ALA
Midwinter, Jan. 2001
Washington, DC
Notes by Shelby Harken
FRIDAY -
Jan.12
12:00-5:00 Capitol Hilton -- Federal A
PALS Users Group –
buffet lunch
Attendees: Becky
Bell, MnSCU/PALS; Gary Johnson, SDLN; Tony Stukel, ODIN; Shelby Harken, University
of North Dakota; David Barton, MnSCU/PALS; John Christianson, TDS; John
Stromquist, Waldo; Teresa Edwards, St. John’s University.
The agenda was
approved with the change of moving the topic of future meetings to near the top
of the agenda. Shelby Harken gave theTreasurer’s report. The account balance is
$873.53. Shelby noted that the bank is taking $11 out of the account every
month. The account must have been set to maintain a balance over $1000. Shelby
proposed the remaining funds be paid to MnSCU/PALS to be used to cover costs
that benefit the WWPALS User’s Group, for example, paying for the lunch at this
meeting. It was so moved, seconded, and passed. Shelby will then close the account. The Minutes were submitted by Shelby Harken and posted on the
ODIN Web page. Minutes were approved as posted.
Future meetings.
We decided to try getting a restaurant with a meeting room. The meeting will be
June 15, 2001 at 3:30 or 4:00 PM and end with dinner. Gary Johnson moved that
the current officers remain in office another year, Tony seconded. Passed.
Becky Bell
reported on the status of PALS sites not present at the meeting.
Atlanta
University Center is still on PALS. They are loading Authorities. They are not
in a rush to leave PALS. They are being courted by Endeavor but they are not
interested in Endeavor software.
Ferris State
signed with III. Chris is working on exporting data. III is having trouble
importing data, particularly holdings which are not being sorted properly.
Chicago Public
Schools is still using PALS.
SUNY/Fredonia
goes live today on Ex Libris. In Oct. they started working on migrating. They
are the first library in the SUNY group to come up live on Ex Libris. All the
data converted. All their data went somewhere including serials checkin (they
didn’t use ACQ). They are impressed with Ex Libris’ expertise and ability to
keep on schedule. No special support had to be given to Fredonia by MnSCU/PALS.
Fredonia set up report writer to generate reports that were converted to UTF8
and no data was lost. If there was not a MARC field for any particular data, it
was stored in MARC 9XX fields or in Ex Libris’ ADM record.
WisPALS had
selected DRA, then deselected it. They decided to go to their second choice
rather than do a new RFP, which was Endeavor. They have had problems with
conversion programs because Endeavor’s conversion program is based on several
versions back of PALS. Locally created
records (MARC Editor 9-xxxxxx) didn’t work so they had to write script to
change the files. As it turns out, that also meant that no record with a prefix
works, so all MARCIVE temp records are also a problem. CalWest was the last
library to tweak the scripts. It is up to PALS libraries to write and/or share
scripts. Endeavor makes no attempt to address the current version, but still
uses the first scripts Tom Briggs at Shippensburg wrote a number of years ago.
South Africa. The
group that provides support is separating from Unisys S.A. They are still
adding libraries to the PALS system and are managing the software themselves.
They are looking forward to PALS ILL using ISO standards.
South Dakota
Library Network (SDLN). Gary Johnson reported they have added lots of K-12
libraries. The biggest problem is staff support in the small libraries.
Winnebago records have been sent to OCLC for conversion. Follette records are a
problem because call numbers are buried in 9XX fields. Small libraries of up to
30,000 vols. are billed $2000/year plus a usage (transaction) fee. Small
libraries historically have not had networks, so SDLN is their network
provider. Military and hospital libraries just added separate lines to the
library to keep firewalls intact. They locally load Books in Print and locally
generated databases. Their RFP committee met recently and will be sending out
an RFP in March/April. Responses will be due before next ALA. A recommendation
will be made to the Advisory Council and a decision made at their Oct. 3, 2001
meeting. The legislature wants to know what they have chosen before they decide
on funding. Money would be available
July 1, 2002 so they would have 1½ years to migrate everyone before 2004. SDLN
is working with ODIN on patron authentication for InfoTrack and ProQuest that
would give the user a single web address to click on based on the IP, it would
then authenticate or require the entry of a barcode which then would default to
the OPAC that goes with the barcode.
Online Dakota
Information Network (ODIN). Tony Stukel reported that name authorities will be
loaded in the next couple of months once the new hardware is stable. ODIN is
also looking at loading MeSH headings. They are FTP’ing MARCIVE and OCLC
records. Six new libraries were added in 2000. Leach Public in Wahpeton, ND
School for the Deaf, Trinity High School, Training Center, Stutsman County, and
the State Historical Society which will be getting documents from a missile
site. Fargo Shanley will join soon. Grand Forks Public has installed a number
of wireless terminals. He was able to get Unisys to give him a 3-year contract
with an option to add 2 more years. Because of the biennial legislature, ODIN
may have to wait until 2005 to get funding. ODIN is implementing a new cost
model over a 5-year period that combines size and transactions. Funding is 2/3
from users and 1/3 from the state. Distance education initiatives may mean use
of DOCUTEC for reserves for NDUS, which may be a use for the NT server he got
with the new Unisys ClearPath.
Traverse-des-Sioux
(TDS). John Christianson. Two museums and several schools have joined TDS as
bibs-only libraries. Next Sept. their Unisys contract is up. He got a
reasonable quote but it is still high. TDS is still working through barcoding
problems resulting from the split among several member libraries. Since the state pays for telecommunications
for libraries that are part of systems, they now have 14 libraries on T1 lines.
All libraries will have 3-5 day delivery for ILL.
CUNY, SUNY,
WALDO. Teresa Edwards and John Stromquist. Both CUNY and SUNY selected Ex
Libris. Teresa said they are reconstituting the RFP committee and hope to have
a decision and move 2003/2004. WALDO has been using MnSCU for three years. They
are negotiating how to go foreward to 2004. One library may go to Ex Libris
early. He would like to see WALDO expanded and get a staffed support office.
