ALA Midwinter, Philadelphia, 2003

FRIDAY--Jan. 24

7:30AM -7:00 Registration - Convention Center - opens at 8A go early - run to Wyndham

8:00AM -4:30 Wyndham Franklin Plaza -- Conference Center BR

Managing Electronic Resources – Training (7:00AM-6:00PM is slot)

Pick up packet and register - Run to Courtyard by Marriott for LC Action

Return until 3:45PM - run to Hilton for SACSEM

Missed a lot of this due to conflicts. I picked up all the handouts and can check URL's.

Davis. Four challenges. 1) Linking diverse content (catalogs, portals, deep web crawlers, google-type searches). OhioLINK is trying to improve and evolve electronic delivery systems including video clips, art images, theses, Sanborn fire maps, OPAC, online journals, etc. Check out: http://www.ohiolink.edu or http://dmc.ohiolink.edu 2) Course management systems and TEACH Act. Check out: http://www.utsystem.edu/ogc/intellectualproperty/teachact.htm or http://ww.unc.edu/~unclng/TEACH.htm 3) archiving digital materials. A partial solution is MIT's Dspace model and OSU's Knowledge Bank. Check out: http://www.lib.ohio-state.edu/KBinfo http://library.caltech.edu/digital http://www.dspace.org/live and http://escholarship.cdlib.org 4) Authentication. A potential solution is SHIBBOLETH with the goal: to develop and architecture and policy framework supporting the sharing of secured web resources and services. It has federated administration, access control based on attributes, active management of privacy and is standards-based. Check out: http://shibboleth.internet2.edu/index.html http://middleware.internet2.edu/MACE/ Another potential solution is OCLC Cooperative Rights and Resolutions Service which would have a central repository for rights information and is intended to be developed to support libraries control their database and journal access; it would be a cooperatively developed database, continually updated, provide management statistics and have rights metadata security. Check out http://www.cofc.edu/cdconference/Charleston_2002_pre-conference.ppt [and OCLC]

Irvins. Stand-alone e-journal management services. Discusses the common problems with a focus on collection development. Some issues: 1) publisher trends are to remove themselves from aggregators, 2) publishers change e-journal host services/aggregator arrangements (e.g. Tay and Francis leaves Catchword/Ingenta), 3) increased use of embargos, 4) retrospective ejournal conversion by publishers and/or aggregators, 5) A&I services providing reference linking (H.W. Wilson and SFX), 6) expect new services from all players, including ILMS, subscription agents, 7) expect vendor consolidation: partnerships ( SerialsSolutions and SFX; Openly Informatics and ISI, etc.)

Tonkery: Three A's - aggravation, agitation, aggregation.

A subscription agent provides "opportunities" you might not have otherwise. We are choking to death and out of control with online resources. To solve it, we need a shared solution with libraries, publishers, vendors, aggregators, but we're not there yet; we are still aggravating over licenses, etc. With print, we had a very simple workflow - a "free" online is not simple and not free. End users want journals on their desktop - free. Everyone thinks it's free - we have obviously failed to get the message across. Technical Services departments in libraries have already dropped staff to save money, and now we need lots of staff for complex and costly management of electronic resources. We are still in an experimental stage. Setup with renewal and maintenance is difficult with many more scenarios, deals, etc.

A subscription agent is someone who will make it easier for the library at a reasonable price. This is possible because of what agents are good at: records management, publisher relations management, data collection, tracking events (e.g. claims, dates for activation, renewal, etc.), knowing if web site is operational, having a backdoor into a web journal so they can turn it on for the client, providing usage statistics, helping build a knowledge database (similar to SFX building and collecting) and helping with workflow.

35% of EBSCO's orders involve electronic resources.

This work required a higher level of competency for library staff to manage the electronic subscriptions.

Mouw (see handout)

http://home.uchicago.edu/~mouw/

Despite the increased cost of staff salary on electronic resources, the resources themselves are very expensive and are supposedly more accessible to users. The presentation focused on OpenURL which embeds metadata in a standard way to send to a server, and then identify and verify the request, and finally send the request (article) back to the user. SFX is an example of locally controlled linking; we want to send users to things that won't bounce, that we support, etc.

Producers who say they support OpenURL seem to forget about it when they change URLs for their journals. ISSN's cause problems with OpenURL linking for older titles, titles with multiple ISSN's, etc. The source side (indexers) understand OpenURLs but lots of target providers don't get it - e.g. Cambridge. Users in his library are clearly using SFX successfully. They catalog titles in the OPAC and also keep an A-Z list. Users predominantly start with indexing databases, then pick the OPAC, and lastly the A-Z list.

8:30-12:30 Courtyard by Marriott -- Meeting Room 104

LC Action discussion - actually starting at 9AM - focus on developing a set of proposed activities intended to encourage the development of one or more the proposals set forth by Marcia in her report (Judith R. Ahronheim jaheim@umich.edu )

It amounts to creating a new "thesaurus" on the web to search catalogs and databases that searches metadata. Perhaps we need to develop a single vocabulary that can be used with a variety of tools: OPAC, web, or databases or as entry vocabulary to do individual or cross-database (thesauri) searching. Design is critical. If you are going to develop a vocabulary to use, it needs to be carefully planned out to allow different applications depending on what is wanted. If we can build it so that at the deepest level we explain the source of vocabulary and relations between vocabularies, but on the top, superficial level, it is a fairly simple list of terms from which users could select. Is there an automated (i.e. a software package) that could begin to create a list of vocabulary words followed by human evaluation and editing. We need to work on identifying key issues for a) display features, user interface; b) for use by those creating the structure. So we are trying to achieve staged access with a vocabulary using clustering - we need to figure out what the structure IS. Bibliographic families is a CC:DA/FRBR thing within the bibliographic environment. User vocabulary is a SAC thing, external to the bibliographic environment.

