Committee on Cataloging: Description & Access

Task Force on Metadata

Draft Interim Report

Summary
The various charges to the TF were discussed on METAMARDA-L and summarized by a TF member during fall of 1999. The next step is for members to work on subgroups, one for each charge, beginning at Midwinter 1999, and with a report due from each subgroup just prior to the Annual meeting of 1999. An interim report, which will include suggested AACR and USMARC changes to that date, will be presented at Midwinter 2000. A final report will be presented at the Annual meeting of 2000.
History and Background
The Task Force on Metadata was formed at the CC:DA Annual Meeting in summer of 1998. The Task Force's charge is given below, in part III. Summarized, it is to make recommendations concerning how libraries may best provide access to data available over the Web, and more specifically to recommend changes to AACR and to MARBI as needed. These recommendations are to be submitted in final form at Annual of 2000. Mary Larsgaard was appointed chair of the Task Force at the Annual meeting; subsequently, Rebecca Guenther became co-chair. John Attig, as Webmaster for the CC:DA pages, kindly volunteered to set up and maintain the Webpage for the Task Force. Since the task force has eighteen members, it took some time for the chair of CC:DA - Daniel Kinney - to get in touch with ask and to request that each serve on the task force; the formal charge and roster was distributed by Kinney on October 14, 1998. In August, Larsgaard wrote strawperson responses to each of the five charges. Larsgaard also at that time asked Brad Eden to begin moderating discussion of each of the five charges on METAMARDA-L, beginning in early September, and finishing in mid-November. Dr. Eden's summaries of those discussions are given in IV below. Larsgaard distributed the charge summaries to the Task Force members, some of whom had not been on the METAMARDA-L email reflector. In mid-December, acting upon suggestions from Matthew Beacom and Rebecca Guenther, Larsgaard requested that Task Force members select one of the charges numbered one through four and plan on each of these subgroups working on a charge completing a study and report by approximately a month prior to Annual of the year 2000. The Task Force will be meeting Sunday, January 31, 1999, 8:30am-12:30pm, in the Clarion South room of the Brentwood Hotel. The agenda for that meeting is:
1. Call to order
2. Introductions (chair, co-chair, others in attendance)
3. Draft interim
4. Plan of work for next 6 months:
  - have 4 subgroups working on each of the first 4 charges; the 5th charge
  - changes to AACR and USMARC that may be required
  - as part of each of the first 4 charges, as applicable
  - draft report from each subgroup due to metadataf@sdc.ucsb.edu by JUNE 1, 1999
5. Full-group discussion of the five charges
6. Breakup into 4 groups to work on each of the first 4 charges; each of these groups at least temporarily chaired by a task-force member; get as much work done here as possible as far as research/tasks (e.g., formulating sample records;
  - all this to aim toward written report recommending best route(s) for libraries to take in providing access to digital data.
7. Meeting adjourned, sometime between noon and 12:30pm.
A brief oral interim report will be presented by Larsgaard on Saturday, January 30, 1999, at approximately 5pm.
CHARGE AND ROSTER
CC:DA/TF/Metadata/1
CC:DA/TF/Metadata/2
Summary of METAMARDA-L Discussion of Charges
[NOTE: This section still needs considerable editing.]
Charge 1
Adds/changes to charge 1:
- The reference below should be to Robin than Martha.
- I tried to characterize two aspects of what I feel to be user needs from information retrieval systems designed to accommodate networked electronic resources. 1) There will be a need for specialized retrieval for specialized files of materials. Example: coordinate-based searching of spatially-organized materials.
  ... emphasize that the scope for these retrieval techniques (and the metadata to support them) is not restricted to electronic resources.
  ...the place of the "library OPAC" within the range of information services offered by libraries has been changing for about a decade now.
  ...the catalog has been moving steadily away from the center of the map of information services.
  ...we are beginning to have considerable experience with what we are now calling interoperability -- making these different systems work together to provide a service. Nor is metadata inherently different from the kinds of descriptive (and other) information we are used to providing about various parts of our collections.
  ...there is a definite need to bring it all together -- to provide information services that span the entire range of resources.