SUNY’s contract was written to allow additional libraries to come in on the Ex
Libris contract but there is concern about SUNY’s ability to operate as a
system provider for a very large group of libraries. John asked about
transaction statistics. The web OPAC results in less transaction counts.
Removing reference files also reduced transactions. They are interested in
TitleSource II as a replacement for BIP but wonder if bibs will download; Becky
would need to see some records. John also asked about NetLibrary – how does one
regard the “one user at a time” situation and are others loading bibs? In
Minnesota they have a file of records that can be FTP’d. Gary wrote a little
script that allows a library to put them anywhere (shelflist) a library wants
them.
MnLINK. Dave
Barton discussed the MnLINK RFP process. They are still struggling with a date
for ending development of major new features, but will probably quit when they
start to implement Ex Libris. They will continue to do patches and follow
national standards changes. They will be moving to new quarters in March and
will then be with Mankato MnSCU computing services. About 1/5 of calls coming
into the support desk are from WWPALS libraries.
Ver. 16
highlights. The ZPALS client has the advantage that records for ACQ and ILL can
be downloaded. The Z client has to have config files set up for other Z client
users. It is one site at a time per session. It will have a cost. It will
bridge to holdings and circ status, like the old loaded reference databases. It
can slow down with very large response sets. ACQ funds and departments will
allow control on the whole fund or dept. rather than just subcode. In
cataloging, 008 will display in addition to the mnemonic display. To change the
CLD for the email address for undeliverable email, you need to enter 10 *’s to
remove one, then enter a new one. You can flag patron records as distance
patrons yes/no. Yes means ILL will operate like an external user but PALS will
know it is your patron by the barcode and patron flag. Each patron needs to be
CPR’d. In the ILL message file, if you add “T” you will get a list of titles.
To get electronic reports, you must select ELECTRONIC as a type in the LRP –
doesn’t work for bibs. The report will go to the email in the CLD. The biggest
change is overnight indexing of MARC editor files or imported records. During
the weekly load, the change will become “permanent”. MED-002 INDX option goes
to the correction file format, to OC 40, to OC43 skipping OC 41-2 which keeps
it from reprocessing every day until the end of the week. OC55 is enhanced to
use the config file.
Version 16 R1.01
has a new installation and creates a new logo. It works with any type of
machine and works with Microsoft 2000. The CLD Accounting interface is for
MnSCU libraries. MMR can be used to create or replace records. DIR … B will
display both item and associated bib record. ILL can print a barcode on a
laser, dot matrix or inkjet. Version 16
for PALSTAC will come out about the same time as mainframe, about 3 weeks
later.
ISO ILL pricing
and availability. They have been too busy to get to it, but hope to soon.
When MnSCU is
done with their RFP negotiations, we can request RFP responses from them. They
used two types of evaluation, 1) did the system answer your questions (RFP
details); 2) unclear issues were addressed in demos.
Respectfully
submitted _____________________________ Shelby E. Harken, Secretary/Treasurer
6:00-6:30 Renaissance Hotel –
Room 4
ME SAC CCS Executive
Committee – Report on to ALCTS
CCS on Program by SAC Subcommittee on Metadata and Subject Analysis with Lori
Robare, SAC Chair.
I was asked to
attend the meeting to present our program plans for Annual. The majority of the
discussion focused on ALCTS’s strategic planning. Examples were that SAC might
want to support continuing education on things like genre headings. Or, by 2002
ALCTS should have a continuing education course on integrating resources,
address the new chapter 3 for cartographic or chapter 9 for computer resources.
These could be institutes, pre-conferences, or PCC-type training programs. SAC
is proposing a subcommittee on best practices for subject reference structures
for vendors and by 2002 is to produce a position paper. They are trying to
follow PCC’s strategic plan outlined on LC’s web page.
6:00-8:00 Convention Center –
Rm. 37
ME SAC – Report on Program at SAC Subcommittee on
Metadata and Subject Analysis
I never made the
meeting because the CCS meeting lasted until 8:00 PM and we were last on the
agenda. (See SAC Subcommittee on
Metadata and Subject Analysis minutes)
SATURDAY - Jan. 13
9:30-12:30 Four Points Sheraton – Franklin C/D
MARBI –
Non-MARC country codes in 043; also in 041; taxonomic hierarchies in 754;
business meeting; LC report
See printouts of
proposals and discussion papers or http://www.loc.gov/marc/marbi/2001_age.html Taxonomic hierarchies seemed useful in a
broader sense and needed minor text revision. Narrators in 508 and 511 is
confusing; they will be in 511 from now on. Types of dates for electronic
formats is to be rewritten. Non-MARC country codes passed so you would use ISO
3166 codes in #c after using MARC geographic codes in #a and MARC sub-entity in
$b (which is not much used). The language code discussion will come back. There is a 2-character and 3-character ISO
language code ISO 639. The idea is you could add a language code to a country
code and end up with a pattern of 2-3 which means the number of characters in a
field wouldn’t match so you’d have to have repeatable fields.
Sally McCallum
reported that a number update MARC format documents have been issued. LCCN
4-digit was adopted Jan. 1, 2001. LC is working on diacritics but is ILS vendor
is having trouble. They have a SGML &DTD ó XML version. They also have XML DTD ó MARC on their web page. There will be a
new XML listserv started after ALA. MARC organization codes are available
online or through SiteSearch. They may never issue them in print again. ONIX ó mapping is also on their home page along
with a record maker they got from ONIX. The web classification pilot is
available for a 3-month trial. There is information on the CDS page. There are
plans for a MARC21 holdings format pre-conference where the emphasis will be on
systems working towards full compliance with MARC21 (I should attend!) June 15,
2001 8:30-5:00. They are also trying to promote the sharing of pattern records
regardless of vendor.