So we are looking at: 1) user access vocabulary with clustering, 2) hierarchical relationships among bibliographic entities, 3) staged displays. The system design has to be set up FOR the vocabulary, not a separate thing. We need to investigate clustering as a method for the subject vocabulary and for bibliographic families. We need to start educating people and se if we can find out what else has been done. We can try being incorporated in other meetings or programs at ALA and try for a program summer 2004.

The clustered vocabulary needs to start where they are and be led by the cluster to where they need to be to find all that wonderful cataloging or web database metadata. Clustering as a method of organizing subjects and bibliographic families for user searching with staged access as a principle has to be incorporated into the way the clustering is done. Sara Knapp's list is horizontal/linear whereas ATT and some LCSH is hierarchical - we need both styles because different users respond in different ways. Knapp's list is limited to social sciences. Perhaps we should start with FAST, then add hierarchical - Marcia disagreed; one committee member thought it a good idea. Somehow you have to get natural language into the cluster. So then do you start with Knapp's list and then add FAST and include word or phrase from the search string?

There seems to be a lot of general research of human psychology stuff, but specifically on information seeking - not so much - nor did anyone seem to know of an ALA activity going on; yet that piece is probably not our committee's responsibility. So we need to start with an iterative design that can be built on, that we could create a prototype from what we know - we need to do something - a mock-up, something people can test, do more mock-up, test again, etc.

So, we need to develop our functional goals and who we think ought to get the money to actually do the developing; vendors won't do anything until the library community wants it. J. Byrum said there is an AVIAC meeting where he could try to get some interest. We also need to plan a program; a list of possible speakers was developed.

Plan: 1) education; 2) functional list; 3) grant proposal to fund a prototype - need to identify testers and a possible granting source - Gates Foundation, Mellon, NSF, IMLS, ERIC, LSTA; 4) report on prototype and testing; 5) RFP to get a vendor to build, hence we need specifications - e.g. what mechanism should be used - metadata enrichment, OAI, etc. to access disparate collections. What is the cost of the process - what aspects will be computerized and what aspects will require human intervention? What form of delivery (i.e. worker entry and user searching/display)

9:30-12:30 Wyndham Franklin -- Philadelphia Ballroom

ALCTS Technical Services Directors of Large Research Libraries DG (Big Heads)

Minutes: http://www.acsu.buffalo.edu/~ulcjh/bh12003min.html

Round Robin: http://www.acsu.buffalo.edu/~ulcjh/bh12003rr.html

4:30-5:30 PM Hilton Garden Inn -- Salon A/B

SAC Subcommittee on Semantic Interoperability – meeting

Report. (See also minutes)

Subject Semantic Interoperability

The ability of two or more systems or components to exchange or harmonize cognate subject vocabularies and/or knowledge organization schemes to be used for the purposes of effective and efficient resource discovery without significant loss of lexical or connotative meaning and without special effort by the user.

1. We decided we need a glossary to go with our report so we will be working on definitions for the glossary.

2. One item in the charge is to identify projects and approaches/methodologies to semantic interoperability. We need to do more research in order to identify as many as possible before we can evaluate their effectiveness. We will try putting out a request for info on Autocat or similar lists. We might try a news article.

3. We also decided to develop a bibliography. This has been started, but more input would be helpful.

4. We need to start developing criteria to measure projects. This can be done by trying to extrapolate descriptions of existing projects or reviews of projects. It would also be helpful if we could identify "experts".

5. There are really three aspects and we may or may not be able to address them all: 1) vocabulary, 2) user access, 3) system design

6. Program ideas will need to be developed and ready for presentation to SAC by ALA Annual.

Since SAC decided to try a program on user behavior, that "psychological" piece probably doesn't need to be addressed by SACSEM; we can probably focus more on the vocabulary and structuring a system

see: http://www.und.nodak.edu/dept/library/Departments/abc/subjhead.htm#SACSEM

7:30-9:30PM Loews -- Z - Washington A/B

OLAC Cataloging Policy Committee

MARBI report / John Attig. Proposal 2003-02: Definintion of subfield $u (URI) in Field 538 (Systems Details Note) in the MARC21 Bibliographic Format requests addition of $u in 538. This was requested by a group working on digital masters. One wouldn't have to include the text of lengthy technical details in the bib record. DP 2003-02: Coding graphic images in leader/06 in the MARC21 Bibliographic Format. At issue is moving vs. still and projected or not and how to code them. In the digital environment they can't be projected so they thought still or moving would work. $g moving $k still. Filmstrips, slides, transparencies would go from $g to $k. The utilities objected because old records for those materials would be wrong. DP2003-03: Adding field 024 (Other standard identifier) into the MARC21 Authority Format. The role of headings in authority records will become more important with FRBR. 024 could be used to help link manifestations with headings. It is likely that more name/title authority records will be needed.

CC:DA report. JSC's working group is looking at expression-level identifiers that allow you to identify expressions of the same work. In Feb. their report will address ch. 1 and 26 to define works and which part of the heading refers to an expression and which part refers to manifestations. Modeling of uniform titles (author/title) looks pretty ugly right now and the only way it will "look better" i.e. work well for users, is if we create many more authority records.

AACR3 may be coming. Ch. 21 will talk about creating relationships, not headings. Ch. 23-25 in new Part 3, will be about authority records so form of entry will be essentially switched to authority control. Ch. 1 will be huge, and the rest small; it may also get to ch. 1 being organized like ISBD. There will be little in the 2003 revision - probably the biggest thing will be a better index. Time and effort right now is being spent on the whole re-write.