  ... to characterize two aspects of user needs from information retrieval systems designed to accommodate networked electronic resources.
  1) There will be a need for specialized retrieval for specialized files of materials. Example: coordinate-based searching of spatially-organized materials.
  2) At the same time, there will be a need for generalized retrieval that works over the entire spectrum of diverse materials.
  ... we are not deciding "What are the resource-description needs of libraries today..." but rather, what value does the library's resource description services provide to users in a rapidly changing information culture.
  ... concern that when integrating metadata into our library catalogs, that we do not destroy what has taken well over a hundred years of labor and expense to create, and that user confidence is eroded once a single author appears in a catalog under two forms of name. We should be exploring ways to create a single catalog for us all to catalog against, instead of our current wasteful practice of maintaining thousands of local catalogs, etc.
  ... we might divide up resource information into categorizations of what a library purchases, what a library archives, what a library creates, and what a library "selects." These "buckets" say something about the library's responsibility relative to different categories of material, and that judgements must be made about the relative value and appropriate level of investment in metadata. She also mentions Columbia's modified information systems as an example worthy of exploration.
  ... one of our key concerns is to what extent do we "add value" through our metadata creation/acquisition activities, and raises the question whether we need to systematically treat networked information resources to the same level of bibliographic control as items in the local OPAC. He mentions some of Martha's concerns, and focuses on her challenge to create a coherent information environment for our users. Again, interoperability is what is enabling us to provide different types of metadata thus far.
Charge 2
... almost everyone is commenting on additions to the conceptual map of the resource-description terrain/landscape, given below:
1. LibraryLand Before the Online Catalog:
  Information accessible via:
  - cataloged/indexed collections:
  - card catalog, [BOOK CATALOGS]
  - [PAPER] indexes and abstracts for periodical articles, technical reports, dissertations, etc.
  - [INSTITUTION-SPECIFIC GUIDES TO PERIODICAL TITLES IN BOOK FORM]
  - collections served by [PAPER] finding aids, such as archives & mss.
  - human beings (library staff at ref desks) to translate as needed
  - uncataloged/unindexed collections; generally info found only through human beings
  some more obvious examples of these collections are:
  - pamphlet/vertical files
  - unindexed periodicals
  - archives and manuscripts without finding aids
  - government publications (I realize that many libs. did catalog these) [BUT MANY NOT CATALOGED]
  - slide collections
  - [GOVERNMENT] maps & air photos (again, I realize that a few libs did catalog)
  - what other biggies have I forgotten?
  - [ERIC MICROFICHE]
2. LibraryLand After the Online Catalog and Online Search Services
  - cataloged/indexed collections:
  - online catalog
  - [CD-ROMS (WILSON INDEXES, ETC.) AVAILABLE THROUGH CD-ROM LAN AS WELL AS OTHER CD-ROM PRODUCTS AVAILABLE ON STAND-ALONE TERMINALS]
  - [PAPER] indexes and abstracts for periodical articles, technical reports, dissertations, etc.
  - [PERIODICAL ACCESS BOTH THROUGH INSTITUTION-SPECIFIC TITLES/HOLDINGS INDEXES IN BOOK FORM AND ONLINE ACCESS TO PERIODICAL TITLES/HOLDINGS]
  - [UNION CATALOGS OF SERIAL/PERIODICAL HOLDINGS]
  - collections served by finding aids, such as archives & mss. [SOME ACCESS AVAILABLE ONLINE]
  - human beings (library staff at ref desks) to translate as needed
  - uncataloged/unindexed collections; generally info found only through human beings;
  some more obvious examples of these collections are:
  - pamphlet/vertical files
  - unindexed periodicals
  - [GOVERNMENT MAP SERIES]
  - [OTHER GOVERNMENT PUBLICATIONS, WITH THE SAME STIPULATIONS AS IN a) ABOVE]
  - archives and manuscripts without finding aids
  - slide collections
  - ERIC MICROFICHE
3. LibraryLand After the Web
  - cataloged/indexed collections: [SOME COLLECTIONS STILL UNDERGOING RECON]
  - online catalog [SOME LIBRARIES STILL MAINTAINING PUBLIC CARD CATALOGS]
  - [CD-ROM (LAN OR STAND-ALONE) ACCESS TO PERIODICAL INDEXES, ABSTRACTS, AND FULL TEXT]
  - [PAPER] indexes and abstracts for periodical articles, technical reports, dissertations, etc. [ALSO, ONLINE ACCESS AND THROUGH WEB, SOMETIMES FOR FEE, AND INCORPORATING FULL TEXT]
  - collections served by finding aids, such as archives & mss. [SOME COLLECTIONS UTITLIZING METADATA (EAD) FOR ACCESS]
