Work Package 1 of Telematics for Libraries project BIBLINK (LB 4034)
Title page
Table of Contents
|
Metadata Formats
Work Package 1 of Telematics for Libraries project BIBLINK (LB 4034) |
Title page Table of Contents |
1. Please indicate the cataloguing system you use.
BnF is building a new integrated information system that will be operational in 1998. We decided to provide you with information available for existing systems. They will be operational until 1998/1999.
BnF is cataloguing only OFF-LINE electronic documents.
As BnF is the French ISSN centre we have started attribution of ISSN numbers to electronic journals (see scoping document). It is transmitted to the international ISSN data base but NOT included in the national bibliography.
A distinction should be made between the format and the cataloguing system.
Format used : INTERMARC
Exchange format : UNIMARC
Cataloguing rules: AFNOR (National French Standardisation body) standards for cataloguing rules.
Cataloguing rules for electronic documents :
We use the French AFNOR standard : Z 44-082 "catalogage des documents electroniques". It is based on ISBD (CF) version 1.
It applies to both on-line and off-line documents.
Databases :
BN-OPALE 1 800 000 records
Electronic documents (for databases) are recorded in this database.
System :
BN-OPALE GEAC
BN-OPALINE
Bibliographic database for specialised departments
Electronic documents (multimedia electronic documents) are recorded in this database.
System :
BULL database system
2. What data elements do you record for an electronic publication in addition to those you record for a traditional publications (see BnF sample records attached)?
All the information below is both for MONOGRAPHS and SERIALS
For electronic documents, following ISBD (CF) we include :
INTERMARC FIELDS
Descriptive fields
AREA 1 = 245 $ d
AREA 3 = 257 $a type $b size
Fixed fields
Leader
position 7 : 2 values of position 7 : s (program) or t (data)
zone 008
position 23 : s (CD-ROM, CD-I)
t (disquette)
Note field
AREA 5
field 336 : subfield $w
field 337 : subfield $ k = requested configuration
subfield $a = technical information on material
For electronic documents holdings : in field 036 = legal deposit number.
We do not include for the moment any data concerning citation, communication...
In future, we will be able to create a link between the record for the electronic version and records for the same document on another different support (electronic ---- print).
4. What data would you like to include in the record that you cannot find or have difficulty finding from the publications ?
For electronic documents the cataloguing is made from the document itself. This means that we "open" the document. In case of a CD-ROM we do install it on a PC to open it and thus create the record from the content of the disk (not just looking at the commercial printed Technical installation).
We have added complementary information for technical information: we add the type of material used for installation in BnF. This information may be different from the technical / marketing information mentioned by the publisher under the section "configuration requested".
We also add the time of installation.
This information will be later on important for access to the documents by the end-user. If you need half an hour to load a document this has to be taken into consideration for the access service.
4.2 Source of description
The information needed to create the bibliographic record is not gathered in one part of the document as it is in a cover page for a book. You need to browse carefully the CD-ROM and collect the information needed at different places in the CD. It is time consuming. There is no structuring of bibliographic information. This could be requested from the publisher to help the cataloguing.
4.3 Multiple author/ multiple functions
For a multi-media electronic document there is a multiplicity of "authors" based on different functions. The rule is to limit to three authors for the same function. But for a CD-ROM you can have more than 10 different functions : script developer, infographist, designer, title manager, music....This increases the workload for bibliographic description of "AUTHORS".
4.4 Information on terms and conditions should also be included.
For our own digitised collection, we mention information on terms and conditions in the holdings zone.
4.5 We need precise date of publication.
6. What library services will use metadata for electronic publications ?
1. National Bibliographic Agency both for legal deposit and National Bibliography.
2. Specific bibliographic by-products for electronic documents
3. Services in charge of access to the collections.
7. What metadata elements do you consider will be required for electronic publications ?
If we consider the Dublin Core metadata element the following elements could be re-used for bibliographic description of electronic documents :
see also question 4.
Subject : this field raises the issue of the list of references (thesauri) used by the producer of this information. But it can be useful for the cataloguer who will then adapt it to the indexing system used in the library.
If not available in Dublin Core the following information may also be useful :
The question of long term availability of the document can also be discussed. As long as there is no national authority in charge of collecting on-line electronic documents a mention of responsibility for guaranteeing the access to the document in case of a change in the URL can be added. Who is responsible ? the author, the publisher, the host ?
8. Will the records for electronic publications need to be integrated in your existing systems ?
For off-line electronic publications there are already integrated in our system. For on-line, it will be the same. We will have a unique multimedia integrated system for the OPAC and national bibliography. This requirement is very important because we will adapt the existing system to new electronic documents but we will not completely change our system or our format in the next five years.
