THURSDAY–-June 23
7:30AM -7:00 Registration
Basic
creation of name and title authorities
Trainee
manual will be used in training CFL staff and ND Funnel participants.
FRIDAY--June 24
Basic
creation of name and title authorities
SAC Subcommittee on Semantic Interoperability – meeting
Tony had proposed some changes to the
Introduction/criteria document. These were discussed. The agreed-to changes
will be made and posted to the web page in the next few weeks. We have seen
some changes to the glossary since the last meeting, but nothing major. Lois
made some notes for Ruth to review. The list of projects was discussed. We
can't find current information on VILIB although it must have existed at the
time it was put on the list. We will try one more time to find information on
it. If nothing is found, it will be removed. Hittite and Hasset are monolingual
and will be removed.
Bonnie and Shannon reviewed the
guidelines and made some suggestions.
It is expected work can be completed
via email/listserv and there should be no need for a meeting at Midwinter.
Meeting adjourned.
7:30P-10:00 Hilton --
OLAC Cataloging Policy Committee (CAPC)
FAST Subcommittee
Talked with Eric Childress
and Shannon Hoffmann - we will be put in contact with Susan Westberg
of OCLC to test: a) side bar in Office2003 for the FAST file and b) ability to
add to records in ContentDM
Ed O'Neill reported OCLC
has been doing cleanup rather than "new" stuff. They are looking
algorithms to make them more effective. They will be moving ahead with the
remaining facets. See handout.
Arlene did some work with
library science students. OCLC make a database of library science topics for
them to work with. Generally over half of them got the same results. Shannon
(BYU) tried using FAST in their projects. They verified realy
headings in LCSH and then input into the terminology list in ContentDM. They would like FAST in ContentDM: a) trying to get to the point that person
with digital images could preview words that match the image before the library
starts working on the metadata; b) their standard is LCSH and when it isn't
there they submit to NACO/SACO which would eventually come back in FAST
Some
issues discussed: Translations vs. translated into; $x after a topic heading
should stay with it, but disappears after all other type of headings and
becomes a separate heading - works well much of the time, but not always; geographics are geographic, form are form - FAST does not
mix types; geographics follow GAC structure - in some
cases the result is uneven; geographics serve to
describe location, topic, and to disambiguate, but this is not always achieved
when separated from their topics.
Test:
Pittaccess/pittaccess to search headings the students
used; try cataloging using FAST
SATURDAY--June 25
CC:DA program -- Cataloging cultural objects (
See handout.
Elisa Lanzi. Artistic creations/works is the focus of the new CCO standard but it also
applies to photos, archival documents, built works, images, installations,
clothing, archeological materials, manuscripts, performance. The goal is to
promote descriptive practe standards and share documentation: data structure,
data values, data content, international content standards, metadata
structures.
CCO follow VRA core; uses AAT, ULAN, TGN. It provides
guidance on how to populate data fields based on VRA core and CDWA. It will map
to Dublin Core and MARC21. It provides guidelines for selecting, ordering, and
formating data and syntax. It is intended to compliment AACR and DACS (the new
APPM). It brings authority control to the VRA group. Publication expected
spring 2006.
CDWA = Categories for the Description of Works of Art
http://www.vraweb.org/CCOweb/
will have examples, training materials, but the actual standard will need to be
purchased from ALA. They are cataloging real and virtual objects and model
relationships. They use multiple toolsets - content standards, subject
thesauri, etc. XML schema at Getty has been set up for OAI harvesting metadata.
Ann Whiteside. There are 17 categories (similar to Dublin Core) in VRA 3.0 and 3.1 in
CDWA - CCO grew out of both. It is intended to be used to catalog single
objects.
|
AUTHORITY RECORDS |
|
|
|
Person/corporate |
|
Image record |
|
Geographic |
Work record |
|
|
Concept |
|
|
|
Subject aboutness |
|
Source record |
|
Subject isness |
|
|
"Class" is terms from an authority file.
"Notes" are general descriptions and physical descriptions.
There are 10 top principles listed on website
Libraries are crosswalking their metadata to MACR21 for the
local OPACs.