  - Webpages; these may include standard indexes and catalogs (e.g., MELVYL is available via web); may be searched via: [MOST ACADEMIC LIBRARY CATALOGS AVAILABLE THROUGH THE WEB]
  - Web search engines
  - [THE BEGINNING OF WEB-BASED INTERACTIVE MEDIA]
  - links to pages from collection managers' webpages [ADDITIONAL WEBLINKS TO ADDITIONAL MATERIAL SELECTED BY LIBRARY, WITH OR WITHOUT FEES OR RESTRICTIONS ON ACCESS]
  - a limited number accessed via online catalog [INCREASING CATALOGING OF WEB RESOURCES UTITLIZING TRADITIONAL MARC FORMATS]
  - metadata provided by webpage producers in some few cases [METADATA PROVIDED BY CATALOGERS IN EVEN FEWER CASES]
  - [GOVERNMENT PUBLICATIONS INCREASINGLY AVAILABLE VIA THE WEB, INCLUDING GEOSPATIAL DATA WITH EMBEDDED METADATA]
  - [BEGINNING OF EXPERIMENTS FOR AVAILABILITY OF ERIC DOCUMENTS THROUGH THE WEB]
  - human beings (library staff at ref desks) to translate as needed
  - [DIRECT ACCESS OF WEB INFORMATION BY PATRONS UNGUIDED BY HUMAN INTERVENTION--WEB PAGES MAY OR MAY NOT HAVE EMBEDDED METADATA]
  - [BEGINNING OF PROJECTS TO DIGITIZE PAPER-BASED COLLECTIONS AND MAKE THEM AVAILABLE THROUGH THE WEB]
  - uncataloged/unindexed collections; generally info found only through human beings;
  some more obvious examples of these collections are:
  - pamphlet/vertical files
  - unindexed periodicals
  - slide collections
  - [GOVERNMENT MAP SERIES]
  - [OTHER GOVERNMENT PUBLICATIONS WITH SAME STIPULATIONS AS IN a) ABOVE]
  - [ERIC MICROFICHE]
4. [FUTURE]
  - [INCREASING PROJECTS FOR THE DIGITAL REFORMATTING OF PAPER-BASED COLLECTIONS-INTELLECTUAL PROPERTY RIGHTS NOTWITHSTANDING]
  - [INCREASED USE OF METADATA FROM BOTH THE PUBLISHER/CREATOR PERSPECTIVE, AS WELL AS FROM THE FORMAL RESOURCE DESCRIPTION COMMUNITIES]
  - [MORE SELECTIVE COLLECTIONS OF WEB RESOURCES AVAILABLE LOCALLY AND TAILORED TO SPECIFIC USER COMMUNITIES]
  - [INCREASING NEEDS FOR THE APPLICATION OF STANDARDIZED METADATA FOR STORAGE, RETREIVAL AND DISPLAY-LOCALLY AND GLOBALLY]
  - [SINGLE WEB-BASED INTERFACES AVAILABLE LOCALLY (INCLUDING ONLINE CATALOGS]) THAT WILL UTILIZE A VARIETY OF METADATA VOCABULARIES FOR INTEROPERABILITY ACROSS DIFFERING DATA RESOURCES]
  - [CONTINUING EFFORTS TO MAKE PREVIOUSLY UNCATALOGED MATERIALS AVAILABLE TO LARGER COMMUNITIES, EITHER LOCALLY OR GLOBALLY]
  - [INCREASINGLY COLLABORATIVE EFFORTS TO REALIZE THESE GOALS]
Everyone who commented agreed that we definitely needed a "future" section. LC, in terms of digitizing paper-based collections, is already utilizing current technology to do this, and utilizing the metadata from the catalog record for the original and linking to the electronic item, providing direct access to the metadata through the catalog and to the item itself. Emphasize the possibility in the future of integrating metadata from various sources in various syntaxes all for the common good of resource discovery. Too much generalization of the conceptual map, using slide collections as an example, and dealing with the idea of what "limited number" means. Too much interoperability; what is needed is more possibilities with mappings and crosswalks in regards to metadata integration into existing OPACS. Need to know the many peoples and organizations that create the various resource description locales on the conceptual map, so that we can help tool builders provide integrated access across those locales, through Z39.50 and any future tools that emerge.
Somewhere between b) and c) belongs the proliferation of multiple formats (implied, but not explicit), and there WILL be human beings in the Future, won't there? .... add to the Future section such efforts as OCLC's CORC Project currently under development for the cooperative sharing of metadata (hey, that was my idea!!)? More generally stated than the preceding and more specific than my "increasingly collaborative efforts ... " how about "increase of national/international efforts for the cooperative sharing of metadata."
Charge 3
Martha Yee helped to get things going, by indicating her nervousness regarding the integration of metadata into the library catalog without the type of authority control that libraries have been doing for the last hundred years. She indicated that authority control is at the heart of the interoperability question.(10/12)
Rebecca Guenther provided an excellent response to some of Mary's questions indicated in the initial email concerning different levels of detail/coding of metadata with MARC format.(10/13)
Stu Weibel mentioned the idea of the vCard that is currently used in the Internet community (10/13). This produced a number of questions concerning what the vCard was, whether it could be integrated into library catalogs, who maintains vCard information, etc. Martin mentioned OCLC's exploration regarding the notion of 3 different authority files (10/14). Stu provided more information regarding the vCard in relation to LC NAF records, along with an Internet Library vParable.