9. Will the metadata need to be manipulated by particular protocols ?
Yes, the catalogue will support Z39-50 protocol. We plan to give access possibilities to other libraries to the BnF national bibliography records via FTP.
10. Please add any further comments about your experiences in cataloguing electronic publications.
BnF has a solid experience for cataloguing off-line electronic publications. The on-line topic is very important to us and we will benefit from BIBLINK results before starting a national project. The French national bibliography is based on the legislation for legal deposit which includes off-line electronic publications. The problem is to include progressively on-line publications but then we come back to the limits. What do we include : electronic journals, home pages (on which criteria ?)...The evolution towards cataloguing "articles" is also an important issue as we do not include articles in the national bibliography.
1. Please indicate the cataloguing system you use (for example, UNIMARC), and which cataloguing rules you apply.different authority
The Koninklijke Bibliotheek, National Library of The Netherlands, is using the cataloguing system of PICA in Leiden. PICA, the Library Automation Centre in the Netherlands, maintains a database with more then 11.000.000 title records accompanied by records from files: thesauri for personal and corporate authors, common and local subjects, printers and editors, etc. This database is in common use by the KB (i.e. for the Dutch National Bibliography), most university libraries, large public libraries and NBLC, i.e. a company supporting all public libraries.
The Shared Automated Cataloguing system (GGC) is used for exporting data to different OPAC's. The Dutch automated Union Catalogue uses this database for Inter Library Loan. For the German market PICA developed databases with the same structure. On-line input to the database is possible by using the PICA cataloguing format. The PICA format as used in the database is like MARC formats built with tags, fields and subfields. The format for the cataloguers is developed to be more user friendly. In this format for the cataloguers many subfields are indicated with the punctuation used in the ISBD rules. In this document PICA tags are mentioned in the internal PICA+ format. The 'cataloguing tags' are added between parentheses. The PICA format is specified to be able to display the data of title records according to the rules for cataloguing (ISBD).
The rules for the use of the PICA format for electronic publications are published in several documents. These are still draft and they are being developed further while we gain experience in cataloguing on-line and off-line electronic publications. Titles of relevant documents are:
2. What data elements do you record for an electronic publication in addition to those you record for a traditional publication? Also, what data elements do you not record that you would use for a traditional publication? (Sample records may be attached.)
Answer to questions 2 and 3.
To avoid redundancy the answers to these two questions are combined. The KB is gaining experience now with cataloguing electronic on-line and off-line documents in the PICA cataloguing system. For the moment two formats are used in the system: a format for On-line Resources, especially used for WWW and a format in use since 1986 for off-line software products but nowadays also used for off-line publications like CD-ROM's, diskettes, CD-I's etc. PICA and KB are co-operating to develop a format for all electronic resources, both on-line and off-line. In the answers below it has been attempted to indicate what is now already fully implemented in the production line and what is in a phase of testing. Below is described the practice today in KB.
a) Offline monograph (ie CDROM encyclopaedia) =
In electronic resources, more then in folio editions, the exact date of issue is important to distinguish updates.
This tag realises links between the records of the electronic and the printed versions of the same publication.
This local counterpart of fields 4083 is used for off-line material when it is copied to a local document server, for deposit purposes and made available via the local library network through a `local URL' tag.
This is technical information concerning installation and de-installation of the material, and possibly also for conversion purposes. This tag is specifically proposed for use by deposit material.
b) Offline serial (ie CDROM journal)
c) Online monograph (ie dictionary on the Web)
This field contains location and file information. URL and file format are proscribed sub-fields. Most other subfields are not mandatory.
This is a repeatable field and subfields are also repeatable to cater for multiple URLs and multiple formats of the same publication.
It is possible that after an evaluation this tag will be totally revised.
Field 4084 contains location and file information of the images linked to the document.
The purpose of this field is to present an image which is linked to the actual document within the bibliographic description. This was a special feature PICA introduced for the WebDOC project.
N.B. If the on-line resource has an image format itself, the information of this image is entered in field 4083.
These local tags have the same structure as tags 4083 and 4084 and are used when the on-line resource which originally could be found elsewhere on the Internet is copied to a local document server.
d) Online serial (ie electronic journal on the Web)
On-line articles
In the PICA system it is possible to describe on-line articles. For lending systems the following tags are important:
e) Online resource (i.e. an organisation's home pages)
With the tags mentioned above and the tags specified for the traditional publications it is possible to catalogue home pages in the PICA system. Home pages are described as on-line monographs or serials.
