National Library of Australia
Kinetica Kinetica
This page is an archive page and will no longer be updated.

Survey on Authority Work

Margaret Kennedy
August 2003

Contents

Executive summary and recommendations
Background
Respondents
Summary of responses to the 2003 survey
Issues arising from the survey
Non-contribution of data to the NBD
Changing environment and resourcing issues
Input and output issues
Data quality
Conclusion
Appendices

Executive Summary

In August 2002 the final report of the Expert Advisory Group on Authority Record Delivery was tabled and discussed at the Kinetica Advisory Committee meeting. One of the ensuing recommendations of the Committee was that a follow-up survey be conducted to ascertain the level of authority work being undertaken in local library systems, but not included in the National Bibliographic Database (NBD).

This second survey was conducted in June 2003. The findings of the survey were very similar to those of the previous survey. Both clearly illustrated the significant quantity of authority work carried out on local systems only, and therefore lost to the NBD and to other Australian libraries. The survey results are also consistent with statistics derived from the Kinetica Amicus database.

The 2003 Survey results indicate that:

  • extensive authority work is carried out, especially in state and national libraries and academic libraries, and by library service providers;
  • the overwhelming proportion of authority work is carried out on local systems, the major factors being staff resources, library policies and priorities, and in the academic sector, avoiding duplication of work;
  • use of the Kinetica Cataloguing Client and lack of a batch uploading facility were also seen to be limiting factors in non-contribution of authority data to the NBD;
  • currency of data in the NBD is seen to be a problem;
  • many libraries source authority data from elsewhere, mainly from the Library of Congress (LC); and
  • some responses indicated a level of confusion about what authority data is, and failed to distinguish between the number of titles added to their databases, or the number of headings automatically generated in their local systems when new bibliographic records were added to the system, and the number of authority records added to the system.

The decline in contribution of authority data to the NBD is of considerable concern. In the year 2002/2003 NBD statistics indicate that only 1605 new authority headings were added to the NBD. Of these 1160 (or over 72%) were added by the National Library and four state libraries, and 180 (or 11%) by a small number of academic libraries. A comparison of these figures with the estimated number of new authority headings created on local systems, 2,933,784 (sourced from survey responses; not to be considered reliable - see last dot-point above), is indicative of the extent of the problem.

This Report recommends that a number of possible solutions be examined to improve the level of contributions. Such an examination should include a cost-benefit analysis of each option in providing a high quality and relevant service to Kinetica customers. Options may include the following:

  • focusing on Australian authority data. Non-Australian authority data is readily available from other sources (including Library of Congress Subject Headings and Name Authorities via KineticaWeb from RLG);
  • entering into an agreement with the National and state libraries (who contribute more than 90% of authority data to the NBD) to facilitate the acquisition of national or state-specific data from these libraries for the NBD.
  • launching a publicity and promotion campaign to educate Australian libraries about the benefits of an Australian authority file, Kinetica authority record standards and the types of services Kinetica can offer. Promote existing Kinetica support for authority data - searching via Client, KineticaWeb and Z39.50; downloading via Z39.50 or email/ftp to a local file;
  • offering a range of products more tailored to the provision of Australian authority data to Kinetica customers (eg only records with references or notes);
  • considering the development of an alternative means of online input of authority records (eg Web update);
  • considering the development of an unlinked authority file which supports batch loading and matching of authority records.

Some of the options detailed above are considered problematic. The option, of providing an alternative means of authority data input, would be of limited value. As stated in the report, the reluctance to use the Kinetica Cataloguing Client is not considered to be the primary factor in the decision to use local systems for authority work. The major factors are identified in responses as resourcing and workflow issues and library priorities.

