American Library Association
Annual Conference (123rd : 2004 June 24-30 : Orlando, Fla.)
Summary of cataloging, etc. meetings

Technical Services Directors of Large Research Libraries Discussion Group (“Big Heads”) (Friday a.m.) * chaired by Arno Kastner (NYU), Bob Wolven (Columbia) will be chair next year, and Karen Calhoun (Cornell) next * membership is the top 25 ARL libraries, based on composite rating (not just collection size) (membership to be reviewed based on last 3 years, will reconsider for next review since Stanford has left ARL) * Big Heads met with OCLC the night before and an overview will be prepared by Glenn Patton

Report on CONSER Summit: e-journals moving focus to article access; opac role in access is changing (about 75% of article access comes outside opac); many libraries providing access by broad topics and everybody’s doing it differently; most collaborations have been two- or three-partied rather than across interested parties; organizing committee had presented 14 recommendations, some of which were fairly concrete (revising ISSN), some concrete but not easy, others amorphous; CONSER may move to facilitation role rather than just database building; need an advisory board with members representing data creators, vendors, ONIX, CrossRef, etc.; identification still important even if access is article-centric; emphasis on access as description becomes less stable; what is shareable about access via opac, ERM, registry, journal lists (URLS and servers are often institution-specific) * January 2005 topic: institution and vendor reports on what’s being done, perhaps Tim Jewell, DLF data elements set and its relation to CONSER record; staff may be realigned as serials checkin changes (NISO and Editeur are working on serial release notification, with Linda Miller of LC on the working group)

Set records: Joan Swanekamp reported on Yale’s experience with creating records for Eighteenth Century Collection Online; not multiple version records; found microform notes still on records, needed 007 and 006 coding * PCC Automation Standing Committee produced standards for set records and they need to be checked; Yale did a joint version record for Evans set; most ECCO records didn’t have subject headings and Yale is looking at taking the subject headings from original print or reprint records * OhioLink is looking at standards for set records; Cornell and UMich have records for Making of America and other institution resources (but don’t promise anything about the records); Indiana is working with Alexander Street Press * Columbia has a partial FTE working on analytics for sets without records * Yale is trying to convince vendors that exporting records to the utilities is to their benefit (sell more of the product) rather than inhibiting sales of the records

PCC BIBCO records and copy cataloging: Wisconsin did a study in April and found that 1.8% of BIBCO records, 1.3% of LC records, and 3.3% of non-PCC copy needed editing * most institutions process PCC copy the same as LC, none admitted to upgrading non-PCC to PCC copy * timeliness is part of quality; try to use level of staff that is likely to apply desired level of scrutiny

Searching across digital collections: mostly project approach so far, needs to become program; issues: number of delivery platforms (e.g. ContentDM, Luna Insight); best practices for collection building and standards; model for unified discovery of collections and items (OAI repository and harvesting, federated search) * need to streamline access so delivery platforms are not impediment, Fedora digital objects know what sort of object they are; Yale developing rescue repository but not insitutional repository

RLG Strategy Focus Group (Friday p.m.) * record loading: Keio has signed off on load; Harvard not loaded for 2 years; UMn backup load being completed, U Cambridge has started process for load; UToronto has multi-branch holdings issues; NYPL does WLN authority processing every night and then sends records to OCLC; LC records loaded daily * RLIN21 deadlines have mostly been met, need to request about as many user IDs as had active accounts * SCIPIO will probably be migrated first * dataloads should not have much of a hiatus as RLIN migrates, though snapshots will be unavailable for a while * ERMs are mostly homegrown and take a good deal of effort * Yale doesn’t worry about URLs on bib records for documents; Columbia does link checking in link resolver, not in opac * Berkeley public services staff cannot agree on links to book jackets

Machine-Readable Bibliographic Information Committee (MARBI) (Saturday a.m.; Sunday p.m.) * papers linked to agenda at

Proposal 2004-05 (Changes needed to accommodate RISM data -- Music incipits): this started out as a discussion paper at Midwinter and was now accepted with addition of $y (alternative text for URI) and $z (public note about URI), and some editorial changes

Proposal 2004-06 (Defining the first indicator and new subfields in Field 017 to suppress display labels in the MARC 21 bibliographic format): indicator and subfields defined to allow for explicit captions for type of copyright registration or deposit number - approved, with editorial changes

Proposal 2004-07 (Applying Field 752 (Added entry - Hierarchical place name) for different purposes in the MARC 21 bibliographic format): 752 has been used for production data and subject coverage is desired; paper had options to expand purpose in 752 via indicators or to re-establish a 6XX (652 or 662) field for subject usage; usage for production or subject is not always distinct; should data be repeated when both uses are desired?; motion of support for 6XX AND 7XX fields for hierarchical place names was supported and proposal will come back (LC/NLC/BL will analyze and make decision on 652 vs 662); parallel subfielding would be desirable in 752 and 652/662 (options: $a as continent or country; don’t want to mix repeatability for hierarchies or for multiples (latter should probably result in repeated fields) * LC Geography & Map Division uses NAF heading as basis for determining levels in 752 though country is always coded (e.g. “Paris (France)” gets $a France $d Paris; “New York (N.Y.)” gets $a United States $b New York (State) $d New York)

Proposal 2004-08 (Changing the MARC-8 to UCS mapping for the halves of doublewide diacritics from the Unicode/UCS half diacritic characters to the Unicode/UCS doublewide diacritic characters): proposal to use one doublewide diacritic for ligature in Russian and other cyrillic languages (and Tagalog doublewide tilde not used since 1898), accepted with editorial changes - OCLC did not support since they felt there are implementation issues; complete Unicode implementation will make the issue less problematic

Discussion paper 2004-DP04 (Use of ISBNs and LCCNs in MARC 21 bibliographic records): ISBNs are used for machine deduping, overlaying, searching and unique values would be helpful; two types of invalidity (does not machine-check, does not belong to item in record though it appears in item); parallel subfielding in control-number fields (010, 016, 020, 022, etc.) would be helpful; straw poll supported adding $y for numbers for other manifestations that appear in an item (paperback, hardback; collection, analytic) * to come back as proposal with $y defined for other manifestations, $z only for machine-invalidity

“Assessment of options for handling full Unicode character encodings in MARC 21” (paper by Jack Cain): something will be lost in moving Unicode encoding back to MARC-8 but this is interim situation; some systems may be able to index but not display characters * system implementations: VTLS Arabic libraries are inputting vernacular in 245 and Yeshiva is doing parallel; III and Aleph import in UTF-8 too

Business meeting and LC report: Understanding MARC authorities now out in revised form, will be webbed later this year * MADS is an authority complement to MODS * ISBN13s will start appearing in 2005 (on U.S. publications, already showing up on some foreign publications); LC will put both ISBN and ISBN13 on records starting in October (either repeated $a or repeated 020s); ISBN13 will be only scheme in 2007; OCLC will put ISBN13s in 024 (as EAN) until database migration * Adam Schiff will be new MARBI chair

... go to more ALA reports ...