CDC Notes


Jump to: Eficiency Recommendations Cataloging and Metadata: Continuity and Change Core Mission Remains the Same While cataloging tools and resources are changing rapidly, the essential mission and values remain the same. The challenge is to analyze tools and workflows, and separate the essential skills from the specific techniques in which they are embedded. One of the ways we've approached this is to maintain an annotated bibliography that supports our ongoing discussion and analysis. I'm providing a selection from this bibliography as a handout. These include opinion pieces and studies by Thomas Mann, Deanna Marcum, Karen Calhoun, Roy Tennant, Barbara Tillett, Arlene Taylor, and others. Knowledge management and authority control still essential It is necessary to respect the past and still acknowledge the ongoing revolution in information technology and find ways to innovate, and this is reflected in our new mission, vision, and goals statement. This document is posted in the Goals section of the Catalog Department home page. What do catalog and metadata librarians do? Traditionally, we do a kind of indexing and abstracting operation with book in hand, creating machine-readable bibliographic surrogates. These surrogates include descriptions, subject analysis, controlled access points for names, places, meetings, etc., and are constructed according to AACR2, MARC21, and other national and international standards. Semantic Interoperability as "Deliverable". Reconciling vocabularies It is vital that Catalog and Metadata Services be involved from the ealiest stages of new initiatives, given continued need to reconcile diverse vocabularies and metadata schemas, design and create catalogs that organize huge amounts of diverse information across academic disciplines. Reconciling heterogeneous metadata MARC has been instrumental to the networked catalog environment used throughout the world. It is no longer enough, however. Today we need to be absorb MODS and MARC-XML from peer libraries, ONIX from publishers, EAD and OAI from mansuscript and archival metadata sources, TEI from the computational linguistics people, and Dublin Core from Web creators in various fields. The underlying exchange language is almost always XML, and it is therefore imperative develop as much XML expertise on staff as possible. Thesauri and crosswalks between and among diverse metadata schemas will be necessary to the success of our online repository. Support federated searching (Cyberinfrastructure Report) The top recommendations from the Provost's Cyberinfrastructure Survey Report (10/3/06) involve metadata that are both robust and precise. Three of the four highest-rated requests speak to the library and catalog department implicity, namely: (1) "Easier electronic access to scholarly materials," (2) "Providing students with digital access to research and instructional materials," and perhaps most importantly, (4) "Providing better search tools to locate materials across all of Yale's holdings and collections." create space for innovation We will continue providing reliable, authoritative access to all materials selected for the Yale library. These include books and serials, and increasingly e-books, networked databases, websites, blogs, and other eletronic media. In FY 2005/2006, we cataloged more than 80,000 titles and 137,000 volume equivalents. We'll need to continue processing print materials at this scale or greater, while at the same time unlocking older collections and expanding bibliographic control over electronic media. The vast majority of these titles were print and microfilm books and serials, but we are mindful that Metalib and FEDORA are gathering steam, and know that these will consume a growing portion of our time and attention. Decoupling of content and presentation is dominant feature of new networked environment. Consider RDA (versus a AACR2/ISBD punctuation/MARC21, with display standards to be moved to appendices) and XML/XHTML semantic tags (versus HTML style tags). James Hilton refered to this phenomenon as "unbundling"; Lorcan Dempsey as the "recombinant library"; Gregory Crane (of Tufts, and Perseus) describes how the network level lets books "talk to each other", especially when metadata is provided at all levels: network, collection, chapter and verse; (Similar to push for XHTML-compatibility; HTML ties content and presentation together too closely; XML (or XHTML) does not). How do we use the unbundling phenomenon to our own advantage? Start by taking control over our own data where ILS isn't serving our needs. This means embracing OSS opportunities such as eXtensible Catalog and Evergreen. Database driven websites, content management systems, wikis and RSS feeds (as opposed to static web pages) are features of the