Conventional terminology. In particular for rule 7.5, a CAPC committee is working on a list for CAPC review, then CC:DA. They are considering functional terms at the expression level [gmd] and functional terms at the manifestation level [smd]. The committee is still considering a standard list, but that it be developed from conventional terms. The idea is to revise the list in the rules rather than pick anything you want. Currently, conventional terminology in ch. 9 is optional in AACR2, but LC has a list it prefers in the LCRI.

See the CAPC page for more documents: http://ublib.buffalo.edu/libraries/units/cts/olac/capc

SATURDAY--Jan 25

9:00-11:00 Convention Center -- Room 108A

OCLC -- Get Connected with Connexion

Several products were described: PromptCat and Bib Notification might work better with Ex Libris. It is 11 cents/record for TOC and 856. Language sets: can purchase materials, cataloging, processing, plus even posters to advertise you materias; can get bok, av in Spanish, etc. at an online store on the web. http://www.oclc.org/languagesets OCLC Automated Collection Analysis allows you to analyze your collection. You get reports showing strengths and weaknesses. You get a comma delimited file. Check their demo site: http://acas.oclc.org Retrospective conversion comes with free authority control processing. MARS Automated Authority Contorl includes an authority record notification service - i.e. you get new or updated authority records. Dewey 22 is coming in July.

CONNEXION

CONNEXION will replace all cataloging and will be updated quarterly. Some features: quick tips, easy way to find out what library goes with a three- or five-letter code, added indices for the save file, dynamically linked headings - easy to verify and when linked are automatically changed in all master records when the authority record changes. In February, diacritics will be placed after the letter per UNICODE and will have a character pad similar to Passport to supply characters. In May, they plan to improve searching; authorities will have derived and command line searching. In list results, the number rather than the title is hot-linked; they broke out display options and listed them in alphabetical order. In editing, if the headings have already been controlled you will see them displayed in blue - you don't have to check them. When you click check holdings, it brings up a hyperlinked list of codes that display the full names.

CatExpress is built for fast copy cataloging - again, may make some sense with Ex Libris. Could export from CatExpress, the validate in ODIN.

CONNEXION 2003: OCLC is adding a title bar to allow quick verification of the title searched. We will be able to shift the Fixed Field to either the top or bottom. Action status ill be more like CatME ina new bar so you can tell if you already exported the records. New diacritics features.

CONNEXION CLIENT 2003: This will be an optional feature adding additional cataloging power. It will allow macros, labels, and interactive online functionality. It will be Windows-based. In third quarter, NACO functionality and electronic resources cataloging will be added. In fourth quarter, it will allow off-line cataloging, local file, batch processing, i.e. the rest of the CatME stuff. Keys will be programmable and one can use mouse or keyboard - your choice. The off-line editor looks like CatMe and can change colors, fonts, etc. The toolbar will be customizable. One can add a line at the end of the record, and reformat to reposition - it will be numerical within tag sets. Alt arrow up or down will allow you to shift within a tag range.

Shift F1 will be help for MARC21. Will be able to cut-n-paste the OCLC number. The client will allow you set display with or without $a. Menus will be similar to a browser. When all the CatME features are transferred to the client or CONNEXION, CatME goes away.

Windows 98 and NT4 are no longer supported. If it works in your library, fine but OCLC won't help if you have problems. OCLC supports only above Netscape 6.1 and Internet Explorer 6. Fees continue to be structured as before. The client is free.

Future: OCLC is trying to leverage WorldCat. They are looking at a cooperative rights database initiative. They are trying to reduce the burden of management rights metadata for electronic and print-based content. Intend to answer: do the rights exist for the content for the library's users? Where is that content? What is the best way to access the content? Another initiative: Cataloging Made Easy - new projects for new services - maybe a copy cataloging wizard. They are thinking about the OCLC knowledge community - something like QuestionPoint but for cataloging.

Questions: 1) what is the keyword search cost in CatExpress - it is not a sca ti, it is really a fin ti but also applies some adjacency. CatExpress alone is a subscription but full authorization use of CatExpress is billed like full authorization. 2) Shift in diacritics - how does it affect export and local systems? It is really a display issue, the content of the record will be as it was before. 3) What format will the rights management be in? It will probably use ONIX for Serials Management, or ORDL. It is just rights management, not a bib record. 4) How to use bibliographic notification? It will go to an online safe file or local save file; it will be easy to find in the product list. You download the file and import it, edit records, then export to local system. 5) What about CJK and Arabic - might be in the future - not soon. 6) Export overlay problems - how do we keep from overwriting records already downloaded? Best option is to save everything to a save file, then at the end of the day, "tag all", then "export tagged records" - i.e. only one file can be handled by one person. The client to append to the file so lots of people can export in a day and not overwrite.