3. Which MARC (or equivalent) fields do you use to hold the data element?
a) Offline monograph
b) Offline serial (ie CDROM journal)
c) Online monograph (ie dictionary on the Web)
d) Online serial (ie electronic journal on the Web)
e) Online resource (ie. an organisation's home pages)
See answers to question 2.
4. What data would you like to include in the record that you cannot find, or have difficulty finding from the publication?
a) Offline monograph (ie CDROM encyclopaedia).
b) Offline serial (ie CDROM journal)
c) Online monograph (ie dictionary on the Web)
d) Online serial (ie electronic journal on the Web)
e) Online resource (ie. an organisation's home pages)
1- Importance of the source of the description
Traditionally the content of a publication is regarded as primary information source of the description. Bibliographic information embedded in the publication is considered more authoritative. We feel this is also true for electronic publications. If metadata is obtained through
other sources (for example through contact with the author) it is put between straight brackets [] in the ISBD presentation.
2- terms and condition/ access information: no special fields
At the moment we use several annotation fields. The format for electronic publications is still being developed. It is well tested and if necessary it is extensible.
For information that we cannot put in special fields, we use tag 4201. Often it concerns information that tends to become outdated quickly. Such information requires frequent updates, which we feel is not feasible in our current manual cataloguing practice. Updates should ideally be made by the producer or third party that offers the product/service.
General remarks related to the described document can be entered in the general note area.
Example:
4201 Database is available free of charge as part of the GPO Access System. Internet users should telnet to leocator.access.gpo.gov and login as leocator. Dial in users should use communications software and modem to call (202) 512-1661. No password is required; a screen message will then instruct the user in searching the database
4201 You can subscribe by sending a message through e- mail to: listserver@nix.surfnet.nl, the first line should read the command: SUBSCRIBE Periodical First name Surname
4201 Back issues of xxx are also available as follows; by e-mail: mailserv@nic.surfnet.nl. Send a message with the following contents: help index xxx
4201 If your institute has a licence agreement with SURFnet, you can order the SURFnet Guide 94/95 through your local system manager -Instellings Contact Persoon- (ICP). The Guide 94/95 is also available in the bookshop for f37,50 (including BTW). The ISBN is 90-74719-01-5.
3- technical information: encoded fields
At present cataloguing practice within the KB is to make a copy of the text on system requirements provided by the publisher/distributor into field 4251, which is an annotation field with free text.
There is however for the development of our electronic deposit (DNEP) an urge to record all relevant technical information concerning a particular electronic publication in a SYSTEMATIC way (encoded field proposal). In the processing model for electronic publications for DNEP, the following activities are aimed to ensure the publication will be readable in generations to come:
The provided system requirements by the publishers + the additional technical data necessary for installation and de-installation are to be recorded in such a way that in a future system, the codes can active software libraries and start applications automatically.
5. How have you resolved or attempted to resolve the difficulty?
By developing test-beds.
6. What library services will use metadata for electronic publications?
Examples: facilitating access (giving terms & conditions, access instructions); record supply.
1. OPAC (local retrieval system)
2. Union Catalogue (NCC) (national retrieval system + ILL)
3. Delivery of electronic resources e.g. published by universities (WebDOC)
4. List of electronic publications as a by-product of the national bibliography
5. (possibly in the future) alerting services.
7. What metadata elements do you consider will be required for electronic publications? Examples: description of content (including mention of sub-units, contents pages); relation to other documents
a) Offline monograph (ie CDROM encyclopaedia)
b) Offline serial (ie CDROM journal)
c) Online monograph (ie dictionary on the Web)
d) Online serial (ie electronic journal on the Web)
e) Online resource (ie. an organisation's home pages)
See answers to question 4 on 1) timely info and 2) technical info.
Also content information will be required - but this will not necessarily need to be different from the way printed material is disclosed by subject. But this issue might need looking into furthermore: what enrichment can national libraries provide here, that publishers don't? Controlled vocabularies and thesauri?
At the moment we are working on a conversion model to map the minimum requirements of metadata we would like to receive from publishers to the extended PICA format.