Hal reminded us that we have to keep ISBD in mind as well, since it underlies the descriptive specifications of AACR which are embodied in USMARC, and his comments on 10/15 provide an essential summary of comments and concerns regarding Charge #3. Further comments on 10/15 and 10/16 mention that we need to distinguish different scenarios for where the metadata that we'd like to make interoperable with our library catalogs comes from, and Hal presents an interesting scenario as the last comment that I have on Charge #3: Is it unreasonable to suggest that interoperability between USMARC and metadata schemes, while conceptually possible, will destroy the value of the catalog as anything much more than a finding list; and that the better way forward is a single search directed at both a fully-featured catalog and a minimally-featured "metadatabase"?
The heart of the matter, essentially, is authority control.
Charge 4
We had a total of two comments over the previous two weeks regarding Charge #4. Diane Hillman posted Monday the 19th that she fears that we are falling into the same stance as the first report -- responding to fears rather than opportunities, not recognizing how much we've learned from the changes we've already experienced, and not acknowledging our own abilities to cope with those to come. Integrating metadata into our catalogs will require a great deal of creativity on our part, but we have a great deal to gain as well as something to lose, in that process. She also mentions that we risk marginalizing ourselves if we chose to marginalize records created via other metadata schemes, and agrees with Rebecca that we may need to develop some "coping strategies."
Hal Cain replied on Friday the 30th that he wonders whether everyone else is as stumped as he is, and that is why there has been no response to this particular charge. Until we look at some scenarios, he mentions, it may not be possible to formulate recommendations for integrating metadata into "current library methods" or to contemplate possible rule revisions. He mentions Diane's comments and agrees with them.
Charge 5
These last two charges (#4 and #5) had very little discussion, comment, or debate. Consensus is from everyone centers around two points: 1) Stu and Diane indicate that the stable world of AACR2R/MARC just isn't ready to respond yet to the prototypical, developing metadata standards as of yet. It is still too early to move towards changing and tinkering with rules; and 2) I think that many of us, from the silence that has ensued since the discussion of Charge #4 was introduced, really do not know where to start or even begin with effecting rule revision regarding metadata schemes.
It would appear that the general consensus, at this point, is to experiment on ways to integrate these metadata schemes into our information universe without focusing on rule revision/change. I like Stu's statement that we should be "gingerly prototyping new systems that explore the operational ramifications of stepping outside the confines of the traditional OPAC... You have to act and respond." Along with that statement, I would like to add that we, as a profession, need to accept and indeed embrace the fact that we will make mistakes along the way doing this. In order to experiment, one has to make mistakes. In fact, we should be encouraging multiple approaches to the integration of metadata, both into OPACs and outside of OPACS, as the appropriate methodology for dealing with the issue.
With that statement, are we not actively expanding this committee's work outside the arena of CC:DA/MARC? Are we saying that, at this time, rule revision/change to AACR2R and MARC is not recommended? If so, do we need to reexamine our charges? Do we need to reexamine committee membership, expand it to other metadata committees currently engaged in similar discussions? Do we need to disband and recommend that committee members, at this time, should work with other ALA metadata committees that are focusing on issues of experimentation, OPAC and ILS integration of metadata, etc., etc. These are all just comments off the top of my head, to wake everyone up and really think about what we as a committee have been asked to do. After all, we still have a year and a half before we need to give a final report. Have we already effectively stated our opinions and recommendations regarding this area? If we are not recommending any rule revisions/changes at this time already, then there really isn't a reason for us to continue meeting for another 1 1/2 years. Comments, opinions, debate, general apathy, and unruly behavior are encouraged.