1:00-5:30 Four Points Sheraton – Rooms 1 & 2
CC:DA – Looks like good stuff starts at 2:30
The revised
ISBD(M) will be out soon on the IFLA home page. Barbara Tillett gave the Library of Congress report (full report
on their web page)_______ There are 7000 new authority records for music;
digital table of contents and reviews are being scanned and stored on their
servers; 20% of new CIP titles have 505’s; they are looking at ONIX and
collaborating with publishers; LCRI’s have been revised and will be in #3 and
#4 updates plus the LC Desktop; new character set characters will begin
appearing in records; they began using CORC in July 2000; cataloging is much
slower with their new ILS – trying to find ways to speed it up; Gary Strawn’s
MARC validator is helping reduce the number of errors; they are still testing
geospatial searching; their goal is to be on Voyager 2000 by May 2001 and have
authority searching up; catalogers are reviewing records revised to Pinyan;
they have made 5 million items in 90 collections available online in the
National Digital Library; they have begun to acquire digital journals for
archiving.
Glenn Patton gave
the IFLA Section on Cataloging report. ______ A concern is the the ability to
exchange bibs created with different cataloging standards and ISBD’s. They hope
to get ISBD(A) and (M) on IFLA’s homepage for free. A committee is looking at
ways to link between authority records that describe the same person or entity
from disparate library systems’ catalogs.
Brian
Shottlaender gave the report from the Joint Steering Committee. The next revision package will contain work
in uniform music titles; AACR2 and ISBD(ER) harmonization except areas 3 and 5
which still need to be addressed; chpt. 9 is being review in final form by
national libraries; 21.1B2 – conference collective role as authorship –
European libraries don’t use it in 1XX fields; retention of nobility terms of
honor; see refs for names of persons; terms of address for married women;
publishers statements; initial articles appendix. Some items didn’t make it through JSC, rule of 3 for authorship,
entry of corporate body, map cataloging area 3, non-roman access, didn’t like
major/minor changes for serials, just major listed; abbreviations
appendix. Political issues include
preparation of an explanation of why we requested an ISBD proto-type of the
rules for user comment; decided there should be a AACR and JSC web site, but
AACR folks are afraid of lost sales; 0.24 multiple versions is now multiple
format variations; Barbara Tillett created a principles document that will be
reviewed in April. AACR2 & ISBD problems – people making proposals will
need to explain how it will cause disharmony with ISBD and the same for AACR2
so as to be prepared for change, challenge, etc. LC50 & 51are new documents
from LC coming before JSC. 50 – series authority records; need to be revisable
for those which reflect bibliographic records that undergo change; 51 –
multipart items in AAR2.
There was much
discussion of major/minor changes in appendix E for title changes. JSC wants
only major listed and wants them re-organized in categories: monograph,
serials, integrating, multipart. They took out the section on reproductions.
Changes to 24.2 corporate names should be in. They paid attention to the ISBD
and ISSN folks. Numbering that begins again isn’t a major change. There was
difficulty with defining significant and substantive; real and actual. They found overlap in 21.2 and 21.3 with the
appendix – where should the information be? Series authority records should be
updated when the bib record changes. Multiple distributors of a single work
(journal) will be separate records, i.e. JSTOR and Project Muse would be 2
separate records. NetLibrary uses the original print record and adds the
publisher in the 533 – is that how we want to do it? Jean Hirons felt strongly
that “change” needs to be addressed in the rules, not in an appendix, or the
appendix needs to refer back to the rules so all the changes are in one place.
There were a lot of questions about the proposal for integrating multipart items
and when it’s a different manifestation, when it is a change “between” or
“within” a manifestation. Is a change in language a different manifestation? –
yes/no because some web sites can display in multiple languages.
5:00-6:00 Convention Center -- Rooms
1 & 2
OCLC ILL Users Group Meeting—Attend for Paulette
OCLC presented
information on recent changes.
Consolidated pricing combines searching, display and IFM into one price,
it incorporates a lending credit and provides a savings compared to separate
pricing. You get 4 free searches to each produce; any more searches and you are
charged a surcharge. Global GAC: allows participating member libraries to
easily identify other libraries willing to lend and send materials
internationally, but you must use IFM, must send air-mail or faster or use
ariel, can sign up on web for resource sharing, there is no charge, you have to
lend to all participants in the GAC. There will be a new web interface as an
alternative to Passport. It is in milestone 4 testing, with 3 more steps left.
Right now you can logon, search, do the message file and apply constant data.
They will move away from passport to the web interface.
OCLC Illiad is
the management system they are licensing to provide all ILL lending and
borrowing functionality within your ILL dept. to OCLC ILL services. It is
client-server based. Union list strategic report service will be available. It
won’t cost more than you’recurrently paying. You get a report of previous 18
months of titles ordered. Reports start by the 15th of the month
following the one in which you registered on the web. It gives a count of the
number of requests, OCLC number, and title. Beginning Feb. 1 the credit you get
for adding Union List data on your OCLC bill will equal a search. Non-referral
days in 2002 will be weekends, Christmas eve and day, New Year’s eve and day.
The new ILLMe 2.1 is available. With it you can reprint shipped versions of the
record after you update and you can reprint mailing labels until the next
update.
WEDNSDAY - Jan. 14
7:00-9:00 Renaissance Washington --
Grand Ballroom South
OCLC breakfast (collections and
technical services)
See handouts.
CORC editing might be easier in the “MARC text area” or “Dublin Core text
area”, but you still have to return to full display to validate. FirstSearch
has local customization options for use of our own logo and our version of the
ILL request form. The NACO macro is an issue everyone brings up when OCLC talks
of its changes. They say they will have something in CatME
9:30-11:00 Hotel Washington
-- Ballroom
LITA Electronic Publishing/Electronic Journals Interest Group
-- SFX - The Inside Issues for
Libraries
At next summer’s
meeting they will discuss digital rights management with Cliff Lynch, James
Neal, Dennis McNey, and Carol Rischer. There will also be a pre-conference on
E-Books and Portal Libraries.