A working group of PICA participants from Germany and The Netherlands is composing a checklist of metadata (inspired by the `Dublin Core'). The metadata of this list which could be embedded in a publication are to be automatically converted to the contents of tags in the PICA format. For the moment the proposal for the list is:
| Labels | Structure | PICA+ |
| Author | <Name>, <First name> <(Addition)> | 028A $d<First name>$a<Name$y<Addition> |
| Next author (repeat.) | <Name>, <First name> <(Addition)> | 028C/0X $d<First name>$a<Name$y<Addition> |
| Corporation as author | Free text | 029F/01 $a |
| Next corporation | Free text | 029F/02 $a |
| Title | Free text | 021A $a |
| Place of publication | Free text | 033A $p |
| Publisher or host | Free text | 033A $n |
| Next place of publication | Free text | 033B/01 $p |
| Next publisher or host | Free text | 033B/01 $n |
| Date, year | dd-mm-yyyy | 011@ $ayyyy$ndd_mm[converted from table]_yyyy |
| Language | In English: <free> | 010@ $axx [converted from table] |
| Next language (rep.) | In English: <free> | 010@ $bxx [converted from table] |
| Edition/update | Free text | 032@ @a |
| Form | Multiple choice | 009P/03 $0 |
| URL | Structured | 009P/03 $a |
| Subject keywords | In English: <free> | 144Z $a |
| Next subject keywords | In English: <free> | 144Z $a |
| Abstract | Free text in English | 047I $a |
For clearness' sake: this a proposal of KB and it is not yet (today is 15 October 1996) discussed in the working group.
Author
Next author
A slight structuring is necessary. The inversion of surname and first indicated by, (comma) is I suppose in common use. With this interpunction we can convert in the majority of cases to the PICA format. This is mandatory for adequate indexing of modern western author names.
E.g.:
Author:Bohemen,Francien van;converted to:$d Francien van$aBohemen(3XXX Francien van @Bohemen); indexed as: Bohemen, Francien van
Author: Van Bohemen, Francien; converted to: $dFrancien$aVan Bohemen (3XXX Francien van @Bohemen); indexed as: Van Bohemen, Francien, and as: Bohemen, Francien van
There must be an opportunity to add different kind of information: on the author as person and about the relation of the author to the publication: composer, editor, translator, etc. I trust it is possible to educate the editors of WebCat publications to enter information between (), parentheses. Conversion of this additional information to subfields f, h or i is possible for different kinds of displaying. I propose the scarcely used $y (addition of title), in cataloguing format to precede by _+_. If there are objections to this solution a new subfield is needed.
E.g.:
Author: Bohemen, Francien van (projectmanager WebDoc); converted to: $Dfrancien van $aBohemen$yprojectmanager WebDoc(3XXXFrancienvan@Bohemen +projectmanager WebDoc). It is the responsibility of the cataloguer to bring the information of the addition in the convenient place of the bibliographical record. Indication as editor, etc. are to be brought in the statement of responsibility of the title field.
The author label is repeatable. Next authors are converted to the field 028C/0X (310X (N), 301X (D).
Corporation as author
It is not convenient to prescribe the WebDoc editors to create any kind of structure in names of corporations. The contents of 029F/01 $a is word based indexed. Therefore conversion from a free text of this label to tag 029F/01 to acceptable.
E.g.:
Corporation: Department of New Testament Studies of the Faculty of Theology; Leiden University; converted to: 029F/01 $department of New Testament Studies of the Faculty of Theology; Leiden University (3121 Department ...); indexed as: department, new, testament, studies, faculty, theology, Leiden, university. After conversion of the contents of the label correct indexing is assured for the time being. In a later stage a cataloguer can transform these date according the bibliographical rules.
The corporation (as author) label is repeatable.
Title
The major problem of the conversion of the label 'Title' to tag 021A $a is the indication of Non Sorting Beginning and Non Sorting Ending. My premise is that we can not ask WebDoc editors to determine articles as non sorting. Accepting this the question is: does PICA have available, or is it possible to develop software to exclude the opening articles of a title from sorting? Is it acceptable to have for a (short?) time titles incorrect indexed? E.g. :De @~civitate Dei, instead of: ~De civitate Dei; The ou cafe, instead of: The ou cafe? Or do we want more intelligent software with a connection to the language? But the indication of the languages is uncertain.
Because of the relative low amount of titles with a wrong NSB/NSE indication I suppose to make simple software: with only a simple checklist of the articles of the most important languages in the WebDoc area (English, German, Spanish, Italian, Dutch) a large majority of titles will have the right NSB/NSE indication.
Place of publication
Publisher or host
I propose to define two labels for publisher and place of the publisher and to make the couple of the two labels repeatable. To distinguish hosts from publishers, host are to be indicated by: [host].
E.g.:
Date, year
In the PICA databases the indexing and display of year (or more specific date) is complex. When the WebDoc editors are following the simple rule 2 numerals for the daytwo for the month4 for the year we realise a convenient conversion to PICA+ . In this case it will be necessary to make a concordance of the numbers of the month and the abbreviation prescribed in the PICA Guidelines for Cataloguing On-line Resources: 01 =Jan .; 02 =Feb .; 03 =Mar .04 =Apr .; 05 =May; 06 =June; 07 =July; 08 = Aug.09 =Sept .; 10=Oct.; 11=Nov.;12=Dec.