recombinant library. It also means sharpening our understanding and use of link resolvers. Someone at code4lib put it like this: "Don't catalog; Resolve". CoiNS (ContextObjects in Spans), unAPI, and other link-resolving and interoperability tools, are instrumental to "effective and integrated access to scholarly resources," as expressed in the Library's Vision Statement . In the "Google Economy" links are citations, the handles by which we identify objects of interest and conversation. HOW CAN WE MANAGE OUR current workload more efficiently? Training XML-oriented Yale catalogers need to overhaul skills and knowledge of emerging metadata standards. This means RDA, but also MARC-XML, MODS, and other XML based storage and retrieval formats. In order to free up time for new training and new assignment, we must find ways to expedite our current workflows. Training must be Project-based "Expert User" model Desginated staff should liaison with the IAC and Metadata Services and manage XML applications and non-MARC metadata schemas across cataloging units, at least during transitional period until general proficiency level has been raised. This group could form nucleus of the Virtual Metadata Services Group proposed b yDPIP Production and Content Integration Working Group's Final Report (p. 21) Attend wider range of metadata conferences and workshops Targeted learning plan funds Encourage metadata-related SCOPA forums and grant applications Assume certain database metadata responsibilities from ILTS (e.g., CatStat reports, Metalib IRDs). Efficiency Recommendations Make non-PCC Core default cataloging level; IS THIS RIGHT? Our recommendation is not to lessen the amount or frequency of authority control. Rather, we suggest a change of emphasis. Authority control should be expanded in the sense that we need to apply it beyond the traditional catalog. MetaLib will need more bibliograpic control as it expands, for example, as will FEDORA. Revisit Criteria for LSF transfers; Reconsider shelflisting (collocation) practices; SRA points out that classification is essential to success of Endeca's faceted navigation (see NCSU catalog). Expand range of duties for E-level staff; and Reconsider treatment of certain titles acquired both in Print and electronic form. Shelf-ready books? But this presupposes shelflisting changes, since otherwise available source copy limited to LC. Long term goal: High-quality vendor-supplied original cataloging New Name Catalog and Metadata Services ... to be announced officially on March 1, following two public forums. Increased Opportunities for Collaboration Common mission of bibliographers and catalogers is to provide efficient access to authoritative information. This has been true for a long time. Today more than ever. Selection and Access Relationship of Cataloging and Collection Development ... Martha Yee has written, in response to the suggestion that Google/Amazon can replace the library catalog ... "catalogs currently ... provide evaluative information, in that the presence of a work in the collection of a major research library implies that that work was deemed of scholarly value. " FEDORA Finding Aids Integration of various findings aids databases in e-repository will require investment in vocabulary reconcilliation and crosswalks. Sakai Metalib Collection Security as Growing Concern "Irrational Exuberance", or, Beware Utopian Digital Dreams Everything Digital? Noise-to-Signal ratio Not everything is being digitized. Even if everything were being digitized, artifacts have their own inherent value that must be treated respectfully on their own terms. They must be accessible and preserved in their own right. As an ALCTS CCS EC statement puts it: "There are no 'transitional' users. Library users comprise both those who sometimes, or always, prefer non-digital resources and those who now prefer digital resources, among others. All are to be treated with equal respect and without condescension. Similarly, the concept of "legacy materials" needs to be treated with caution. Whatever the values conferred by digitization of "non-born-digital" resources, we realize that those are added values, beyond those inherent in the resources as originally created. In other words, resources which are not universally available are not inherently of lesser value. We do not assume that non-digital resources are simply waiting to be digitized. World Wide Web as Vanity Press Digital Preservation Assuming all of Yale's holdings were indeed digitized, the issue of digital preservation has not been adequately addressed. In any case, digital preservation begins at the point of object capture/creation. Intellectual Property Rights Similarly, given the current copyright regime, it is unclear whether items scanned by Google et al., will be