SFX was developed
by Herbert Von Stempel to “create a network of knowledge.” Miriam Blake from Los Alamos described how
they are using SFX with science databases. They were a beta site although any
library can purchase it and use it with any system. The problem is there are
lots of things libraries try to offer to users that are all over the place,
some intra-net, some internet, and some locally. How do we get it all into one
search? The easiest is a hyperlink, e.g. 856 to publisher’s site, but that
still might not be the licensed site. Sometimes you find more citations in a
document. The key is to get the user to the appropriate copy and for reference
linking to work to sites they are authenticated for. It uses Open URLs – open
protocols – a NISO standard. It is a big thing being used by many publishers;
localization is based on the user’s incoming URL, and points back to the
institutional service component. It is recognized by CrossRef and DOI handle
system. How it works: information is pushed into the base URL – a service
component – via a cookie, via a dLF certificate, etc.; a unique open URL is
constructed for each object containing ISC (my library) location and identifies
metadata from the source. Example: http://server.lanl.gov/menu?id=doi:issn____-____
SFX is the service component software which provide context sensitive localized
service. You (library) decide the services, e.g. appropriate copy full text. It
is dynamically available based on locally defined parameters; once an
administrator adds a title, it is immediately available to users. At Los
Alamos, articles are added every few minutes and they are available
immediately. “Sources” are references/citations/abstracts you start with and
want to link from. “Services” is the list of things you can do, can retrieve,
etc. “Targets” are where you end up, e.g. full-text repository, journal, etc.
SFX must have
metadata to present services. Metadata from external sources would be metadata
coming from CrossRef, the publisher, etc. Internal sources provide more local
control over metadata retrieval. Example: SFX cookie is set to present user
with a button, metadata is retrieved from source database to dynamicallyl
present a list of possible services. Los Alamos renamed the button LinkSeeker
for their library. They got a team together to decide what services to provide,
i.e. what targets they wanted their users to go to; what is the syntax to go to
that target (is it Z39.50, HTTP, ISSN); what content will the user receive
(table of contents, abstract, full-text, etc.); alert if password required.
Parsers are created by ExLibris and included in the product although you may
need/be able to create only locally. The vendor on the other end needs to be
“Open URL aware”. Once they do that,
they don’t need to address linking to your holdings because you make yourself
Open URL aware and set it up to go to your OPAC to get holdings.
A MySQL database
controls SFX parameters and thresholds. If the target does not match the
library’s subscription, it (SFX button) doesn’t show up to the user. SFX runs
on a 500 mhz intel/LINUX/<1 GB. Ex
Libris hosts for you in most instances. Staff resources at LANL, ¾-1 FTE to
initially implement, ¼ FTE for ongoing. The new release requires even less
time. The could be considered electronic journal listing work which you would
quit. Technical knowledge needed is UNIX shell, MySQL, PERL scripts for
reports. Librarians and non-technical people do most of the work. They started
with an Excel database to batch load so they keep a list in Excel of
new/ongoing/dead titles and batch load to keep current. Once loaded you can get
cool reports. User issues: begin conservatively so you don’t overwhelm user;
ask users what they want; once in a while you get a 404 but it is not really a
problem; you have to train users how to use the SFX button and services screen,
but then you use that for all your databases. The majority of titles are in Ex
Libris’s database, but if you have different start dates, you can put that in
the Excel spreadsheet. Swets/Blackwell has agreed to provide a file with start
date information for subscribers. CalTech said their biggest problem has been
vendors who fail to keep up with ISSN’s for title changes. If you create your
on databases, it will work with SFX as long as you create metadata; the better
the metadata, the better it works. Los Alamos set up a link to the OPAC for
their print titles to show library holdings. You can include both print and
electronic ISSN’s for a title.
http://lib.www.lanl.gov/lww/add.htm
11:00-1:00 Convention Center – Exhibits (other times
I could fit in)
Library of
Congress. American Memory site has their RFP used for the digitization project
and copyright policies.
OCLC Illiad.
Watched a demo of the borrowing half. OCLC is licensing the software as it’s
ILL product.
CORC. Stopped to
see if there was something new. They were demo’ing pathfinders.
Ex Libris. Acq
was the one area at the Minnesota demos I thought were a little weak. As it
turns out their three most recent clients have requested some of what I was
unsure of in their RFP’s.
NetLibrary. There
are 30,000 titles with a group focused for academic libraries. I tried to get
into a “private” presentation of NetLibrary for Academic libraries which I had
registered for, but they said they were full.
Web Feet was
mentioned at the CORC meeting. Intriguing. They claim they have cataloged loads
of web sites and you can buy MARC records from them. Seemed school focused, but
they did have “academic” .
I tried to find a
vendor with Rocket Readers, etc. but none seemed to have them to try out.
1:00-3:00 Grand Hyatt Hotel – Independence F &
G
Extending the OCLC Cooperative – OCLC Metadata Services – Learning about OCLC’s sweeping new strategy to
extend the OCLC cooperative, particularly the Metadata progrm, which will
transform OCLC’s traditional cataloging service to a comprehensive metadata
creation and management service. The session was to describe new and future
initiatives and tools that will help create a distributed metadata repository –
Extended WorldCat – through the cooperative efforts of libraries, archives,
museums, publishers and other organizations.
Libraries are
changing. Libraries have been custodians, handled mostly one medium, owned
their collections, got materials to users in reasonable time, everything
in-house, local reach. Now we deal very much in multiple media, we have
libraries without walls, service is just-in-time, we do more outsourcing, we
have a global reach. The library needs to come to its users. Ask Jeeves gets 20,000
questions a day. A group of librarians tried 12 questions, and Ask Jeeves
failed them all. Sell ourselves: Believe in Libraries – we have expertize,
resources of value, and that have stood the test of time.