E.g.;
Date, year: 29091996;converted to: 011~at $al996$nO1 Sept. 1996 (1100 1996 $ 01 Sept. 1996).
Language
Next language
The PICA language codes are requested for the system. For practical reasons it is impossible to ask WebDoc editors to use the PICA language code list. Therefore a concordance English/German terms to the language code is required. For the English retrieval system of the Dutch Bibliography a machine readable concordance is made: acoAcoliAcoli ...zweZweedsSwedish. This concordance, or a part of it with for our purpose the most important languages, can be used with addition of variant terms: old Greek= grk; Classical Greek=grk; Modern Greek=grn; New Greek=grn. When the label language contains a term not existent in the concordance the PICA language code onb (onbepaald, undefined) is to be generated. [N.B.:the examples of codes are from the Dutch situation.]
E.g.:
Edition/update
Edition/update is here in the traditional meaning: as in paper publications: edition, printing, Ausgabe, Auflage, editie, druk, etc. In electronic publications it is possible that the indication of edition/update is equal to the term of the date of publishing. It is also possible that there are connections with the next label Form. E.g. the contents of the second edition in the Postscript form can be the same of the third edition of the HTML form. The cataloguer must combine the data from the labels Year, Edition and Form for creating bibliographical data according the rules.
Form
A kind of standardisation of the terms used is desired. I refer to the description of the elements in the Dublin Core: 'examples might include PostscriptII document, Windows 3.1 executable file, HTML file, or WordPerfect 6.1 document. The PICA subfield $0 from tag 009P/03 precede subfield $a, the URL in the Dutch set of subfields in tag 009P/03.
E.g.: 009P/03 $0html$ahttp://www.oclc.. (4083 html =A http://www.oclc..)
URL
The URL is constructed according delimited rules. The cataloguer has to control the data converted form the label to the PICA field.
Subject keywords
Next subject keywords
For the moment I suppose to convert the contents of the label Subject keywords to a tag form the local level . The local cataloguer judges the proposal of WebDoc editor, if possible in relation with the abstract, and brings the subject indexing to the common level (basis classification, 045Q). The German participants are inspecting the possibilities of aliassing free subject terms to authorised term.
Abstracts
The maximum length of field 0471 $a must be the same or greater then the length specified for the contents of the label.
8. Will the records for electronic publications need to be integrated in your existing systems?
Yes, the records of electronic publications produced in the PICA format are in the same database with the other title records. It must be possible to have both kind of records in the OPAC's of the library and e.g. in files of the National Bibliography. A total integration is required.
9. Will the metadata need to be manipulated by particular protocols e.g. Z39.50, DE, etc.?
PICA exploits since the end of 1995 an Z39.50 entry to the central database GGC. In view of this it is important to investigate the pros and cons of manipulating metadata by the Z39.50 protocol in comparison with the use of the PICA protocol.
Nowadays we buy tapes with bibliographic records which are integrated in the common database maintained by Pica. As output, tapes are sent to our CD-ROM production company in the United States. Possibilities to use FTP for transfer of the records of the national bibliography for the CD-ROM are investigated.
10. Please add any further comments about your experiences in cataloguing electronic publications.
The cataloguing of CD-ROM and diskettes does not present specific problems. Except for the difficulty in some cases to judge whether a hybrid publication (book + diskette) is mainly printed or mainly an electronic publication.
Concerning on-line resources, especially homepages: At this moment three cataloguers are making bibliographic descriptions for home pages. One for the scientific collection, one for the union catalogue and one for the national bibliography. Further, since a common cataloguing systems is used they can also extract records form the common database and they can see how cataloguers in other organisations, especially one university library catalogue e.g. homepages. The problems with homepages are : author unknown, publisher unknown, date of publication unknown, change of even disappearance of the URL.
BOff-line electronic publications are catalogued and published in the national bibliography.
11. Wish list - anything else?
Not so much wishes but some additional thoughts that have come up here when tackling the cataloguing issue:
1. Importance of the presentation
For the National Bibliography and other metadata products and services of the library, the consistent and user-friendly presentation of records on screen or on paper is very important. ISBD plays therefore an important role. ISBD is a standard for presentation of bibliographical records. In the international library community a proposal for an ISBD(CF), the presentation of metadata of computer files, has been commented upon. In December 1996 the final version is expected.
The terminology of CF (computer file) has already been adapted to ER (electronic resources). In the meantime we have defined ourselves an 'ISBD'-like presentation for 'On-line Resources' and PICA has implemented this in its system. When the new ISBD(ER) will become a standard PICA and KB will as a matter of course adhere to this new standard.