searchable by other libraries. Usability of e-Resources Moreover, it is unclear that mongoraphs will ever give way to e-books. Thomas Mann makes a good case for the monograph as the continuing gold standard in scholarly communication, requiring extraordinary concentration on part of aurhors and readers, and encouraging well-organized, comprehensive treatment of topics. Quotations Increased Emphasis on Unique Holdings "Human intervention in cataloging will shift to a focus on unpublished, often uncataloged material—material that fills the shelves of special collections, archives and institutional storage facilities" PCC Mission Statement Task Force assumption 8 (2005) "In allocating cataloging resources, invest fewer resources in describing easily discoverable materials (common titles on Amazon, eg) and more resources in describing unique specialized materials (special collections, abstruse languages, etc)" . (University of California Task Force. (Dec. 2005) p. 51. Cataloging in the Age of Google "... in the age of digital information … how much do we need to continue to spend on carefully constructed catalogs?" Deanna Marcum's 2005 The Future of Cataloging ". "Technology tends to unbundle bundled endeavors; and, man, is education bundled" --James Hilton, CIO University of Virginia; heard at Definitely Digital symposium. "When books start to talk to each other ..." -Gregory Crane, editor of Perseus project, Tufts University. "Even if all other aspects of the system worked perfectly, poor quality metadata would degrade the quality of the resulting library." (p. 3) ... Minimally descriptive metadata, like Dublin Core, is still minimally descriptive even after multiple quality repairs. We suggest that the time spent on such format-specific transforms might be better spent on analysis of the resource itself--the source of all manner of rich information" [which sounds like a plea for more expert cataloging] (p. 7). -- -- Lagoze, Carl, et al. (2006). "Metadata Aggregation and 'Automated Digital Libraries' : a Retrospective on the NSDL Experience." Retrieved from http://arxiv.org/abs/cs.DL/0601125 on 7/12/06. Metadata in the Digital Library "Much hope is placed on the ability of an institutional repository to rescue scholarly communication, yet in no way can it become part of this conceptual global system unless there is a strong backbone of cataloging and metadata" (p. 5). Indiana University White Paper (Jan. 2006) "To be successful in the long run, technical services must play a central role in digital library design and development and in e-resource management .” (Karen Calhoun, LRTS, 2003) "Catalogers must learn several metadata schemes and organizational structures beyond AACR and the MARC record. They must free themselves from thinking in terms of flat files and linear access and begin to think in terms of multi-scheme data registries, new record constructs, and relational data models… They must envision a new spectrum of authority control that includes many types of identifiers along with the more familiar names, titles, and subjects. And most critically, catalogers must actively participate in the development of system architectures and data registries. Only this level of activity will ensure that catalogers play a key role in the development of authority control systems for electronic resources.” (Sherry Vellucci, UCSD presentation, March 2006 Jan. 2000 LRTS article (44:1; 33-43) on the 2nd slide) Metadata at Yale "DPIP can encourage staff working on specific projects to ensure that their metadata is reusable by higher-level systems such as harvesters and compatible with external projects such as the DLF Aquifer initiative." (DPIP Report p. 24) "Strategic planning groups in the library advocated a 'federated approach' to Yale's integrated library and argued that a separate digital library unit would create a digital elite and discourage mainstreaming of digital activities." (DPIP Report p. 11) "Easier electronic access to scholarly materials" ... "Providing students with digital access to research and instructional materials," ... "Providing better search tools to locate materials across all of Yale's holdings and collections." Yale Provost's Cyberinfrastructure Survey Report (10/3/06) "To segregate this intellectual work based on format would require the Library to build costly redundant systems and limit our agility in dealing with shifting priorities." Catalog Dept. Strategic Directions, 2007. "Form partnerships on and off campus ... expand staff expertise in non-MARC metadata, with understanding that "the best training is by doing" (p. 17) ... study and prepare for evolution of OPAC, including decisions about whether tiered levels of cataloging are being applied appropriately." Indiana University White Paper (Jan. 2006) top