OCLC strategy for
the future is to be a leader for global libaries, help libarires serve, and
provide economical access to information. WorldCat is the heart of the strategy
with 2 million records added per year. Their goal to to keep extending it and
be a globally networked/linked resource including images, video, etc. They
intend to support metadata, and discovery and navigating (find it) and
archiving and fulfillment (get it).
Metadata. The
source of metadata will go beyond libaries, to publishers, etc. for 1) online
creation, 2) expanded coverage, 3) early capture, 4) just-in-time, 5) local
management, 6) checking, updating. Example: click on an icon that goes to
Microsoft Publisher to create metadata and bring in the Dewey class and add
Dewey links. Dewey has been licensed to organizations to organize wesb sites,
eg. Suite101. In CORC the PICA grey literature file can be searched for foreign
materials, which can be reformated to English and cataloged.
Archiving and
content management. OCLC intends to archive both data and the software used to
create it for conversion. Local management example: working with GPO, at the
time they catalog a document, if they select the “archive” button, OCLC will
pull it and put it in a digital vault and add a second URL. It is being
designed for anyone to use that way.
Discovery and
navigation. The idea is to make libraries ubiquitous and try to meet users’
needs where they are. In Google, the international library symbol shows up if
you check “use library”. There is a way to profile it for yourself. You can get
to OCLC e-catalog where you can borrow or buy, sort by delivery time or by
price. Plan to work on Ask a Librarian and is working with the Library of
Congress’s digital reference efforts.
Fulfillment.
Borders has: connect to your library, helps you find your library, and may let
you see parts of chapters.
The demand for
quality metadata has exploded. We need to link metadata to resoruces, to
networks, to content, to content providers. We need to transform cataloging
services, by mvoing to web-based, from proprietary to open, but still support
the functionality we have had. We should not go backwards. The web environment has the advantage of
providing linked authorities and classification; you can use multiple metadata
formats, MARC, DC, etc. You can do URL checking. They intend to extend the
value of WorldCat by adding table of contents, author information, digital
images, reviews, etc. Multiple script support increases the global reach. One
can have flexible, customizable view of WorldCat.
Migration from
Passport and Prism to the web. Plans will be provided, but the goal is mid
2003. CatMe will be enhanced somewhat to carry certain activities while
Passport is phased out. CD-ROM products are already gone. Passport will be the
next to go. Eventually Unicode will
handle multiple foreign scripts and phase out CJK and Arabic products. CORC and
CatExpress will handle web cataloging providing an integrated metadata and
management service. Z39.50 will continue separately. The result will be a
simpler pricing model and software that will work with MAC’s, Microsoft, Linux,
palm pilots, wireless networks, etc. It will by up 2x24x362 and be
standards-based for interoperability.
CatME is an option to replace Passport to a point, but it doesn’t have
NACO functionality. They will give us 12 months warning before removing
supported software. ILL will also go to the web. 37,000 libraries are in OCLC,
2.5 % of the world’s 1.5 million libraries. OCLC intends WorldCat to become a
linked set of repositories, one of which will be OCLC bib records, with
multiple viewing mechanisms.
Open Name Service
– Open URL: network of providers using standard identifiers that will work
behind the scenes to access authorized resources. Try out http://names.oclc.org They plan to use open URLs which include
more metadata to narrow you search. You can go to a vendor, bib, etc., e.g.
AddALL which is an aggregator of titles in 34 dot.coms. They are talking with
vendors about using OCLC billing to buy books when you can’t use a credit card.
3:00-5:00 Grand Hyatt -- Roosevelt
Room
MetaLib – Ex Libris
MetaLib is a digitool for digital creation and management. It
is part of Aleph which is software used by 540 libraries in 42 countries in 20
languages. It is used to: 1) look for
information, 2) defines how to navigate between/among/within databases, 3)
obtain actual materials, 4) tool to harness information . It is a information gateway to all the resources in
your library. MetaLib is the front door; SFX is the back door. The wish list
they started with when they developed MetaLib: 1) unified interface, 2)
simultaneously search databases, 3) result list separate OR together, 4)
continuously search among databases, 5) controlled by librarians, 6) handle authentication
and authorization, 7) provide for customization at the institution level, 8)
allow personalization by the users.
The gateway allows you to search using different protocols,
Z39.50, http, Aleph. It takes care of protocols, syntax of queries, character
conversions stored in Unicode UTF8. I has four components: find, present,
combine, find duplicates. The Resource
Store uses the cataloging fucntion with MARC or Dublic Core so you can record
information about your resources. You can configure information per 1 metadata
target. There is an “i” button to store your results. The connect button has
filtering options. You can use your own authentication mechanisms or Ex
Libris’s. You can create a personal page with an “E-Shelf with your choice
lists, Alerts can regularly give you updates of what is new, retain search
history, locate path – where you have rights to borrow and access.
SFX does the linking. MetaLib extends the context sensitive
analysis of a documents, e.g. author citation to further links, or services
like email , etc. SFXtargets enables “deep” linking, not just to basic home
page. Once you get to a resource you can start wandering away from MetaLib or
go back to it. SFX takes you to the authenticate version of a resource. From a WebSPIRS result, you can go on to:
citation database, full-text, Ulrich’s, OPAC holdings, subject gateways, and
resources like PubMed.
MetaLIb is a resource discovery tool that is Open URL aware.
We have control over the data access through MetaLib administration. If the
library switches database providers, it is changed in Admin software, once. If
a user logs in, he gets to resources he is allowed to access or you want him to
get to. Patron information can be populated from the library’s patron files.
