLA Annual Conference in Québec City.  Both the code and the RDA online product generated considerable interest.

The Joint Steering Committee worked through a number of remaining issues, completing consideration of all of the priority issues from previous constituency reviews, and forwarded the content for the full draft of RDA to the developers near the beginning of September, 2008.  The full draft was finally released on November 17 for a brief constituency review.  It was not possible to issue the draft in a pre-release version of the online software, as we had hoped, and the process for producing the text for the PDF files released for the review was far from smooth. However, the draft is now out and is being reviewed.

The JSC will be meeting from March 12-20 in Chicago at ALA Headquarters. At that meeting, we will make decisions on any outstanding issues, specifically those raised in the responses to the full draft. We plan to turn over as final text to the developers early in June 2009. At the March meeting, we will also begin to consider how to maintain RDA, including how to deal with a long list of issues that we had decided we could not include in the first release of RDA.  We expect that there will be substantial updates to RDA in the coming years.

Product development:  A prototype of the RDA software was demonstrated at the ALA Midwinter meeting.  The Co-Publishers expect that RDA will be released during the third quarter of 2009.  They have begun to release some preliminary information about pricing and licensing, which includes both one-time and subscription prices.

RDA testing:  The Library of Congress, National Agriculture Library, and National Library of Medicine will be testing RDA once it is released.  They are looking for about 20 institutions or individuals to join them in the test.  Preparation for the test (training of participants and practice work) will take about three months; the test will take another three months; after that, the results will be evaluated and announced.

 

John Attig

ALA Representative

Adam Schiff, chair of Examples Group II sent me some comments on some of the ALA comments in our response to the full draft, and asked me to share them with you:

Form of work (6.3) is required only when needed to differentiate a work from another work with the same title, which is not the case here. It also seems redundant to apply in this case, since the preferred title is "quintets" and the form is "quintet". This is an instance when RDA haphazardly veers between calling for explicit recording of data previously inferred from context and retaining those omissions, presumably in the interest of legacy data.

This example would also be more helpful if it were to include the construction of a preferred access point.

Adam's comment:

In this example, the form of work is required to distinguish a variant title for the work from the preferred title for another work by the same composer.  The form of work also is needed to distinguish other variant titles for the work from variant titles for another work by the same composer.  The only way to include the qualifier (Quintet) in the variant access points (see the three 400 fields in the MARC example on p. 83-84 that include this qualifier) is to have recorded this qualifier in the Form of work element.  Schubert also wrote a song titled Forelle, and that song is also known by the titles Truite and Trota.

So in this particular case, while form of work is only required when needed to differentiate one work from another work with the same title, it is recorded here (optionally) to distinguish one work from an identical variant title of another work.

9.2.2.5.3

9.2.2.5.3 b), examples Li An (p. 12) and Ang Lee (p. 14): There is a director whose name is established as "Lee, Ang, 1954- ." Users of RDA aware of this director will believe the examples misrepresent the instruction. A further note should be added to clarify that the examples are referring to the Chinese author Li An.

This example actually is for the motion picture director.  I do not see how this example misrepresents the instruction.  Ang Lee the director has made Chinese language and English language films. The systematically romanized form of his name is Li An.  The well-known name established in English language reference sources is Ang Lee.  The fact that the authorized heading for him has a date in it, is immaterial to the examples here.

0.10:  "Specified elements" refers to the lists of transcribed elements vs. recorded elements and the different ways in which things like numbers and dates are treated in RDA.

0.11:  This instruction allows for alternative vocabularies -- which is what allows use of both codes (treating the MARC code lists as alternative vocabularies) and terms in languages other than English to be recorded.

Wow, those were my only comments on the Introduction.

The first draft of the response is not complete; editing and fine-tuning remain.

2.7.2.6, etc.: The instruction at 1.4 does in fact address the language and script for supplied data.  If you are supplying data in addition to a transcription, you follow the language and script of the transcription; if you are supplying the element, you choose "the most appropriate language and script".

2.9.2.6, etc.:  I omitted a number of comments about what to do when information is not identified.  These seem to me to be based on assumptions about the interrelation of the production, publication, distribution, and manufacture statements which RDA does not assume.  Just because you have publication information, that does not prevent you from recording (for example) "Place of manufacture not identified"; in practice, you probably wouldn't want to, but RDA does not -- and should not -- forbid doing so.

Chapter 3, Describing Carriers

3.4.5.2, 3.4.5.10, 3.4.5.14, etc.: The JSC did discuss instructions for scrolls, rolls, accordion-style folding, and other formats common in East Asian resources. We ran out of time before we had reached final decisions, but it was felt that the information provided by ALA formed a basis for the necessary instructions for the extent and dimensions of such materials.  ACOC agreed to compile a discussion document, and I am assuming that this will either be included in their response to the full draft or prepared for discussion at the JSC meeting in March.  I am confident that this will be resolved, and have not included any comments in the ALA response.