9:30-12:30 Loews - Washington A/B

MARBI - http://www.loc.gov/marc/marbi/mw2003_age.html 

P2003-01: Defining subfield $2 in Field 022 for ISSN Center code {approved}; P2003-02: Definintion of subfield $u (URI) in Field 538 (Systems Details Note) in the MARC21 Bibliographic Format {approved with amendment - subfield I and 3 include; redefine definition and scope; define in holdings format}; DP2003-01: Data elements for article level description {will pursue a proposal looking at 4.3 and 4.6 sections in the paper as possible solutions}; DP 2003-02: Coding graphic images in leader/06 in the MARC21 Bibliographic Format {will be forwarded to communities for review. Reason: first lets decide if projected/non-projected is the right place to start or is it still/moving it ?}; DP2003-03: Adding field 024 (Other standard identifier) into the MARC21 Authority Format {International Standard Text Code (ISTC) - will come back as proposal - relation of authority and bib formats addressing issue of transcription of number from (or not from) the item itself}

11:30-1:00 Convention Center – Room 108

ALCTS Electronic Resources DG -- Electronic theses and dissertations: Brian Surratt (Texas A&M) uses a Perl script to generate MARC records for an electronic database; Gail MacMillan from Virginia Tech talk about archiving preservation of electronic dissertations

MacMillan

Gail said annually 400,000 masters and doctoral theses are being produced. Often some show up in journals, but no in complete form. They don't circulate much with the library or library to library. Virginia Tech wanted to help students learn how to publish electronically. They got locally created journal authors to send a copy to the library online. For theses, they are required to submit them electronically; most use PDF. They are no longer quired to bind a copy for the library. When the library catalogs the thesis, they include the abstract from the thesis to enhance searching. Virginia Tech takes the document and creates a documents with ETD markup to allow searching within the online database. Now that they are available online, they are reacing people with all kinds of email extensions and lots of different countries.

Virgninia Tech process: The grad student creates the thesis electronically (e.g. Word document). The student committee approves. Student submits online with a completed web form (they can choose type of access - open or withheld for a period of time). The grad school approves - email notification is sent to the author, advisor, and UMI. The library stores and provides access on services that is backed up. It is the library's server and search engine software. The library catalogs for OPAC and OCLC. Virgninia Tech has free software we can use or adapt for the process. It includes submission and management scripts, log files, surveys online data gathered from the author. They started with freeWAIS, but no uw InfoSeek software - UltraSeek. See http://scholar.lib.vt.edu/theses So far, publishers don't seem to be afraid to publish from the ETD's or portions of them. Elsevier, says they want to publish first, then open access is okay. What about format migration? - yes it is an issue but the software isn't changing as fast as it used to. Adobe is open source. OCLC is looking at being a digital archive for theses.

Surratt

Brian has developed a perl script to create MARC records. It mines the ETD system software from Virginia Tech for text to go to the MARC record. It doesn't create a call number or subject headings. When the "switch" is turned on to make it public, the draft MARC record is created for review by the cataloger who has to assign call number, other notes, subject headings, do NACO authority work, etc. They then have to upload to OCLC. The ETD.pl (public) activates generation of the MARC record and ftp's it to staff server in the library. The thesis office (in grad school?) has to have the perl script on their server too. The URL takes the user to the server where a description screen pops up and the pdf file is there to click. Handout has the script any programmer should be able to work with. It could also be used with other projects. Students still fill out a UMI form to send along with auto notification to UMI of the URL - UMI then can get the file and do their "thing". Texas A&M is going to try it with a digitation project. http://lib-testing.tamu.edu/users/bsurratt

Networked Digital Library of Theses and Dissertation (NDLTD). You can join for free. There is a listserv. http://www.ndltd.org

2:00-5:30 Wyndham Franklin Plaza – Philadelphia Ballroom

CC:DA – Liaison

http://www.libraries.psu.edu/iasweb/personal/jca/ccda/301-agen.html

Announcement:

The agenda contains links to CC:DA documents that will be discussed at the meetings (with one exception, which is available only to CC:DA members and representatives).

The Saturday afternoon meeting will feature reports from the Library of Congress Representative, Barbara Tillett, from the IFLA Cataloguing Section, by Glenn Patton, and from the National Information Standards Organization (NISO), by Paul Weiss. There will be a report from the ALA Representative to the Joint Steering Committee, Matthew Beacom, who will report on the September 2002 meeting of JSC in York, England. Finally, the Task Force on Major and Minor Changes will present its draft publication "Differences Between, Changes Within: Guidelines on When to Create a New Record."

The Monday morning CC:DA meeting will include reports on a program and a preconference at the 2003 Annual Conference in Toronto, from ALA Publishing Services by Donald Chatham, from the CC:DA Webmaster, and from the MARBI representative. The meeting will also include discussion of several new rule revision proposals:

A proposal from the Map and Geography Round Table to correct an example in rule 3.1F1

A proposal for an additional example in rule 3.5B

A proposal to simplify rule 12.1E1
The Monday meeting will also include reports from two Task Forces:

The newly-formed Task Force on Reconceptualization of Chapter 9, chaired by Michael Chopey, will present an interim report; note that the Task Force is meeting on Saturday morning from 10:00-12:00;

The Task Force on Consistency Across Part I of AACR2, chaired by John Attig, will present rule revision proposals covering Areas 2 (Edition) and 3 (Material-Specific Details); the Task Force will be meeting Friday afternoon from 3:30-5:30 PM.

Notes:

The task force on the Rule of Three is basically finished and CC:DA sent in its comments. We will start a new task force if work in Ch. 21 required work. Officially moved to have a task form on FRBR terminology. The task for on special characteristics is finished. Kristin said all liaisons are welcome to comment on the discussion list either as liaison or as their own knowledge can contribute to the discussion.