Meeting Minutes, 1/30/99

Clarion Suites: Bentwood Room, 8:30-10:30 a.m.

Mary Larsgaard, Chair
Judith Hopkins, recorder

The report on actions to date drafted by John Attig and made available to Task Force members via e-mail in the week prior to the conference will be shortened and presented to CC:DA. The sub-groups designated to deal with the first four charges of the Task Force [see Appendix for text of charges] were asked to have interim reports ready by May 1999.

The subgroups consist of the following persons:

Sub-group I.

Sub-group II.

Sub-group III.

Sub-group IV.

Vianne Sha, a visitor, called to the attention of the Task Group that the unhyphenated word "metadata" has, since 1986, been a registered trademark of the Metadata Company. Only the versions with a hyphen or as two words are not trademarked. Mary Larsgaard said she would check up on this.

The Chair asked the sub-groups to focus on the following topics (based on Matthew Beacom's conceptualization):

Sub-group I: Provide a coherent environment for users, perhaps with the OPAC not too far from the center.

Sub-group II: Focus on the model for the future that was devised during discussion on the Metamarda list prior to the Midwinter Meeting. Diane Hillman said we are not sure what the future environment will look like and will have to make assumptions. Matthew Beacom suggested that the sub-group prepare a bibliograpy of materials that have tried to determine what the future will look like.

Erik Jul suggested that the sub-group should prepare cogent scenarios or models with stated assumptions.

Bill Fietzer said the sub-group should try to make contact with other groups within ALCTS that are also working on metadata issues, e.g., the Committee on Networked Resources and Metadata; on groups outside of ALCTS but within ALA; on library groups outside of ALA, e.g., OCLCs CORC Project; and groups outside of libraries, e.g., the Alexandria Digital Library, ASICOMAR (in the database research community), and the Metadata Working Group which is trying to determine what metadata elements are needed to describe unique identifiers. Rebecca Guenther, the Task Force co-chair, wondered if there is any effort on the part of ALA to bring these various groups together. Bruce Johnson noted that, within ALCTS, the CCS Executive Committee would be looking for recommendations from its SAC subcommittees on metadata as well as from our Task Force. Someone made the point that the sub-groups recommendations should be theoretically sound and practically feasible; cataloging should be both practical and cost efficient.