Maria Oldal. She feels MARC works well to put CCO into. She has done it for nearly
everything successfully.
Note: can ContentDM accommodate CCO? Can I crosswalk? Could
we create a separate base in Aleph and use a separate edit_doc_999 to
accommodate display and not activate authority control?
Object description
Object naming: Workk
type: ojbect. MARC mapping: 245h, 300a, 655a, 246 allows variant names or 247
former names - they capitalize titles in 245
Authority control: makes sense in CCO except for 245 titles -
they are too fluid and locally created. "Main" entry is not used -
just all creators and their role; information can come from any valid source.
Anonymous (or similar work) is prescribed - e.g. Anonymous $c Nederlandische $d
16th century $e artist
Stylistic and chronological information: concept of school,
style creator's place or culture (rather than the object's place of creation -
often too hard to determine)
Subjects: use from authorized sources but not in strings and
they use both plural (for multiple objects) and singular (for a single object)
forms.
See: http://www.getty.edu/research/conducting_research/standards/intrometadata/
Jonathan Furner. Subject access.
Subject access is 3-fold: description, identification,
interpretation. First they write the description. Next they created the 3 types
of subject headings as keywords but from authorized sources.
Example: flowers in a vase
subject description: still life, flowers, peonies, primrose,
tulips, roses, wine, ledge
subject interpretation: senses, smell, beauty, life, Passion
of Christ (scholars say that is what the painting portrays)
Example: Civil War picture of
subject description: portraits, army, officer, president,
tent, camp site
subject identification: NACO form of name for each person,
Civil War,
"Class" can be assigned from Dewey Decimal
Classification because the number and its caption are semantically always the
same
Challenges: What does "subject" mean? What kinds of property of works
should be indexed? What is used to determine "subject"? Need to
select and do authority control and identify the metadata elements.
1. Subjects: e.g. people, things, events, places, concepts
- objects (works) - e.g. art works, buildings, documents,
collections. A) descriptive cataloging: what objects are; b) subject
cataloging: what objects are of / about
- images - e.g. photos, slides, digital files. a) descriptive
cataloging: what images are: what images are of; b) subject: what
images are about.
- texts - e.g. books. Descriptive: are and subject: about
- representatal (figurative) - e.g. narrative stories,
episodes, non-narrative: people, animals, activities, places
- non-representational - e.g. abstract works, buildings,
furniture, decorative arts. Subject: content, meaning, form/composition,
function, purpose, use
Results in ofness and aboutness: generically of is
"descriptive" and specifically of is
"identification" and about is "interpretation"
Example
Descripptive: nude woman holding a dagger
Identification: suicide of Lucretia
Interpretation: virtuousness
2. Determine subject analysis
Offness - who
what when; work from generic to specific
Aboutness -
meaning, what is expressed by the work, intention of the creator,
interpretation by scholars, historical what does it symbolize?
3. Term selection - what kinds? How many? Depends on:
resources you have, needs of users, importance of work, unusual details,
capabilities of system (e.g. ability to link to broader/narrower terms)
Note: it is better to be right about generalness than to be
wrong about specifics
4. Authority control
Use preferred names of real people and real places
Use preferred names of genre terms (from various thesauri)
Use preferred forms of generic subject terms (")
Use resources such as AAT, Inconclass
http://www.incolass.nl, etc.; also see
CCO for list of sources for terms
Link you 6XXs within your ILS
5. Metadata element sets - subject metadata elements in CCO
a) description - 1/record, free text
b) subjects - required, controlled, repeatable
c) extent - designating what part of object the subject
applies to
d) subject type (descriptive, identification,
interpretation)
"iconographical research into aboutness"
- something that needs to be done to determine subjects - history of the
object. Can try AMICO database (digital images) and use the notes there for
subject interpretation
Challenge: in-progress works. E.g. a performance with
different actors over time, an installation as it grows over time
9:30-12:30 Hyatt -- Grand Ballroom F
MARBI -- http://www.loc.gov/marc/marbi/
Proposal No.