Agenda item 5. See for chair's report on CC:DA motions: http://www.libraries.psu.edu/iasweb/personal/jca/ccda/chair4.htm

Agenda item 6. LC report. (See handout) Z39.50 access to authorities is still in the works. LC currently has no budget and the proposal is to cut 15%. For cataloging policy on remote electronic resources and loose-leafs see http://lcweb.loc.gov/catdir/cpso/train.html Authority records including 3,250 free-floating subdivisions have been distributed to subscribers. Subdivision authority records may be access in ClassificationWeb. American Memory is growing and access is free; good to add to your OPAC. LC is exploring including name authorities in ClassificationWeb. Planning has begun for the implementation of Unicode standard for LC's MARC21 bibliographic, holdings, and authority records - it affects every record they have. LC has organized an initiative, ZING to evolve Z39.50 to a web platform protocol. LC's Z39.50 gateway now contains more than 500 databases on 400 servers, 145 are non-US. Metadata Object Description Schema (MODS) is now out for review - it is a lightweight version of MARC using language-based tags rather than numeric ones. Harmonization of MARC21 with UKMARC was completed. LC will not be implementing the new Leader/05 Status value of "obsolete" that was added for FAST. They will also be deferring 148, 448, and 548 until they can be indexed by Voyager. Use of $u in 670 is also delayed until 2004. A file of 40,000 item-level still picture records were created in the Prints & Photographs division. These could be added to OPACs because there is an 856 to the image, free access. LC is storing 1.2 million books and bound journals sorted by size. The MINERVA Web Preservation Project was establish to collect and preserve primary source materials. One can now search Bibliotheque nationale de France authority records and RAMEAU subject headings http://www.bnf.fr see also http://www.bnf.fr/pages/zNavigat/frame/infopro.htm

Agenda item 7. IFLA. The CC:DA page will have a link to IFLA's new webpage on FRBR when it comes up. A number of countries are rewriting rules to incorporate FRBR. There is a Working Group on the Use of Metadata Schemes working on guidelines for essential elements to include in any metadata scheme for descriptive metadata. The Working Group on a Multilingual Dictionary for Cataloging Terms and Concepts continues to work on setting up the English vocabulary that will form the basis for the other languages. The Working Group on Functional Requirements of Authority Numbering and Records (FRANAR) is working on the extension of the FRBR conceptual model to authority information. Paris Principles are being reviewed worldwide.

Agenda item 9. ALA has a representative to NISO. ALA decided ALCTS should pick a person to be the rep. Paul J. Weiss reported on recent activities of NISO.

Agenda item 10-1. JSC report. Three parts: 1) what's been done since the last CC:DA, 2) a few things needing CC:DA action, 3) future for Apr. JSC meeting. Chair's motions 3,9,20 - FRBR terminology "manifestation" went in during Sept. and out during Dec. Just changing words doesn't work because of the conceptualizations that go with the words. Chair's motion 9 - Ch. 9 will take a lot of work to evolve to a new set of rules for catalogers. 21.30J and 24.20C1 were approved with slight revisions and will be in the 2003 AACR revisions.

Agenda item 10-2. FRBR terminology/3 - not much point to more discussion until Pat Riva finishes her work. It was proposed that a consultant be hired to write a draft to comment on because the task is huge. Comments: a) it is critical to use FRBR words in ways that was FRBR's intention in the rules, and not just sit there and try to do a substitution and possibly lose the conceptualization or relation to other chapters; b) are we proposing scrapping AACR and starting all over? Well, not totally, we should include/incorporate recent changes. If we're going to deconstruct the code just to fit FRBR in, we'd better know where we're going and that the library community wants it or thinks it will be better; c) JSC needs to delineate what CC:DA should be doing.

Agenda item 11. a) the document on when to create a new record heavily used FRBR terminology and that may now need to be retracted. Points: 1) it's already convoluted enough, 2) refer to rules - don't reproduce them in this document, 3) should they mention URI's? They are in LCRI's but not AACR…? Yes, include anything that defines a major attribute whether or not in AACR2. Yes, go beyond the rules to help people to do a good job of cataloging in conjunction with the rules. The checklist sort of gives you a sense you can march down the list without thinking about the principles or content. Actually the use of "difference" and "change" isn't always clear even though they are different.

SUNDAY--Jan. 26

7:00-7:45 Marriott -- Salon E (7:00-9:00)

OCLC breakfast - rsvp'd - leave early

Only heard the very beginning overview. See handout.

8:00-9:30 Loews -- Regency Ballroom C

CONSER-at-Large -- discussion of one record for online journals

The new CONSER Cataloging Manual is out - probably not being shipped to despostories so I will need to order. It will be in Catalogers Desktop (which I can't access or print from even though we pay for it). Regarding publication patterns, they are looking at two notification initiatives a) with OCLC some sort of bib-note, b) subscription agents. A committee is looking at electronic journal holdings format - how they should look and how they would be used. Diane Hillman is working on Universal Holdings for long term storage and technical issues.

FRBR - Everrett Allgood. They are looking at citations for work and citations for expressions in authority records and FRANAR's work on numbering in authority records - will there be able to be a connection? They also need to look at the attributes of entities in FRBR and how they apply to serials.

510 - what should be done with the? A survey was 50/50. They thought about stripping them out or storing them at OCLC. NLM and Chem Abstracts are updating their records still.

Electronic journals in aggregators. From option B in the survey, the committee including Jean Hirons, came up with Option B+. The idea is one generic e-journal record as it appears at the publishers site, but somehow use for inclusion in aggregations by multiple vendors. The problem of multiplicty of records was identified by Big Heads. The idea is that the local library would add what's appropriate for them but OCLC would be a generic record. Background: time to stop making separate records for same title from multiple aggregators. B+ is a record described from the publisher with ULR(s) from aggregators. How do you then handle ISSN's? How do aggregating companies identify records? What will the SerialsSolutions-like companies do? The agenda lists some assumptions to add to B+