2. Linking between different versions of electronic publication
We would wish to link all versions of the same publication transparently and present the user with all versions available. Another reason for linking different versions is the possible economy of cataloguing efforts. This is extremely difficult however. First of all one must be sure 2 versions of the same publication are exactly the same content wise!
At the moment we have a linking facility between records of the printed and electronic versions of a publication. The record of the printed publication is the master record. The record of the electronic version is appended via a PPN-link. This is the internal PICA record number (PPN). This linkage does not work properly due to indexing complications.
Links between records of different versions of an electronic publication (CD-ROM/on-line) is not possible. Different formats of the same electronic publications are catalogued in the same record: the URL-tag is repeatable and caters for multiple locations and formats of the same publication. Links between journal and articles of the journal (whole/part relationship) should also be realised in the system. In the PICA system this is also done with the PPN.
1. Please indicate the cataloguing system you use (for example, UNIMARC), and which cataloguing rules you apply.
Cataloguing system: BIBSYS-MARC, based on the USMARC
Cataloguing rules: Norwegian cataloguing rules, based on AACR II:
Katalogiseringsregler : Anglo-American cataloguing rules, second edition / oversatt og bearbeidet for norske forhold ved Inger Cathrine Spangen. - Oslo : Norsk bibliotekforening, 1983. - 610 s. - ISBN: 82-990932-0-1
In addition a Norwegian supplement for electronic document (machine readable files), which is mainly used on off-line documents (not totally satisfactory for on-line documents):
Maskinlesbare filer : reviderte regler for beskrivelse med tilhørende eksempelsamling ; samt Revisjoner og tilføyelser til Katalogiseringsregler 1983 / ved Inger Cathrine Spangen. - Oslo : Norsk bibliotekforening, 1990. - 86 s. : ill. - ISBN: 82-90790-01-5
And we are using:
Cataloguing internet resources. A Manual and Practical Guide. Nancy B. Olson, editor.(OCLC 1995) ISBN 1-55653-189-3. - http://www.oclc.org/oclc/man/9256cat/toc.htm
2. What data elements do you record for an electronic publication in addition to those you record for a traditional publication? Also, what data elements do you not record that you would use for a traditional publication? (Sample records may be attached.)
[I have listed all the data elements we record for electronic documents. Data elements in addition to those we record for traditional publications is written in bold-faced type]
a) Off-line monograph (i.e. CD-ROM encyclopaedia)
Same as a) (As far as we know at the moment, we have not catalogued any of these so far)
c) On-line monograph (i.e. dictionary on the Web)
Same as a) (As far as we know at the moment, we have not catalogued any of these so far)
c) On-line monograph (i.e. dictionary on the Web)
d) On-line serial (i.e. electronic journal on the Web)
Same as c) and...
Same as c) (As far as we know at the moment, we have not catalogued any of these so far)
3. Which MARC (or equivalent) fields do you use to hold the data element?
[I have listed all the data elements we record for electronic documents. Data elements in addition to those we record for traditional publications is written in bold-faced type]
a) Off-line monograph (i.e. CD-ROM encyclopaedia)
Same as a) (As far as we know at the moment, we have not catalogued any of these so far)
c) On-line monograph (i.e. dictionary on the Web)
We do not add any links to printed version (if any) of the documents.
d) On-line serial (ie electronic journal on the Web)
Same as c) and...
e) On-line resource (i.e. an organisation's home pages)
Same as c) (As far as we know at the moment, we have not catalogued any of these so far)
4. What data would you like to include in the record that you cannot find, or have difficulty finding from the publication?
a) Off-line monograph (ie CD-ROM encyclopaedia)
b) Off-line serial (i.e. CD-ROM journal)
Same as a)
c) On-line monograph (i.e. dictionary on the Web)
d) On-line serial (i.e. electronic journal on the Web)
Same as c)
e) On-line resource (i.e. organisation's home pages)
Same as c)
5. How have you resolved or attempted to resolve the difficulty?
By asking the publishers and/or the staff in the Information technology department and/or librarians cataloguing the same kind of material in other libraries.
6. What library services will use metadata for electronic publications? Examples: facilitating access (giving terms & conditions, access instructions); record supply.
All library catalogues, the metadata will probably have to be refined
to meet the requirements of the National bibliography.
a) Off-line monograph (i.e. CD-ROM encyclopaedia)
In addition to data elements recorded for traditional documents and elements listed in «2» and «4»:
b) Off-line serial (i.e. CD-ROM journal)
Same as a)
c) On-line monograph (i.e. dictionary on the Web)
Same as a) and:
d) On-line serial (i.e. electronic journal on the Web)
Same as c)
e) On-line resource (i.e. an organisation's home pages)
Same as c)
8. Will the records for electronic publications need to be integrated in your existing systems?
Yes.