3.12.1.4:  Shi Deng (for CEAL) proposes adding some East Asian binding terms.  This is new to AACR2/RDA and will require a separate proposal after RDA has been released.  It should also be considered whether these are comparable to the traditional western imposition formats or whether they are more appropriate as item-specific carrier characteristics (3.21.2); this would not require a proposal and examples could be added.

3.16.2.3:  "Record details" in RDA implies "in a note" but avoids using the term "note" as that implies a particular presentation of the data.  Mentally adding "in a note" is one of the things that catalogers will have to learn to do when applying RDA in an ISBD and/or MARC environment.

Chapter 7, Describing Content

Note that 7.17.1.3 says to record the presence of color "using an appropriate term".  This is stated in this way -- rather than prescribing any particular term or spelling -- precisely because of the refusal of CC:DA to agree with the rest of the constituencies on a standard spelling.  In this case, the examples have to say something, and it is perhaps not inappropriate that most of them follow the British/Canadian/Australian spelling.  The fact that some of the examples use the U.S. spelling should be taken as further indication that no particular spelling is prescribed.  If this is inconsistency (and it is), it is because we insisted upon it!

7.22.1: Kathy raises the question of recording both total and track timings.  The instruction seems to allow both, so you can do either -- or both.  She also raises the question of how to associate the timing with the content of a particular track.  The standard JSC answer is that you have to do this in your encoding; RDA as such provides nothing that supports such internal linkages.

Chapter 6, Identifying Works and Expressions

6.1.3.3:  I had to modify Kevin's proposed wording.  Remember that this chapter is in a section on describing works (and expressions), which under scenario #2 would be recorded in authority records.  Therefore I don't think we should talk about access points for additional persons, etc.; these would be in the bib record, and might appear in references in the authority record.  Instead of that, I just said to retain the old access point as a variant access point.  I also made a change to the first paragraph of 6.1.3.3 -- hopefully, a change consistent with the other changes.

Further thoughts on changes in responsibility or title for integrating resources:  If we take care of changes in responsibility or title in the manifestation/bib record, then we end up with two distinct access points, each of which (ideally) should be linked to an authority/work record.  In that case -- which is current practice and is what I think RDA was trying to address -- then in fact you are creating an access point for a new work.  That leaves you -- in the case of integrating resources -- with two (or more) works, but only one manifestation ... which is strange but plausible.  However, it makes more sense and is more consistent with the general treatment of changes in RDA to take care of changes in a single work record, revising the preferred access point to agree with the new information, and treating the old information as a variant access point. The alternative use of added entries and see references to provide access to variants is an old issue, but I think that what we are suggesting here makes sense in the context of RDA ... at least for integrating resources.  I'm not sure how to apply this same approach to serials.

6.3.1.3:  Shi Deng (for CEAL) again has an interesting suggestion about additional instructions on formulating preferred titles for moving image works.  The JSC is unable to consider proposals for additional instructions at this time, but would entertain such a proposal after the initial release of RDA.

Note: I'm going to stop commenting on suggestions that I believe call for additions or changes to AACR2 rules.  All require formal proposals, and the JSC does not have the time to deal with them before the initial release of RDA.

6.10:  In the text of RDA, a centered caption "Other Identifying Attributes of Expressions" should appear before 6.10.  Content type does in fact apply to expressions, not works.  The disappearance of these centered captions from the text of RDA is a problem that must be fixed. 

6.10.1.3: The instruction says to record as many terms as are needed.  If a single term does not adequately categorize a resource, then use two ... or more.  A video game is a computer program and a two-dimensional moving image (or whatever).

I have resumed work on the ALA response to the full draft.  There wasn't much interesting in Sections 4-9, so I resume with Section 3, Recording attributes of persons, families, and corporate bodies.

Remember that I'm working backwards.

Chapter 11, Identifying Corporate Bodies

I wasn't paying attention, so I have nothing to report. I'm paying more attention now.

Chapter 10, Identifying Families

John M. saw a contradiction between 10.2.3.5 and 10.7.1.3.  The former deals with the use of a hereditary title as the name of the family; the latter deals with the hereditary title as a separate element.  In the former case, inversion seems justified; in the latter, it does not.

Chapter 9, Identifying Persons

Several of the comments ask for justifications of what are in fact AACR2 rules.  I'm going to exclude all such comments.

Shi Deng (for CEAL) suggested a new provision at 10.2.2.10.3 to deal with pseudo-compound pseudonyms.  This is a change to AACR2 and requires a separate proposal after the initial release of RDA.

Chapter 8, General Guidelines

The distinction between bibliographic and authority records applies only to scenario 2 implementations.  There is a clear layout of what that means in the general introduction, but RDA should not be limited to current implementations.  The instructions in Section 3 apply to describing persons, families, and corporate bodies -- whether these are stored in bibliographic or authority records or in something that is neither.

Shi Deng (for CEAL) makes some interesting comments about removing the distinction between preferred and (at least some) variant names.  This would be a major change for RDA and I doubt is supported generally by CC:DA.  So I'm not going to include these comments, although I think that a discussion paper on these issues after RDA has been released would be worthwhile.