Discussion points at the meeting: a) libraries want to provide access to serials in aggregations via the OPAC - Big Heads agreed there should be a bib record, but just one for the electronic journal (as long as it is the same thing - not reshuffled articles, etc.) b) libraries need and will contiinue to need record sets, c) creators of record sets need base records that can be customized, d) the base records should be CONSER records, e) the base records should be separate (from and additional to print) records that reflect the online version, f) separate records may be either created by catalogers using Option B+ or machine-derived from existing records, g) the Task Group on Journal in Aggregator Databases should define the fields for a machine-derived separate record, h) for titles for which there is no record, records should be created by CONSER, I) for changes involving the aggregation themselves, i.e. titles added/deleted, maintenance will be handled by serial management companeis (SMC) and OCLC, where possible (i.e. SMC notifiy OCLC) j) for bibliographic changes to the serial, e.g. title changes, changes will be handled by CONSER

Hope to develop guidelines for discussion and approval at the CONSER meeting in May, with work starting after that. What would catalogers take out to make the records "neutral"? What would you need to retain? What happends to links in print records - 776, 856, etc. You'd no longer use double 130 qualifiers. 246's might be used if various titles (in title changes) are used by aggregators (put together differently on aggregator pages from the publisher) 506 might be more generic or non-existant. 538 should be generic. 710 and 730 for specific aggregator would be dropped. 500 description based on - would need to be specific. So how do you gather aggregated titles - use URL's - one line to remove would be easy. Start dates, what ought that be? 25,000-30,000 records are currently in OCLC; 5000 appeared to be dups with 2/3 being CONSER records. It could take 1-2 years to merge, but there are possiblities of maaching manipulation of cleaning up 130, 710, etc.

An interim state is needed so we don't keep making multiple records. In OCLC, only do URL? Aslo need guideline on $z and $3 - probably not but need to determine if $u will do adequately. Wondering if a decision could be made on what form of aggregator name is in $2. CONSER's policy has always been to create separate records. Some libraries will continue to use option for single record. At some time, need to determine what you can put on print in the CONSER database so as not to confuse. As more titles drop print and if a title changes there will not be a print record to add a URL to. Authority records, etc - we need to collocate and differentiate at the same time. Do we need to decide on what ought to be qualifiers, e.g. always add (Online) ? ISSN polic is one (1) ISSN per record, i.e. not two 022's. 022 is designed to use $x of 776 but vendors don't seem able to use it to find the print. MARC21 will allow two 022's but the ISSN people were very against something like that (like 022 nnnn-nnnn (print) 022 nnnn-nnnn (Online). We need to educate vendors about suing the correct ISSN - best practices. Work level identifier may be developed - ISTC - to identify journal content no matter what the format is so there would be a number overall and separate numbers for each format and the ISSN agency could assign both at the same time.

Option B+ http://www.loc.gov/acq/conser/optionbplusdec2002.html

9:30-12:30 Marriott -- Salon K/L

SAC - General Meeting

OCLC FAST update http://wcp.oclc.org/fast/

See handouts.

Agenda item 5. MARBI decisions were re-reviewed to better meet FAST needs in Dec.

Agenda item 7. LC's report. [In the booth,I learned the LC Desktop on the web - beta testing may be available as early as March]. SACO web proposals will be given priority to encourage that means of submission. http://lcweb.loc.gov/catdir/pcc/saco.html PCC is considering folding SACO into PCC - perhaps more like NACO ??? Free-floating subdivisions will be in ClassificationWeb - but there will still be no automated means of knowing correct order of the string or whether it should be $x or $v when both are possible. When you look at SACO records on the web, the not done ones will say (Proposed) or changes will say (Proposed change). When you search the web lists of subject headings and nothing is there, it could be 019 SP - being sent to editorial board, on its way to tentative list, getting reviewed, etc. Plan to start working on genre headings - 155's - in small chunks. Just cuz LC is not doing 148,448, 548 doesn't mean the rest of us can't. Since 1999, information on the form of geographic subdivision has been included in records for all newly established or revised geographic subject headings that are also authorized for use as subdivisions. In MARC21, this data is carried in the geographic subdivision linking field 781. If a geographic headings is not authorized for use as a subdivision, a statement to that effect is added in at 667 field. Use of "Great Britain" as a geographic subdivision will be changed so that is it used ONLY when the topic applies to all countries. Otherwise, the individual countries should be use, England, Ireland, Northern Ireland, Wales.

Agenda item 8. LC is accepting member copy for Dewey call numbers for 082 ^4. WebDewey subscribers automatically get quarterly updates to class numbers. Semantic interoperability of English Dewey and German Dewey. "Forest Press" is going away - it will just be OCLC.

FAST - see handout. In phase I there are 6 facets: topical, geographic, period (chronological), form (genre), personal names, corporate names. In phase II, authority records will be created for uniform titles and conference and meeting facets; the geographics facet will be enhanced. 900,000 authority records have been created so far. OCLC will use the obsolete code approved by MARBI. The geographic area code has been added to geographics and hope to at latitude and longitute to headings; they thing 90% are possible. Alpha test so far contains errors, cross-references are not all there. They are open to comment. There doesn't appear to be a way to put FAST back together. However, NLM has their records stored in separate facets and an algorithm puts them together for bib record distribution - maybe it works because their structure is simpler than LCSH. There might be some testing in the near future. FAST does not need to be limited to electronic resources. Research database is at: http://fast.oclc.org

Agenda item 13. Anthing to be posted on an ALCTS website has to be in .doc file.

Agenda item 18. SAC is author of GSAFD. They will be doing a final report by the next meeting.

Agenda item 14. Shelby reported on the task force meeting. The direction is primarily one of creating a system of clustering terms or bibliographic families for searching web, OPAC, or databases; doing some education on the issue; getting funding for a study grant; and enticing a vendor to take it on and develop it. SAC is interested in seeing Marcia Bates' report.