The search history (held over time for the patron) holds 40 of the hits it
found; for more, you tell it to redo the search which it stored. A user loging
in can go to the previous profile he used, or start again and add more
resources. You can set up a shared identity so an instructor could assign
something for a class and they would all get the same thing. Remote users would
use a personal login to a proxy server on top of the SFX server. User
information is password and ID protected.
4:30-6:00 Mayflower Renaissance Hotel --
Senate Room
BASIC and NISO are sponsoring a joint standards
program at Mid-Winter ALA.
A session
discussing international identifiers, ONIX and the AAP standards for ebooks.
Late to this
meeting – long walk. The proposed ISTC
was being discussed. It would identify a work with all its manifestations under
it. It would be a numerical identifier with uniform title and with metadata
embedded. ISBN is used by the book industry. ISCT could be assigned to older
works. Any time someone registers a new version of an old work, the old work
would also get a number. It could be applied to any text work/object, but
probably would be assigned only when needed. The first working draft is
available; then will put RFP by end of Feb; hope to have the agency set up by
the end of theyear and implement in 2002. The INDECS vocabulary should be used
with metadata for ISTC. ISTC is getting interest from authors and many other
groups.
Regina Reynolds
on ISSN, SICI, ISBN. ISSN is being used in SFX. Since table structure allows
for either ISSN (print or electronic), resolution linking results in few
failures. They are hoping to have fewer reasons for new titles and hence fewer
ISSN’s. There is agreement on continuing resources and they will be given ISSN
numbers. They will start with those that provide access as thought is is a
journal. ISSN as URN – there is a plug-in at the ISSN International Center
page. The plug0m takes you to a split
screen, with the top being the journal and the bottom is ISSN metadata. ISSN is
developing a portal service, with links to publishing agencies, union catalogs,
subscription lists, e.g. Project Dieper. Issues related to lack of use of SICI:
lack of publisher support, character support problems, many think it will be
replaced by DOI, except a DOI with a SICI in it is very long. ISBN will be 13
digits by Jan. 2005. That will increase the availability of the number of
ISBN’s and will conform to EAN?UCC format. U.S. ISBN agency is advocating
separate ISBN’s for each e-book rendering, e.g. 1 for palm pilot, 1 for Rocket
Reader, etc. Core metadata needs to be developed to be associated with each of
these formats.
Sandy Paul. ONIX
and the Aap metadata standards for e-books provides traditional bibliographic
data PLUS author bio’s, cover image, all marked up with XML tags in an
internationally accepted standard. The EDItEUR people started e-commerce
standards for the world and developed EPICS. The American developed a subset
and by Sept. of 2000 it became an international standard, ONIX. See http://www.EDItEUR.org/ONIX.html A push-pull for print needs to be set up
between you (library) and a publisher. Then they moved to e-book metadata
standards they added to ONIX. See www.publishers.com
Also US organiztions and their implementation is at http://www.bisg.org
Richard Stark,
Barnes and Noble with a publisher’s viewpoint. You can’t sell books online with
just the author and title of a book.
You need price and format, you need descriptive texts, critical reveiws,
tables of contents, etc. He would like TOC tagable. ONIX only works if you have
clean, quality data. He doesn’t see it as a replacement for EDI or MARC21. Using
ONIX, they see lots less errors than in comma-delimited files.
Sally McCallum,
Library of Congress. ONIX IS a book seller’s standard. But there is info useful
to libraries: product numbers, form of item, title, author, extent, set/series,
conference, language, audience, prices, subjects employing a variety of subject
schemas, linked products (images, etc.) There are also other things of
interest: rights management, announcement dates, discount price info,
transaction info – publisher addresses, print run, etc. ONIX has 2 levels, one
that is basic and required, and one with lots more info that is optional. They
have personal and corporate authors, but conferences are not authors. They
don’t do “main” and “added” entries.
They use a lot o controlled lists. The impact on library processes:
selection might be enhanced and information electronically delivered; ordering
can be linked to EDI data be extractable into order forms; receipt – the
description is thorough for verificaiton; cataloging would have access to
fuller data; It facilitates earlier notice of publication; It supports EDI for
ordering; mapping is possible. http://www.loc.gov/MARC/ONIX2MARC.html Now we need to figure out how to get CIP into
the process. ONIX is being used to transmit back and forth between publishers,
vendors, etc. DOI is an acceptable identifier and some publishers are using it
for e-books. ONIX uses Unicode so it can be in English, Danish, Spanish,
Portuguese, etc.
6:30-8:30 Hotel George -- Bistro Bis (15 E. St. NW)
Ex Libris
reception
Met the sales rep
for ND and SD, Susan Buchanan (susanb@exlibris.usa.com)
Talked with prospective and new customers. Visted with Gary Johnson and Tom
Bremmer.
MONDAY - Jan. 15
7:00 Grand Hyatt Hotel –
Constitution A
OCLC CORC Users Group Breakfast meeting (A working meeting to discuss
administsrative /management issues regarding the CORC Users Group as an
organization, such as by-laws and mission. Additionally, the organization and
role of special interest groups will be address by officers Suzanne Pilsk,
Pres, and Jackie Shieh, Pres-elect. There will be information given by OCLC
staff concerning new developments in CORC.
description into
the public services world.
Jim Simms from OCLC. See handouts. Intend to transform traditional
cataloging into a comprehensive metadata and management service. The goal is
one entry point … single clik of the browser or Z39.50 and what you need will
be pushed at you. They are concerned about productivity rate. Goal is
completion by 2005. Intend to eliminate stand-alone products and eventually
form a metadata desktop. In 1 ½ to 2 years passport will be gone. Nee to move
ILL to the web and figure out NAD and UL. CORC will be the cataloging
environment; hope to improve the harvester. They are starting to figure out out
how to do NACO in CORC, at least add, replace or add a reference. Can’t do the
NACO macro in the web but are thinking of what to do. Indexing changes:
proximity operators, etc. need tweaking; search strategy needs to stay in sync
with FirstSearch. Next meeting will be in April – discuss issues from the LC
Bib Control conference and issues such as multiple versions of e-books, with
some online, some downloaded, some sitting on a network.