Cataloging and Metadata: Continuity and Change

Core Mission Remains the Same

While cataloging tools and resources are changing rapidly, the essential mission and values remain the same. The challenge is to analyze tools and workflows, and separate the essential skills from the specific techniques in which they are embedded. One of the ways we've approached this is to maintain an annotated bibliography that supports our ongoing discussion and analysis. I'm providing a selection from this bibliography as a handout. These include opinion pieces and studies by Thomas Mann, Deanna Marcum, Karen Calhoun, Roy Tennant, Barbara Tillett, Arlene Taylor, and others.

Knowledge management and authority control still essential

It is necessary to respect the past and still acknowledge the ongoing revolution in information technology and find ways to innovate, and this is reflected in our new mission, vision, and goals statement. This document is posted in the Goals section of the Catalog Department home page.

What do catalog and metadata librarians do? Traditionally, we do a kind of indexing and abstracting operation with book in hand, creating machine-readable bibliographic surrogates. These surrogates include descriptions, subject analysis, controlled access points for names, places, meetings, etc., and are constructed according to AACR2, MARC21, and other national and international standards.

Semantic Interoperability as "Deliverable".

Reconciling vocabularies

It is vital that Catalog and Metadata Services be involved from the ealiest stages of new initiatives, given continued need to reconcile diverse vocabularies and metadata schemas, design and create catalogs that organize huge amounts of diverse information across academic disciplines.

Reconciling heterogeneous metadata

MARC has been instrumental to the networked catalog environment used throughout the world. It is no longer enough, however. Today we need to be absorb MODS and MARC-XML from peer libraries, ONIX from publishers, EAD and OAI from mansuscript and archival metadata sources, TEI from the computational linguistics people, and Dublin Core from Web creators in various fields. The underlying exchange language is almost always XML, and it is therefore imperative develop as much XML expertise on staff as possible.

Thesauri and crosswalks between and among diverse metadata schemas will be necessary to the success of our online repository.

Support federated searching (Cyberinfrastructure Report)

The top recommendations from the Provost's Cyberinfrastructure Survey Report (10/3/06) involve metadata that are both robust and precise. Three of the four highest-rated requests speak to the library and catalog department implicity, namely: (1) "Easier electronic access to scholarly materials," (2) "Providing students with digital access to research and instructional materials," and perhaps most importantly, (4) "Providing better search tools to locate materials across all of Yale's holdings and collections."

create space for innovation

We will continue providing reliable, authoritative access to all materials selected for the Yale library. These include books and serials, and increasingly e-books, networked databases, websites, blogs, and other eletronic media. In FY 2005/2006, we cataloged more than 80,000 titles and 137,000 volume equivalents. We'll need to continue processing print materials at this scale or greater, while at the same time unlocking older collections and expanding bibliographic control over electronic media.

The vast majority of these titles were print and microfilm books and serials, but we are mindful that Metalib and FEDORA are gathering steam, and know that these will consume a growing portion of our time and attention.

Decoupling of content and presentation is dominant feature of new networked environment. Consider RDA (versus a AACR2/ISBD punctuation/MARC21, with display standards to be moved to appendices) and XML/XHTML semantic tags (versus HTML style tags). James Hilton refered to this phenomenon as "unbundling"; Lorcan Dempsey as the "recombinant library"; Gregory Crane (of Tufts, and Perseus) describes how the network level lets books "talk to each other", especially when metadata is provided at all levels: network, collection, chapter and verse; (Similar to push for XHTML-compatibility; HTML ties content and presentation together too closely; XML (or XHTML) does not). How do we use the unbundling phenomenon to our own advantage? Start by taking control over our own data where ILS isn't serving our needs. This means embracing OSS opportunities such as eXtensible Catalog and Evergreen. Database driven websites, content management systems, wikis and RSS feeds (as opposed to static web pages) are features of the recombinant library.