2005-04R: Hierarchical Geographic Names in the MARC 21 Bibliographic
Format
Summary: This paper proposes expanding the definition of
Field 752 (Added Entry — Hierarchical Place Name), adding new subfields to 752
and the newly defined 662 and making some current subfields repeatable to
enable a hierarchical approach to subject-oriented geographic coverage.
MARBI action taken: Approved with significant editorial changes. Discussed
shifting subfields to allow for $e being relator.
Proposal No. 2005-07:
Revision of subfield $b in field 041 in the MARC 21 Format for Bibliographic
Data
Summary: This paper proposes changing the coding convention
of field 041 (Language code) subfield $b (Language code of summary or
abstract/overprinted title or subtitle) for audiovisual materials by removing
the phrase “when they [the languages of subtitles] differ from the language of
the soundtrack.” It also proposes changing the terminology used for subtitles
and clarifies that captions are also covered in the definition of subfield $b.
MARBI action taken: Approved with minor changes
Proposal No. 2005-08:
Changes to accommodate IAML coded data in bibliographic fields 008/18-19, 047
and 048
Summary: This proposal discusses the changes needed to
incorporate IAML form and genre, and medium of performance codes in MARC 21.
MARBI action taken: Broke variable fields out. They want to restrict
these to 3-letter codes. 047 to come back.
Proposal No. 2005-06:
Addition of Subfields for Relator Terms/Codes for
Subject Access to Images
Summary: This paper proposes defining subfield $e in fields
630 and 651 and subfield $4 in fields 630, 650, and 651 in order to use relator codes and terms to enhance the retrieval of visual
materials.
MARBI action taken: Need to find a new subfield for 611 and 711 $e. Need
to defined $e for 654 and 662.
Voted to approve
resolution: Equal access to nonroman resources - i.e.
utlization of Unicode. Unicode has 4 possible
implementations and the expectation is that the ALA membership will be asked to
vote on it to encourage use of Unicode by all system vendors.
Discussed FRBR. There is a
need for identifiers for work and expressions and how would we do authority
records. Sally McCallum prepared a document proffering two options. Should
6XX's be included and what is the impact of needing to keep them updated?
2:00-5:30 Hyatt – Grand Ballroom A
CC:DA – Liaison
ALA representative, Jennifer Bowen. Resources Description and Access (RDA) should take a more progressive
approach. We need to involve other communities/stake holders.
Timeline:
May-July 2005: Development of prospectus
Oct 2005-April 2006: Completion of draft of part I, and
constitutency review
May-Sept. 2006: Completion of draft of part II, and
constitutency review
Oct 2006-April 2007: Complete of draft of part III, and
constituency review
May-Sept 2007: Completion of general introduction, appendix,
and glossary
2008: publication
See: http://www.collectionscanada.ca/jsc/0410out.html
- Feedback on part I included comments on both past AACR2 and
future "AACR3". The new outline is not what CC:DA had discussed so
CC:DA commented at ALA and was generally favorable: purpose and scope at the
beginning, relation to the resources descriptions, and chapter 1 with
terminology and where to start when cataloging.
- PCC in its stratetic planning is thinking the catalog will
be discovered from various sources and the ILS client will be more for library
staff managing data and making sure a larger federated searching model will
easily harvest data we have in traditional OPAC records.
- RDA is intended to focus on content - not how our OPAC
displays data, i.e. not a display standard.
- FRBR has been worked into the new outline
- RDA should allow a cataloger to catalog something htat will
be compatible with ISBD, although arrangement is not exactly in ISBD order,
e.g. all rules about the "title" might be together, but the resulting
elements be show up in different places with ISBC.
- IFLA is reviewing both FRBR and ISBD so they are in flux
- Prefered source of information is an important issue -
chief source and/or prescribed source - LC's response was discussed
NISO report. 5 standards, 1 ISO, and 1 registration were all
acted upon since the last meeting.
- ISSN is being revised - they have reached a concensus
(TISSN): there will be medium version (e.g. online, microfilm, print) while
retaining ability to keep the original/super ISSN
- Functional Requirements of Authority Records (FRAR) is in
final draft and will be reviewed in the next few months.