9:00-11:00 Renaissance
Washington – West A
OCLC CORC
and the Public Services Librarian -- Learn how OCLC
CORC brings resources description into the public services world.
A variety of
purpoted reference sites were discussed, some better than others. The reason
for using metadata is because it is structured data, supports administration of
copyright, enables retrieval, and describes resources to improve discovery and
navigation. Search engines have back away from metadata because they worry
about its authenticity. The metadata provides semantics (associated meaning
with a label), syntax (it’s like grammar, order and rules for arrangement), and
interoperability (Z39.50, http, xml, etc. standards supported) So why CORC? –
it uses Dublin Core which is simpler than MARC, it has semantic
interoperability, it is extensible (then it is not so simple), and has
international concensus. Dublin Core is a simple, rich standard. We need to
share with other communities (MARC is library-centric) e.g. internet, vendors,
creators, distributors, etc. We need to become part of the larger community. We
can create structured access to valued web resources. We are experts at
providing information packages, whether a bib record, a handout, or BI class. A
goal of OCLC: weaving libraries and librarians into the web.
10:00-12:00 Grand Hyatt – Roosevelt Room
SFX -- Ex Libris Presentation
SFX tries to
solve how to access multiple electronic formats under varying authorities (authentication)
and how to inter-link them all. The “front door” approach is to provide access,
navigation, among/between resources (even if I don’t know which ones I need);
and use profiles. The “back door” approach – I know what resources I want my
users to get to, so I use SFX to provide the navigation. Current approach
restraints: static linking – I have to put the right (authenticated) URL in my
bib record or my web page; it is not context sensitive – you can get to a site
but not the one you are authenticated for; multiple vendor solutions –
approaches by vendors vary, thus confusing users; librarians have no control
over secondary links.
With SFX the
linking is separated from the resource and stored and manipulated in the SFX
server that you configure for acess, authentication, etc. The linking is then
controlled by the librarian. Example: the user searches and gets an A&I
database response screen; on it is an SFX button; the SFX button brings up a
box with a list of services: Table of contents at the Institute of Physics (the
library doesn’t have access to full-text); holdings in local OPAC for print;
full-text via a local ILL service; list of citations (the library can define
search engines); and users original search for editing and re-issuing another
search. The process opens lots of windows, but they will again find SFX buttons
and only three windows actually remain open.
The trick is to train users to opt for the SFX button associated with
hypertext links. That will keep them within the library-defined parameters. The
ILL form downloads data.
The server 1)
accepts Open URL from the information resource, 2) analyzes content in the Open
URL, 3) evaluates appropriate services you are profiled for (can be based on
patron info from Aleph), 4) dynamically computers link to a target. You can
define groups: public users, reference librarians, tech services, a class, etc.
that would each have different services. The server can be Linux or Solaris, it
works in single or multiple-site environments, comes with software and tools
and some pre-existing files you can use or alter by adding/removing titles or
changing dates for which you have access.
At Los Alamos it took 5 months for .5-.75 FTE to set up. Maintenance is
about .25 FTE. It tracks statistics, e.g. how many times SFX was clicked in a
resource, how many times full-text was available, if it was what other services
did the user select. You will find out which titles get used the most and which
services users like.
SFX is compatible
with other linking services, CrossRef and DOI (persistently identifies digital
objects, usually journal articles). It uses an Open URL framework. It is a
communication of metadata about the resource. It is created dynamically at the
time of request and allow interoperability between information services. The
SFX server creates a base URL, adds elements of metadata about the object or a
pointer to it. SFX identifies users as having ability to access the resources.
SFX is currently going through NISO. Example: http://sfx.aaa.edu/menu?genr=article&ISSN=1234-5678&vol=12&issue=3&Spage=1&Epage=8&date=1988&aulast=smith&aufirst=paul
Sixty-three
publishers are now using CrossRef. It is a service/system that is built on DOI
to see links to other journals that might be published by a different
publisher. It is built on an 1)authoritative namespace (unique identifier), 2)
mechanisms to register and assign identifiers, 3) resolution of DOI links. So a
publihser 1) registers metadata of their objects (articles) and get an assigned
DOI, 2) publishers obtain and put it in their electronic article, 3) the DOI’s
are provided to users as links that are resolved by the DOI server (Handle
System)
The problem with
DOI: CrossRef could take you to a place you are not authorized to have access,
i.e. wrong URL; or , you may have print in the library and you want the user to
know that. DOI only deals with full-text, not other types of e-sources; ;there
is a lot of other stuff. So how does SFX and DOI work? DOI is a namespace, SFX
makes it Open URL aware to do the authentication, identify the user and list
his services.
http://www.sfxit.com/crossref/prototype1.html
http://www.sfixt.com/CNI2000.html
You can create
local databases with large numbers of digital objects. As long as you make them
Open URL aware, you can use SFX and MetaLib.
2:00-4:00 Wyndham Hotel – West Room
ALCTS/SS Committee to Study Serials Cataloging -- CC:DA report, Mary Grenci; MARBI report;
LC & NSDP repots; subject access, Dr. Lois Mai Chan; CONSER, Jean Hirons;
catalog as portal, Becky Culbertson; Chpt. 12, John Attig
Cataloging the
Web will be published in March by ALA Editions (from the 2-day pre-conference
last summer)
CC:DA report. JSC
reviewed CC:DA Task Force report and the committee will now rework based on
their coments.
Jean Hirons, LC.
Library of Congress is starting to archive digital journals. Revised chapt. 12
will be made available soon; the appendix and other chapter revisions are still
being worked on. Multipart part title
changes – prefer one record using earliest title with newer ones in 24X’s.
Title added entries – the 4 exceptions listed in the rules will be made
optional.
Regina Reynolds.