9. Will the metadata need to be manipulated by particular protocols e.g. Z39.50, EDI, etc.?
Yes, certainly by Z39.50. And by other relevant protocols (even if we not use e.g. the EDI protocol).
10. Please add any further comments about your experiences in cataloguing electronic publications.
11. Wish list - anything else?
The quality of the metadata elements should be ensured, especially by controlled vocabulary, e.g. name of persons and institutions, subject keywords.
1. Please indicate the cataloguing system you use (for example, UNIMARC), and which cataloguing rules you apply.
For the time being, the Biblioteca Nacional only catalogues off-line publications, both monographs and serials. Deposit law requires deposit of off-line publications but not of on-line publications. Monographs are catalogued in the Computer File module of our cataloguing system, but not serials, due to problems in the linkings between the computer files and the paper records. Serials can be catalogued by the Serials Cataloguing Section provided that they are received through legal Deposit, or by the Spanish ISSN Centre if the publisher asks this Centre for the assignment of ISSN. The same serials format is used for cataloguing for the Biblioteca Nacional database (ARIADNA). Although the fields used in MARC and ISSN sometimes differ, a conversion program from ISSN to MARC has been developed at the Biblioteca Nacional.
The cataloguing of on-line publications has not started yet. If an editor asks for an ISSN for a given on-line journal, he is required to record the information in a diskette, which is used as the cataloguing source.
The cataloguing system we use is IBERMARC, a variant of the MARC format. The cataloguing rules we apply are the Spanish Cataloguing Rules, based on the ISBD.
2. What data elements do you record for an electronic publication in addition to those you record for a traditional publication? Also, what data elements do you not record that you would use for a traditional publication? (Sample records may be attached.)
2. a) Off-line monograph: Specific material designation, File characteristics, Technical data, File type, Related files, Other formats available, Linking entry complexity note. (Sample record attached)
b) Off-line serials: Specific material designation, System requirements, Linkings.
c) d) e) They have not been catalogued so far
3. Which MARC (or equivalent) fields do you use to hold the data element?
3. a) Off-line monograph: 245$h, 256, 538, 516, 530, 580.
b) Off-line serial: 245$h, 500, 775
4. What data would you like to include in the record that you cannot find, or have difficulty finding from the publication?
4. In the case of monographs, sometimes is difficult to determine the technical data (753), the ISBN and the Legal Deposit numbers because they are not shown in the disc label or elsewhere.
In the case of serials we have problems to know if the issue we have in hand is the first one, because it is produced not by an established publishing house but by some kind of firm that has experience in data processing and not in publishing.
In addition, questions as "statements of responsibility" or "edition" are sometimes difficult to determine.
5. How have you resolved or attempted to resolve the difficulty?
In most cases we make a phone call to the publisher and ask for information about frequency, date of publication, title, etc.
6. What library services will use metadata for electronic publications? Examples: facilitating access (giving terms & conditions, access instructions); record supply.
Most library services will use metadata. A problem to be regarded with deposited electronic publications is copyright and a possible unfair use.
7. What metadata elements do you consider will be required for electronic publications? Examples: description of content (including mention of sub-units, contents pages); relation to other documents
8. Will the records for electronic publications need to be integrated in your existing systems?
Yes. The National Library wants to have a single database.
9. Will the metadata need to be manipulated by particular protocols e.g. Z39.50, EDI, etc.?
Yes. At least, Z39.50, ISO 10162/10163, EDI, and any other protocols established by the DG.XIII.
10. Please add any further comments about your experiences in cataloguing electronic publications.
We have few experience on this type of publications but we are very interesting in all problems with this material.
The responses to the questionnaire have been gathered from three different sources. They have been collated into this document.
1. Cataloguing system and tools used - Washington Library Network system (to be replaced in the near future, UKMARC, AACR2r.
2. Fields that differ between electronic and print records - For all types of electronic resource a notes field giving system requirements is needed. (542). Title notes (514) are also needed to give the source of the title.
An Internet resource should have an 856 field giving the URL. Currently, for remote resources, either serial or monograph, no physical description field is given (300).
Electronic resources should also have a 258 field describing file characteristics which is not currently featured in the UKMARC manual.
3. MARC fields for electronic publications: - If the data element is taken as being the information in the record which indicates that the item is an electronic resource, where this appears will vary from item to item.