SUNDAY--June 26
7:00-9:00 Hilton -- Grand Ballroom - registered
OCLC breakfast
OCLC will be using WIKIs for review, reader advisory service,
&information able to be added for books in WorldCat. WorldCat is growing
rapidly: 58 million bibs in March, 59 in April, 60 in May.
New platform able to use multiple sources
|
MARC Dublin Core IFLA
LDAP EAD |
SOAP XMP OAM-PMH
SR |
|
Full- |
Text |
|
New contributors |
Links |
|
Uni- |
code |
|
Graphics sound |
motion |
- Netlibrary has 91,000 titles. They also have digital
audiobooks.
- Will add AskJeeves to Google and Yahoo toolbars.
- Terminology services pilot - 18 libraries are testing. Uses
CONNEXION browser, Microsoft Office 2003 research pane, and thesauri.
- E-journals holdings pilot - setting e-journals holdings
automatically; partners include Ex Libris
I tried to find out information about PromptCat. Buser says
they will have a new form soon. She said my profile is still there but needs
revising. I will need to figure out fields in Aleph. She has names of Aleph
users. A librarian using Baker and Taylor says is works great. She does
everything but labels. BNA is very experienced and can include barcode and
order number in 949 or whatever field depending on Aleph.
9:30-12:30 Palmer -- PDR 16
SAC - General Meeting
See handouts.
Sears will soon be
available in MARC21. Will we want in ODIN? Will be able to save records in
UTF-8 (Unicode) and bring into Aleph.
LC
report. See handout. Web access to series - p.4; Automated web cataloging -
p.4; CIP - Cornell doing ECIP - p.5; three modes/levels for digital materials:
1) AACR2, 2) MODS, 3) Webguides - p.6 #D; PCC page redesigned; NACO liaisons -
p.8; change in headings for parks - p.9; Indians and tribes - corporate names
for nations using names of tribes found at Bureau of Indian Affairs website -
p. 9. There is a proprosal to allow adding dates to personal names - respond by
July 22. [ I did]
MLA's
75th anniversay convention will be in Memphis Feb. 2006. Focus will be
form/genre for music and music information retrieval.
AALL's
report focused on what is and isn't inherently legal and effect of incorrect
subfielding in OPACS.
Discussed new schedule format. Bruce will ask for: Sun. 10:30-12:00 and Mon.
1:30-3:30 and 4:30-5:30
_#_#_#_#_#_#
I stopped by the OCLC booth to ask for a demo of the
Terminology Project but only one person had heard of it and couldn't demo it.
1:30-4:30 McCormick Place - S405
ALCTS/LITA
Authority Control in the Online Environment IG -- "XML and
Authority Control" (see below)
See handout
Program Description ACIG http://www.library.yale.edu/cataloging/authorities/acig2005program.html
Information traditionally stored in MARC authority records in library catalogs
is increasingly being used in interesting ways in XML environments. This
program will look at standards that have emerged to facilitate the XML uses of
authority data, projects that have begun to implement these standards,
experimental research being conducted in the field, and general issues
surrounding the use of controlled vocabularies in metadata descriptions.
Speakers:
1. "MADS (Metadata Authority Description Schema), a MODS Companion"
Sally H. McCallum, Chief, Network Development and MARC Standards Office,
Library of Congress
- In XML
"formats" become schemas which specify the tags, attributes, etc.
Newer protocols like OAI harvesting and SRU/SRW prefer XML. METS packages
metadata (prefers XML) including rights data and all other stuff. Most XML
applications convert to another XML application. MARC21 -> MARCXML ->
MODS -> Dublin Core -> MADS; once something is in MARCXML it can go back
and forth. MODS is less detailed than MARC21 but highly compatible and has rich
linking capability. MADS pub ver. 1 Apr 2005: highly
coordinated with MODS - the schema specifies high level elements with unique
substructure with much of it really points at MODS or links to authority records.