Library of Congress is ¼ done redoing their serials checkin in their new ILS.
Voyager 2000 will be implemented in May. A new CONSER subgroup for remote
access journals helped revise Module 31 on the web site. Harvard’s effort to
add pattern records to 25,000 – 30,000 journals is progressing. Definitions,
models, minor title changes were discussed at the Harmonization meeting in
England and they finally reached agreement on many issues.
Lois Mae Chan.
Subject Access to Networked Resources and Implications for LCSH. Much of what
she said is at: http://www.alaorg/alcts/organization/ccs/sac/MetaRept2.html
which came from the work of my SAC subcommittee. Now we need to deal with many
languages, schemes, mappings etc. We need to consider display and should
develop guidelines. IFLA is working on this. Trained professionals should be
working on the intellectual subject analysis and computers should do
programmable tasks.
ISBD(CR)
Continuing Resources is a result of the Harmonization meeting. We will start
seeing revisions published. See handout. Continuing resources will use latest
entry or iteration (a continually updated database) whereas serials will use
earliest (the situation where one page gives access to current and older
titles)
CC:DA’s
definition of a continuing resource will put it in Chapter 12: lack of a
predetermined conclusion. It includes finite integrating and continuing
integrating.
Regina Reynolds.
Partnerships to Mine Unexplored Sources of Metadata. Libraries have: big bib
files, experience, authos, controlled subjects. Libraries need: metadata for
lots of stuff. We cannot spend the same amount of time on all stuff. For lower
level stuff, we need some automation. We still need to include it in our OPACs
(even if it doesn’t look like “good” cataloging) We can repurpose metadata.
Lots of metadata already comes into LC – ISSN, ISBN, DOI, ISTC, copyright
registration, Catalogoing-In-Publication, etc.
We need templates that we construct that will encourage creators to
follow our standards (perhaps unbeknownst to them) that creates metatdata we
can re-use. The beauty of it being electronic is we can return it in text or
html depending on type of resource. ONIX-XML is emerging among publishers as a
standards. There is also ISTC
(International Standard Text Code) and DOI. How could forms be improved to get
information we need? Pull-downs, definitions, develop better conversion
programs, interactive instructions, fix cataloging rules, focus on facts, not
form or transcription, eg. Capitalization or Minn. Vs. MN
Becky Culbertson.
California Digital Library. They created a 10th virtual library.
Multiple people from all institutions take part in the work. See handouts. They
basically follow GPO’s approach to add to print records if they exist. They put
all the bibs in a separate file (so print use and electronic use of the record
don’t conflict) They removed local information to make the records generic,
added 920’s to list owning libraries, and used OCLC PURL software to create a
URL for their (CDL) access – the resolver creates a numeric PID
ISST. ISST would
replace key title and may uniform title. It would be a stable title for the
title until a “major” change occurs. You’d get a new ISSN, if a new ISST is
needed. What would be cool is that the person cataloging could push a button
that would send your record to the ISSN staff to add the ISSN or form up the
ISST format.
TUESDAY - Jan.16
9:00-11:30 Library of Congress – Madison Building – National Digital Library Learning Center
Registered
On the CDS page
there are demo files. http://lcweb.loc.gov/cds/cdroms1.html
Leader is Bruce
Johnson bjoh@loc.gov
Listserv:
subscribe desktop firstname lastname to: listserv@loc.gov
Library of
Congress selected FOLIO a number of years ago. It has subsequently been bought
out a couple of times.
Shadow file. A
mechanism through which you do customization of the database. It is a template
and you enter information and make a copy.
Record. A segment
of information; a piece of an infobase; a record in ClassPlus is a line.
Infobase. A database
even though it looks like text.
In the menu bar,
the Window pull down shows you what you have open. Tools is another useful pull
down, but be careful because you can mess it up bad enough to force you to
re-install the software. Under Tools, there is Options, then Query. The first
box, uncheck it, otherwise if you run a search, the previous search is in the
way. Tools, Options, Internet – it automatically selects Internet Explorer so
if you want to use a different browser, click BROWSE, find in Program Files the
program you want, click OK. In Tools, Customize, Toolbars, menus equals pull
down menus. You can change this stuff but it is not recommended; if you do, it
changes the registry of the product (like *.ini files) and you may need to
re-install to correct errors.
The top tool bar
is for functions; the bottom tool bar is for searching. The Document bar gives
you the biggest viewing space. If you want to use Contents, press the number 1
for the main contents, 2 for the next level, 3 for again the next level down.
If you select Browse, you can navigate forward and backward.
If you have lots
of databases open, it should lock up any more. However, the license is BY FILE
– i.e. by Infobase, so if my number licensed users have it open, no one else can
see it. So, close it when you are not using it.
Some common
activities are on function keys. Advanced query = F2. Use quotes around a
phrase, e.g. “uniform title”. F3 is next hit. F4 is previous hit. F5 is
previous operation or at the top it is the left arrow. F1, a separate window
opens and gives help depending on where you were when you hit F1; just X out
when done (the internal window).
If you don’t keep
up with installs, which extends your license, your license will lapse.
Color is
important. Purple means web. Rust-brown means you have access. If you don’t
have it, you get that annoying box.
Searching. F2.
The top left box is a “word wheel”. Use quotations around a phrase to do AND
and ADJ simultaneously. Question mark (?) is a wild card character. (*) is a
wildcard for many characters. F2 to get the window back, then do PREV if you want to modify the search you
just did. The query template is used to replicate a search for an infobase. The
binoculars with the red X eliminates previous searches. The first binocular is
F2. The next 3 query template buttons display depending on the infobase you are
in. Ctrl-Tab goes through all the infobases you have open.
Lots of
customization is possible, but the general message was, it can get messed up
easily and cause more problems than it is worth. So, don’t set up shadow files
or jump destinations (Ctrl-J)
For technical
support help, go to the web FAQ