1:00-3:00 Convention Center - Room 103

OCLC's Rights and Resolution Initiative - manage rights to content

OCLC is trying to address the problem of rights management. The problem: content is available in many places, but which is appropriate (i.e. which one we licensed) to our user. Library access rights management is resource intensive - i.e. lots of staff time. What they are trying to design is a unique cooperatively developed central repository for rights information and holdings information for both print and electronic resources. It will be coupled with linking solutions to help libraries, e.g. SFX. It will attempt to answer the questions: a) do rights exist for this content for this library's users? B) where is the content?, c) how do we access it? This could work with FirstSearch, Proquest, etc. It creates an OpenURL - can use any OpenURL resolved (SFX) - go to the OCLC rights database - an XML packet goes back to the search service you started with. By doing it cooperatively with libraries and vendros and publishers, current data should get input automatically or manually.

Possible models. Between libraries, linking solutions and OCLC reights database. It could serve as a stand alone knowledge database that works with SFX. It could be integrated into the library's OPAC. It could be integrated into the library's discovery service of choice (OPAC, Gale, what?) Whatever the model, the library would be in control of access and authentication. They could work with the shibboleth model. OCLC is not proposing to keep our user ID's and passwords.

Librarian's interface

Home	Library des..	Vendor des..	Library profiles	Reports
	List of aggregator and what titles each has		Includes drop downs of aggregators info	Usage stats by aggregator, by title, not filled, etc.
	Include local electronic content - course packs		Member libraries could add aggregators and not listed	Can link out to journal
	Would include information from publisher, coverage dates, etc			Provider listing: Ingenta, ECO, Kluwer
	A small box for local information			Frequency
				Cost
				Coverage

Other things OCLC will look at: where to put: what about ILL rights, what about library specific information in a license, embargo information.

2:00-5:30 Loews – Washington A/B

MARBI -- XML and MARC

024-DP2003-03. The paper proposes the addition of Field 024 (Other Standard Identifier) to the MARC21 Authority Format. This would allow for the recording of standard identifiers relevant to what is identified in the 1XX field in the authority record. There are a number of efforts to develop stnadard identifiers for works as well as for people and organization, and asthey are implemented, there will be a need to record them in authority records. In addition, there are efforts to establish authority records for "trademarks". Particularly in Europe. Examples are the International Standard Text Code (ISTC), International Standard Audiovisual Number (ISAN), International Standard Musical Work Code (ISWC), International Standard Recording Code (ISRC), and the International Standard Authority Data Number (ISADN) Discussion -> expression level is intended. The number could be used in FRANAR. The idea is that these numbers could be stored in our records and ILS vendors could use them to create displays of related records. Do we always make an authority name/title for everything or only if multiples exist. What if I don't know about the other multiples? Where are we going to find a number while creating a NACO record?

XML (see handout). Organization codes only online. Codes for relators and 655, 656, 657 codes will be online only eventually. MARC 2709 with Unicode … ? will come out after MARCXML. Toronto program is same time as SAC - 8:30-12:00. Then in Orlando preconference, 1 ½ days with Thurs PM more general whole FRBR, Friday full day - demos. "Don’t be dysfunctional - put FRBR in your future.

4:30-6:00 Radison - Warwick - Crystal Ballroom

NISO Program -- ONIX: What's in it for Libraries?

Couldn't attend, but a friend picked up the handout.

4:30-5:30 Loews - Ballroom C

RUSA meeting <LC Action 2.3> -- Marcia Bates. Research design review : improving user access control to LC's catalog and portal information

Marcia spoke on her report commissioned by the LC Action 2.3 task force. What she tried to do was describe potential tools, communicate ideas to potential vendors, educate librarians why we need front-end mechanisms above the bibliographic record level. What she described is for use in search engines and portal mechanisms (how different is a portal from European subject gateways?). She is convinced there is a mismatch between catalogers [words] and uses [searching]. Users don't always know what they want or that there might be a special word to use. Once they think of something, then they can't think of anything else. Statistically, most popular terms are very widely varied and only 20-33% would be successful. Catalogers know what the book is about and what distinctions need to be made - it is a mixture of your cataloging background, knowledge of tools, and the actual item is in hand and knows what it is about.

How do we bridge that? The user should only need to hit the side of the barn, not the knothole. How do we lead them to the knot hole? - the word the cataloger used? A big bridge needs to made between the two. She is proposing a cluster vocabulary. It would include variations in spelling, misspellings, abbreviations. Entering any one of those would bring up the related words and can include words from LSCH or other controlled vocabularies. The searcher needs all the terms and variations brought together. So, she proposes starting with Knapp's list, then with good system design, bring up words to pick from that ARE actual subject headings from a thesaurus.

Her description of bibliographic families are FRBR-esque. For some works, lots of books, etc. are available surrounding an original work. She referred to Barbara Tillett's work. We need a system design that will show all the works with their variations and related works, adaptations, paper, web, AV - linking items and creating bibliographic record families. These probably follow the Bradford distribution, i.e. a few books have lots of related records and it progresses to only one book with nothing related. She suggests staging access in an online system in the 1:30 ratio - when you are trying to display records on the screen, 1 to 30 number of words, the journal article is 30 times as big as the title; i.e. users will look at 1/30th of something first so lets develop displays in those ratios.

Judy Aronheim said we need to get a developer to build a tool, we need to specify what we want. 1) who are we building this for? How it looks to tech services people to work on is very different from how it will look and work for a user, 2) something to work with whole web or just a library portal, 3) where will it live? - if only one, maybe LC; could people update if from anywhere? Could one use something like OCLC where it could be cooperatively built and more standardized? 4) how would we "cook it" - collecting additional subject terms; what mechanism could be used to add in established thesauri, 5) who would maintain it? Just to keep it working? How would additions to a particular thesaurus get migrated to the megathesaurus. 6) who manages it?it costs to keep it up; would there be assessments to contributors building it?