Sub-group III. Consider the library in the context of a larger or more general information environment or environments.It should provide information on what is being done in this area by various communities, e.g., the geo-spatial community. Authority control is one important issue for this group to consider as it is a library speciality.

Bill Fietzer asked if they should restrict their attention to cataloging standards, e.g., the cataloging standard used by the museum community in AMICA, or the Getty Information Institute publication edited by Martha Baca Introduction to Metadata: Pathways to Digital Information which describes standard ways to describe objects.

Erik Jul said that raised the question of whether cataloging is a part of metadata or whether metadata is part of cataloging. In response Judith Hopkins called attention to the forthcoming book by Arlene Taylor The Organization of Information which treats cataloging as a type of metadata.

Mary Woodley asked if the sub-group should restrict itself to the description of electronic resources. Judith Hopkins said No, but Bill Fietzer said the task could become unmanageable unless they did so limit themselves. Mary said libraries have always been involved in selecting whether we catalog things fully, minimally, or not at all.

Sub-group IV. This sub-group was asked to make a list of prototypes and to determine conceptually sound ways of evaluating them. As Mary Woodley said, we need to come up with something that can help us provide better service to users of our libraries.

Erik Jul phrased it as both doing new things and doing old things in ways in which we havent done them before. By identifying what is inside a universe, you also identify what is outside it and that is more important.

Mary Woodley said the focus should be on how to make information available to users, starting with what doesn't work with our current methods.

Brad Eden said the word "current" in charge 4 is too restrictive.

It was decided that sub-group IV should focus on user needs.

The Task Force then broke up into concurrent meetings of the sub-groups with visitors participating in the sub-group of their choice. When the Task Force reunited half an hour later the sub-groups made the following reports on what they had decided to do.

Report of Sub-Group I on Charge 1 (Matthew Beacom reported): The Group decided to review existing principles (e.g., Cutter, Ranganathan) and to determine fundamental principles that inform the aims of all libraries, to define users and their expectations, and to look at the context in which we are describing our resources, i.e., to review the mechanisms we are going to use to deliver resources, e.g., cataloging, indexes, webpages, etc.

Report of Sub-Group II on Charge 2 (Diane Hillman reported): The Group decided to look at list of information [pieces?] and see how they have changed over time; see what models have worked well.

Report of Sub-Group III on Charge 3 (Bill Fietzer reported): The Group plans to devise definitions of Metadata and Interoperability, and then to take these definitions and see how they apply to existing cataloging rules and USMARC (now MARC 21).

Report of Sub-Group IV on Charge 4 (Brad Eden reported): The Group decided to look at the user tasks described in the IFLA document Functional Requirements for Bibliographic Records and then to try to compile a world-wide list of prototypes and to evaluate them. (They defined prototype as "Virtual search access to integrated information and relevant retrieval from the user point of view".)

It was agreed that the Sub-Groups should communicate over the Metamarda List (METAMARDA-L@CORNELL.EDU), placing the Sub-Group Name in the subject line. When they have something ready for the world to see they should send it to John Attig who will post it on the Metamarda web archives.

The meeting was adjourned at 10:34 a.m.

APPENDIX: Text of charges:

The Task Force on Metadata is charged with, but not limited to:

Analyzing resource-description needs of libraries, seeking input from interested librarians and discussion via the metamarda-l email reflector;
Building a conceptual map (or maps) of the resourc-description terrain / landscape and developing models for accessing / using metadata both within and outside the library community;
Devising a definition of metadata and investigating the interoperability of newly emerging metadata schemes with the cataloging rules (AACR2R) and the USMARC format;
Recommending ways in which libraries may best incorporate the use of metadata schemes into the current library methods of resource description and resource discovery; and
Recommending, as needed, rule revision to enable interoperability of cataloging (with AACR2R) with metadata schemes.

The Task Force shall present CC:DA with interim reports at the 1999 ALA Midwinter Meeting, the 1999 Annual Conference, and the Midwinter Meeting in 2000. The final report of the Task Force shall be presented to CC:DA at the Annual Conference in 2000 and shall be sent to the Chair of CC:DA no later than June 1, 2000.