Heading elements in authorities include: name, title, topic, temporal, genre,
geographic, hierarchical geographic, occupation. Attributes indicate related
types: earlier, later, parent organization, broader, narrower, equivalent,
other. Other attributes indicate variant types, e.g. expansion. There are other
elements: notes, affiliation, fieldOfActivity, URL,
identifier, extension, recordInfo, etc. http://mods.loc.gov
2. "XML for Authorities at NLM: the Groundwork for an Integrated Authority
File" Diane Boehr, Head of Cataloging,
National Library of Medicine
see handout
3. "XML Name Access Control Repository at the Hong Kong University of
Science & Technology Library" Louisa Kwok, Head of Cataloging, HKUST
Library
HKUST - XML for Chinese/English database. They feel name access control is
better than authority control. In Chinese, a person can have multiple names and
multiple characterizations/scripts of the name. They can retrieve names from
OCLC and modify the record to have the English form in the 100 and 700s with
other variants.
http://alcme.oclc.org/eprintsUK/services
4. "Organic
Authorities: XOBIS and the Metamorphosis of Library Data" Kevin S. Clarke,
Digital Projects Programmer, Firestone Library, Princeton University
see handout
5. "Web Service
Experiments with Authority Control" Thomas B. Hickey, Chief Scientist,
OCLC
Need persistent URLs. They tried using DSPACE - linked with "control"
button and it brings up name from authority record. It is free, simple, can
construct a link to a full authority. Example: http://errol.oclc.org/laf/n82-####.marcxml
It needs to be moved to a standard protocol SRW/SRU, would be better linked to
LC, ranking is needed for multiple hits and an ability to check frequency. The
basic service needs to be free.
Completeness - NACO
doesn't have all names and it's not easy to get them there from non-NACO
entities, eg. Museums. Experiment: could ProQuest these map to OCLC? If all went through NACO work
they could be compared and cleaned up.
VIAF - trying mapping;
need to use the bib record information to help identify different authors with
the same name. VIAF would connect separate authority files. Looking at OAI-PMH harvess but need stable URI or SRW/U for names, other text
in records, control numbers. RDF is designed to support this mapping.
ONESAC is using RDF/OWL -
OPAC network shared authority control. RDF imports MARC in RDFXML. http://errol.oclc.org/laf/n8254463.html
6. "Hong Kong Chinese
Authority (Name) Project: the HKCAN XML Version" Joanna Yi-hang Pong,
Cataloguing Librarian, Run Run Shaw Library, City
University of Hong Kong
see handout
4:30-6:00 Embassy
Suites - Salon
A-C
OCLC PromptCat Users Group
Robin Buser
- new developments at OCLC with potential impact for PromptCat
users followed by Arlene Klair (University of Maryland) - how they use PromptCat in conjunction with Bibliographic Record
Notification. - REGISTERED
There will be new forms in the next few monoths. The
Cataloging Partnership Program is for outsourcing difficult formats, can even
get NAXOS. PromptCat can be used for all formats except serials. The library
orders and is billed through OCLC (i.e. MTX). Most libraries have a 97.4 match
rate and have their materials shelf-ready. Bib note (bibliographic
notification) send a file and report daily for minimum to full level bib, TOC,
and new/revised URL - this would even work for the GPO records being updated
per our profile.
See Maryland (USMAI) handout (Ex Libris library)
They do both Promptcat and Bib Note. They check only 245-260
and send the book on its way. One person is able to do all the PromptCat
titles. They tried having acquisitions staff do the work but found it wasn't
successful so now it is a trained copy cataloger who checks the records. Their
profile is just for DLC/DLC records. [The librarian doing Baker and Taylor did
more libraries but not UKM - fewer DLC records for kids books]. USMAI has been
successful in matching on the OCLC number only in Aleph. They have a purchasing
profile with BNA - accept everything sent.
They are not entering orders until the item arrives. If you are an
enhance library you'll get it back in
Bib Note unless you fix the profile to not get them. Bib Note starts for
records cataloged after you start the subscription. Do a separate loader in
Aleph because you want those to overlay.