It also means sharpening our understanding and use of link resolvers. Someone at code4lib put it like this: "Don't catalog; Resolve". CoiNS (ContextObjects in Spans), unAPI, and other link-resolving and interoperability tools, are instrumental to "effective and integrated access to scholarly resources," as expressed in the Library's Vision Statement . In the "Google Economy" links are citations, the handles by which we identify objects of interest and conversation.

HOW CAN WE MANAGE OUR current workload more efficiently?

Training

XML-oriented

Yale catalogers need to overhaul skills and knowledge of emerging metadata standards. This means RDA, but also MARC-XML, MODS, and other XML based storage and retrieval formats. In order to free up time for new training and new assignment, we must find ways to expedite our current workflows.

Training must be Project-based

"Expert User" model

Desginated staff should liaison with the IAC and Metadata Services and manage XML applications and non-MARC metadata schemas across cataloging units, at least during transitional period until general proficiency level has been raised. This group could form nucleus of the Virtual Metadata Services Group proposed b yDPIP Production and Content Integration Working Group's Final Report (p. 21)

Attend wider range of metadata conferences and workshops

Targeted learning plan funds

Encourage metadata-related SCOPA forums and grant applications

Assume certain database metadata responsibilities from ILTS (e.g., CatStat reports, Metalib IRDs).

Efficiency Recommendations

Make non-PCC Core default cataloging level;
IS THIS RIGHT?
Our recommendation is not to lessen the amount or frequency of authority control. Rather, we suggest a change of emphasis. Authority control should be expanded in the sense that we need to apply it beyond the traditional catalog. MetaLib will need more bibliograpic control as it expands, for example, as will FEDORA.

Revisit Criteria for LSF transfers;

Reconsider shelflisting (collocation) practices; SRA points out that classification is essential to success of Endeca's faceted navigation (see NCSU catalog).

Expand range of duties for E-level staff; and

Reconsider treatment of certain titles acquired both in Print and electronic form.

Shelf-ready books?
But this presupposes shelflisting changes, since otherwise available source copy limited to LC. Long term goal: High-quality vendor-supplied original cataloging

New Name

Catalog and Metadata Services ... to be announced officially on March 1, following two public forums.

Increased Opportunities for Collaboration

Common mission of bibliographers and catalogers is to provide efficient access to authoritative information. This has been true for a long time. Today more than ever.

Selection and Access

Relationship of Cataloging and Collection Development ... Martha Yee has written, in response to the suggestion that Google/Amazon can replace the library catalog ... "catalogs currently ... provide evaluative information, in that the presence of a work in the collection of a major research library implies that that work was deemed of scholarly value. "

FEDORA

Finding Aids

Integration of various findings aids databases in e-repository will require investment in vocabulary reconcilliation and crosswalks.

Sakai

Metalib

Collection Security as Growing Concern

"Irrational Exuberance", or, Beware Utopian Digital Dreams

Everything Digital?

Noise-to-Signal ratio

Not everything is being digitized. Even if everything were being digitized, artifacts have their own inherent value that must be treated respectfully on their own terms. They must be accessible and preserved in their own right. As an ALCTS CCS EC statement puts it: "There are no 'transitional' users. Library users comprise both those who sometimes, or always, prefer non-digital resources and those who now prefer digital resources, among others. All are to be treated with equal respect and without condescension. Similarly, the concept of "legacy materials" needs to be treated with caution. Whatever the values conferred by digitization of "non-born-digital" resources, we realize that those are added values, beyond those inherent in the resources as originally created. In other words, resources which are not universally available are not inherently of lesser value. We do not assume that non-digital resources are simply waiting to be digitized.

World Wide Web as Vanity Press

Digital Preservation

Assuming all of Yale's holdings were indeed digitized, the issue of digital preservation has not been adequately addressed. In any case, digital preservation begins at the point of object capture/creation.

Intellectual Property Rights

Similarly, given the current copyright regime, it is unclear whether items scanned by Google et al., will be searchable by other libraries.