Mary Woodley's discussion. Assessment tools - are there any that could be used to test effectiveness? Who is the audience? Do they all test? What is an effective method to get vendors interested?

Questions:

Won't the clusters get too much information? Huge search response? Marcia still thinks this works because the cluster should include a few identifiers that would be the right one the user wanted or bring up a new cluster with better words.

Doesn't FRBR address the bibliographic family? Aren't there other approaches being worked on? Marcia said, by all meas use FRBR. Marcia is more focused at the system design level - pretty displays, graphic, bubbles, etc. We don't do a good job of displaying an array that can display links to related works rather than just lists.

How do you handle hierarchy? A reference librarian felt we should show that a narrower term would be better. Marcia avoided hierarchy on purpose because she wanted to include words and phrases. Marci says she is just trying to help the user fee, those other words they didn't think of that may get more specific without really knowing they did it.

Wouldn't they huge vocabulary be too big for a small library? Marcia thinks the single user could still think up nearly any word, but the library would have to mask out what they don't have. The purpose is to create something that is the user's vocabulary; not trying to help catalogers; once the user gets into the cluster, some of what they'd see is an LCSH heading and the do a better job of recognizing what a subject heading means thank thinking one up.

Would ISO standard for topic maps be worth considering? Marcia hadn't looked at them yet.

Marcia's paper will be on the LC Action 2.3 page.

6:00-8:00 Pennsylvania Academy of the Fine Arts -- 118 N Broad Street (only 1 ½ blocks from Convention Center)

Ex Libris reception - RSVP'd

MONDAY--Jan. 27

9:00-12:30 Wydham Franklin Plaza – Philadelphia Ballroom

CC:DA -- Liaison

Agenda item 14. CC:DA sponsored program, Knowledge without Boundaries, June 19-20th - 2 full days for $425. FRBR will be Suan AM June 22 8:30A. Pre-conference in 2004 Orland will be 1 ½ days.

Agenda item 15. ALA publishing report. 85% of the orders for the new AACR were for bindrs. The year of updates will be included in the footer. The entire index will be reviewed for the next update.

Agenda item 16. Correction to 3.1F1 example approved. The new guide Cartographic Materials is complete and will be out by Toronto or shortly after.

Agenda item 17. Adding examples how to handle various or unpaged atlases approved. Revision and simplification of 12.1E1 is tabled.

Agenda item 18. Receptualization of ch. 9. Will begin on scope paragraph of ch. 9. Then will look at other part 1 chapters. The are aiming to make ch. 9 less of a carrier chapter and more of a content chapter. They want language materials, cartographic, etc. to be discussed in their chapters. The might look at OCLC's list for "type" and LC's CPSO list for "e" or "m". ISBD(ER) has a longer list with more description or definitions that might be included. Lots of catalogers want to give physical description, yet we just took it out -- .e.g. pdf files. They also thought of "300 1 online resource" or 1 constant word or a field in fixed field or leader. John A. proposed that what ever is "additional" electronic characteristics be in a separate chapter or Everett wondered if it would be better to do more ISBD are organization to the rules. M Chopey said many want area 5 info for electronic stuff in each ch or at least in ch. 1.

Agenda item 20. Task Force on Consistency. Sections .0 and .7 are the areas with the most overlap cross-linking between chapters. Area 5 may need renumbering - otherwise the familiar numbering standards seem to be generally working. Area 2 and area 3: they looked across rules to see if things are consistently stated rule to rule and hence would make more sense to say it in ch. 1. They looked for consistency that could still abe applied when the rule necessarily varies in each chapter. Where John tried to FRBR-ize terminology might have to be taken back to the original. They had trouble with multiple edition statements and using a work or bibliographic resource or what you're cataloging to explain which thing you're cataloging and to which the statement applies.

Agenda item 21. CC:DA site will be changing to include some FRBR links. ALCTS pages are being redesigned by ALA. For now use John's page at Penn State as official CC:DA site.

Agenda item 22. MARBI report. (see handout)

Relator/relationship identifiers were discussed. 21.0D has optional rule to add designations of function. The LCRI says don't apply except illustrator of certain types of materials. The address relationship issues of FRBR, does it need to be broader than the items listed. If interested in task force, email Kristin.

2:00-4:00 Loews - Z - Congress A/B

SAC -- General meeting

Agenda item 16. Report from LC Action Plan forum. Diane Dates Casey heard pretty much the same thing about 2.3 that Shelby reported. She also reported two other sections.

Agenda item 17. Sara S. Lane said the draft report on Subject Reference Structures in Automated Systems will be on the SAC list. Program will be Sat. 1:30-3:30 (conflict with CC:DA) in Toronto.

Agenda item 19. Lori reported being close to a first draft of training materials.

Agenda item 20. Shelby's report on Semantic Interoperability (see above)

Agenda item 21. David discussed the role of or adding liaisons.

Agenda item 22. OCLC MARS will include GSAFD. Bibs in WorldCat is soon going to reflect 2nd ed. of GSAFD.

David brought up the issue of user behavior in searching. PCC wil have a new task group on what a user needs in a bib record - is there any relation here to David's idea. Shelby said it would be easier to focus the Semantic Interoperability program if we don't have to address the psychology of users. Decided to formulate a program.

2:00-4:00 Convention Center – 201B

ALCTS/SS Committee to Study Serials Cataloging - watch for reports