MONDAY--June 27
8:30-12:30 Hyatt – Grand Ballroom A
CC:DA -- Liaison
Don Chatham talked about publishing RDA. A project manager
will be hired for the publishing of RDA with the ability to promote communication
among interested parties. Need to address how RDA will related to the
Cataloger's Desktop. Map cataloging is moving to the Desktop. Jennifer Bowen
said the goal is to create a web product first. JSC is looking for input on how
we want it to work. We'd like to be able to find e.g. all rules needed to
catalog a globe - will need complex XML attributes. The information is really
very hierarchical in print, but may not need to be so deeply with XML where it
can be linked in multiple ways. A number of design suggestions were given.
J. Bowen continued her report. At Big Heads she described the
process as one of trying to be helpful and improve communications as ALA
representative rather than just CC:DA's rep to JSC. People seem to be really
caring about RDA. CC:DA site will have presentations. The RDA product will need
Alpha and Beta testing. A document on how to test will be devised. For LITA
this could take the form of usability, viewability, navigation - if you ask for
a task force to review, they must comments or not be on the task force and must
reply by deadlines.
Discussed LC's response. Again, the issue of rules together
came up. If we have rules together about the title, it should give all -
descripton, source, notes, etc. The principle would be put the information
where the elements are. Another: stuff about the source for title be spsecific
and evertying else can be from anywhere, e.g. determining the source of the
title can be difficult for multiparts and then you need a priority list to follow.
Following a priority may be different for some communities that have had
different priorities. Most attendees liked LC's response on this issue. You
almost always need a source of title note.
Another issue of concern is whether an item is published or
not. In the proposed revision A1.4C it says "optionally identify the place
of publication … " - why not
instead address this in levels of description, e.g. minimal, standard, all.
This area needs to be flexible for a broader group of works, e.g. place of
creation for artistics works, because theoretically since they aren't
published, they can use the rule for place. In the proposed revision, A1.4E.
Date is sounds like one can only do publication date (i.e. only for published
works) but the intention was all dates as we have done in the past. Discussion
is on Confluence.
Discussion of 5JSC/chair/5. The general feeling was the rules
in 21 need to stay. There is an expectation that other communities will have to
write up instruction manuals and libraries will need to buy them, despite the
fact the original goal of JSC was to create rules that could be applied by
anyone for anything without additional manuals. JSC intends to refer out to
these other manuals.
What about LCRI's? LC will no longer produce a document tied
to AACR/RDA. It will continue to provide guidance. LC is continuing to review
chap. 21. "Primary access point" idea will replace "main
entry". Current OPAC performance should not drive our decisions.
5JSC/chair/6. A task force has been formed to address the
GMD/SMD issues.
CC:DA's response to Part 1 draft will need to say
"blah" is okay and "blah-blah" is definitely not. JSC will
compare responses.
A task force has been set up to review Rules for Technical
Description of Digital Media, with a focus on data elements.
There will be a pre-conference on Cataloging Cultural Objects
(CCO) at 2006 Annual.
The task force on "Differences between, changes
within" document for serials is working on updating it.
MARBI report. See MARBI
DCRMB is asking CC:DA for comment. A task force has been
formed.
12:00-1:30 Hilton – International ballroom - need invitation -
registered
OCLC Luncheon
Update similar to breakfast with a
few more details.
2:00-4:00 Palmer - PDR 16
SAC -- General meeting
MARBI report. See MARBI
Bonnie needs feedback about the
issue of including subjects in expression authority records by midwinter. Can
we all add to? Is it required or suggested? Is this in place of putting on
individual bibs? Need to coordinate with a similar IFLA task force chaired by
Marcia Zeng.
There will be a FAST program in
2006 Sat. 1:30-3:30
FAST geographics - OCLC is going to
look at BGN names.
David Miller reported that he was
able to successfully use the Reference Structure report with the III Users
Group to affect some changes.
ARLIS liaison to SAC will probably
be approved.
NLM is changing how records are
being distributed. They had separated subject trees, and then reassembled them
for bibs. Beginning Dec. 2005 they will not be reassembled and will look like
they do in Locator Plus.