Usability of e-Resources

Moreover, it is unclear that mongoraphs will ever give way to e-books. Thomas Mann makes a good case for the monograph as the continuing gold standard in scholarly communication, requiring extraordinary concentration on part of aurhors and readers, and encouraging well-organized, comprehensive treatment of topics.

Quotations

Increased Emphasis on Unique Holdings

"Human intervention in cataloging will shift to a focus on unpublished, often uncataloged material—material that fills the shelves of special collections, archives and institutional storage facilities" PCC Mission Statement Task Force assumption 8 (2005)

"In allocating cataloging resources, invest fewer resources in describing easily discoverable materials (common titles on Amazon, eg) and more resources in describing unique specialized materials (special collections, abstruse languages, etc)" . (University of California Task Force. (Dec. 2005) p. 51.

Cataloging in the Age of Google

"... in the age of digital information … how much do we need to continue to spend on carefully constructed catalogs?" Deanna Marcum's 2005 The Future of Cataloging ".

"Technology tends to unbundle bundled endeavors; and, man, is education bundled" --James Hilton, CIO University of Virginia; heard at Definitely Digital symposium.

"When books start to talk to each other ..." -Gregory Crane, editor of Perseus project, Tufts University.

"Even if all other aspects of the system worked perfectly, poor quality metadata would degrade the quality of the resulting library." (p. 3) ... Minimally descriptive metadata, like Dublin Core, is still minimally descriptive even after multiple quality repairs. We suggest that the time spent on such format-specific transforms might be better spent on analysis of the resource itself--the source of all manner of rich information" [which sounds like a plea for more expert cataloging] (p. 7). -- -- Lagoze, Carl, et al. (2006). "Metadata Aggregation and 'Automated Digital Libraries' : a Retrospective on the NSDL Experience." Retrieved from http://arxiv.org/abs/cs.DL/0601125 on 7/12/06.

Metadata in the Digital Library

"Much hope is placed on the ability of an institutional repository to rescue scholarly communication, yet in no way can it become part of this conceptual global system unless there is a strong backbone of cataloging and metadata" (p. 5). Indiana University White Paper (Jan. 2006)

"To be successful in the long run, technical services must play a central role in digital library design and development and in e-resource management .” (Karen Calhoun, LRTS, 2003)

"Catalogers must learn several metadata schemes and organizational structures beyond AACR and the MARC record. They must free themselves from thinking in terms of flat files and linear access and begin to think in terms of multi-scheme data registries, new record constructs, and relational data models… They must envision a new spectrum of authority control that includes many types of identifiers along with the more familiar names, titles, and subjects. And most critically, catalogers must actively participate in the development of system architectures and data registries. Only this level of activity will ensure that catalogers play a key role in the development of authority control systems for electronic resources.” (Sherry Vellucci, UCSD presentation, March 2006 Jan. 2000 LRTS article (44:1; 33-43) on the 2nd slide)

Metadata at Yale

"DPIP can encourage staff working on specific projects to ensure that their metadata is reusable by higher-level systems such as harvesters and compatible with external projects such as the DLF Aquifer initiative." (DPIP Report p. 24)

"Strategic planning groups in the library advocated a 'federated approach' to Yale's integrated library and argued that a separate digital library unit would create a digital elite and discourage mainstreaming of digital activities." (DPIP Report p. 11)

"Easier electronic access to scholarly materials" ... "Providing students with digital access to research and instructional materials," ... "Providing better search tools to locate materials across all of Yale's holdings and collections." Yale Provost's Cyberinfrastructure Survey Report (10/3/06)

"To segregate this intellectual work based on format would require the Library to build costly redundant systems and limit our agility in dealing with shifting priorities." Catalog Dept. Strategic Directions, 2007.

"Form partnerships on and off campus ... expand staff expertise in non-MARC metadata, with understanding that "the best training is by doing" (p. 17) ... study and prepare for evolution of OPAC, including decisions about whether tiered levels of cataloging are being applied appropriately." Indiana University White Paper (Jan. 2006)

top

This file last modified 01/25/07