For physical format fields, the item is described in the 300 field, as follows: 300.00 $f1$ncomputer laser optical disc to indicate 1 CD-ROM.
Absence of a 300 field and the presence of an 856 field or notes fields (5XX) indicating a service provider and login name and password, or system requirements will indicate that the item is not locally held.
The 008 field indicates in both cases that the item is an electronic resource, as the 008 $p contains a K.
Generally speaking, home pages are treated in the same way as on-line monographs are treated, and a date is applied which reflects either the time of cataloguing, or if given, the most recent date of modification.
The General Material Designation [computer file] is not applied as part of library practice at the moment, but may be in the future. If it were to be applied, it would be present in all the types of records mentioned.
4. Data which is difficult to find often includes the following (true for all types of resources, but in particular the on-line variety!) -
Publisher name and location and clear statements of responsibility which describe what an individual's contribution has been.
On some resources (both off-line and on-line) it isn't easy to determine the correct title, a variants on boxes, accompanying information and internal sources often conflict. Some items have no internal title at all (not usually true for CD-ROMs, but certainly true for less "public friendly" productions like databases).
Version information is not always clear, particularly when an updated software version may be released with the manual for the previous version.
Sometimes the stated systems requirements differ between internal and external sources, often not distinguishing between required and preferred hardware and software.
5. Resolving the difficulties. In general, cataloguing rules do not allow you to assume anything, so if information is not available, statements of responsibility are left blank and the appropriate terms for "not known" are inserted.
If a title is not present, the cataloguer is allowed to create one that they feel is descriptive of the item and a note is added to say that it is a given title.
A trial project cataloguing off-line publications has been run and a larger one is planned in support of the extension of legal deposit to non-print items. Provisional policy guidelines are being developed.
6. Library services: - Obviously as the main cataloguing department, we need to be providing metadata for the collections, if the collections are collecting electronic items. Whether we use metadata that has already been encoded as our base record, or whether we use metadata that is made available to enhance our own metadata is a policy issue which as yet hasn't been dealt with.
In addition to facilitating access and record supply, these library functions will also be served by the appropriate metadata -
7. Metadata elements that will be required:- File size and type of content is always useful information (whether there are audio/graphics files, what type of files these are, whether the product is designed to interactive or not).
On-line resources are always more reliably catalogued if they include a date/time stating when the last update occurred.
For both types of serial, it would be useful to know what the frequency is, and whether this is regular, or whether new papers are posted as soon as they are approved for publishing (more likely to be the case with on-line journals).
It would also be helpful to know whether an item bears any relation to another item as this sort of information is included in records when it is available.
For works which already exist in a different form, (e.g. the digitisation of a novel) it would be helpful to know the extent and type of input from individuals other than the author(s) of the original work, to identify the additional access points.
Offline monograph: contents page data; revision/updating arrangements (e.g. via the Web); multimedia content; hardware/software requirements; networkability
Offline serial: contents page data; abstracts; relationships with other publications; hardware/software requirements; networkability; frequency and regularity
Online monograph: contents page data; revision/updating arrangements (e.g. via the Web); multimedia content
Online serial: contents page data; abstracts; relationships with other publications; frequency and regularity; policy regarding timing of appearance of new papers (at regular intervals or as soon as paper approved).
Online resource: frequency of updating
For the British National Bibliography a summary of the contents would be useful.
8. Integration into existing systems: -Yes, as far as practicable and to the same basic bibliographic and format standards. The library is aiming for more uniform access to its collections, so any additional records would be expected to integrate into the library-wide system.
9. Manipulation by other protocols:- They may need to be manipulated by Z39.50 if the library is collaborating on Z30.50 projects. Also probably MARC, HTML
10. Comments and Wish list:- The most important issue for us at the moment is the whole problem of installing an item to catalogue it and not having clearly defined instructions to de-install it. Simply deleting the files is not sufficient, as there are sometimes attachments which disable the CD-ROM or floppy disk for a second installation, and sometimes floating files are left clogging up disk space on the PC. This makes their use in the Reading Rooms impossible. Of 29 items included in the investigation, only 1 had clear instructions in the user handbooks that often accompany the item. Much time can be wasted in looking for the relevant instructions. Quick installation procedures are desirable - the faster the installation process, the faster we can catalogue these types of items, and certainly at the moment, it takes longer to catalogue an electronic item than it does to catalogue a book.
Attachments include 3 examples of serial records, all of which are non-BNB because the current exclusion policy for BNB states that items which are predominantly computer files should not be included. Also attached are 3 monographic records and 2 for kits, also non-BNB for the same reason. (BNB records would have LCSH subject headings, Dewey Classification and availability information.)
| Next | Table of Contents |