The option of batch loading of authority data is problematic as it is much more complex than batch loading bibliographic data. Authority records have very few reliable match points. If the Authority file is to be linked to the NBD bibliographic file then accurate matching also requires a comparison of the bibliographic record(s) linked to the database heading (or authority) and those linked to the incoming authority records, and of course the incoming authority records will not be accompanied by bibliographic records. This scenario may also require matching of an incoming authority record with only a heading (not an authority record) on the NBD. If batch loading of authority data from Kinetica customer libraries is to be attempted then an unlinked authority file may be the only practical option. Irrespective of the approach taken, batch loading of authority data could impact on the degree of duplication in the NBD, and would result in a significant increase in resources required for database cleanup and manual review.

A combination of options is recommended. Our major contributors of authority data are the national and state libraries. Working closely with the these libraries and acknowledging the contribution they make to the NBD, even to the extent of providing monetary incentives for the contribution of authority data, would reward those libraries who currently contribute in excess of 90% of authority data to the NBD, and would encourage those who do not currently contribute. The contributions of these libraries would provide a unique, comprehensive, reliable and up to date source of Australian authority data for libraries, and costs could be passed on with the availability of a range of Australian authority products that would be of relevance to them. This is a market that is not adequately served by other sources of authority data. Meanwhile, we can concentrate on encouraging contribution of bibliographic and holdings data from all Kinetica customers, both online and via Batch*Link, in the interests of resource sharing.

Background

In 2002 the Kinetica Expert Advisory Group on Authority Record Delivery (EAGAD) conducted a survey of Kinetica users and reported on a range of issues relating to authorities and the National Bibliographic Database (NBD). The final report of the group is available online at http://www.nla.gov.au/librariesaustralia/eag_ad.html. One of the EAGAD recommendations was "That a more extensive survey be conducted to identify the volume of authority records created on local systems, and to determine the extent of the market for NBD authority records".

The 2003 survey has been carried out to gather information on the extent of authority work being undertaken in Australian libraries on their local systems. The results of the survey provide a valuable base for the National Library and Kinetica Advisory Committee to consider the scope and range of issues relating to authority work that needs to be addressed in the future. This Report outlines the results of the survey.

Respondents

Seventy-four respondents completed the 2003 survey. Of these, twenty-seven responses were received from academic libraries, nine from government libraries, seventeen from public libraries, one from a school library, eight from special libraries, six from state and national libraries, and six library service providers.

Respondents by Library Type
Figure 1: Respondents by Library Type

Some respondents did not comfortably fit into any of the categories listed on the form; these have been separated into the additional category of Library Service Providers.

Summary of responses to the 2003 Survey

  • There is a strong correlation between the results of the 2002 and 2003 surveys. Both drew responses from more than seventy Kinetica users (1) , representing a range of library types, and both surveys highlighted similar issues regarding the contribution of authority data to the NBD, and the use of the NBD as a source of authority data.
  • Most of the respondents from State/National libraries (five of six) and Library Service Providers have authority policies (five of five). Five of seventeen public libraries, eight of twenty-seven academic libraries, five of seventeen public libraries and three of eight special and school libraries also had policies (2). Only one government library respondent (of nine) stated that the library had an authority policy. Some respondents stated that they realised the need for an authority policy, and were developing the policy at present.
  • There was clearly a lack of understanding of what authority control entails. Consequently there were inconsistencies in responses to a number of questions relating to authority control. Only thirty-seven of the seventy-four respondents gave a positive response to the question of whether they carried out authority control. However, sixty respondents claimed to carry out authority control on their local system, and fifteen claimed to carry out authority control on the NBD. From some of the comments on authority control, it appears that some respondents considered that the automatic generation of headings when new records were created on or downloaded to the local system constituted authority control.
  • Library service providers would appear to carry out the most rigorous level of authority work on behalf of their clients. Authority work in academic libraries is thorough and far-reaching, with many of the respondents checking all new headings and carrying out necessary authority work on them. In addition, on-going maintenance work is carried out to resolve conflicts. While authority work in academic libraries is considered desirable, even essential, several respondents cite a lack of resources as the reason for limiting their authority work to conflict resolution only.
  • The situation is similar in state/national libraries. In addition, there appears to be little authority work carried out on headings created as a result of copy cataloguing, because of lack of professional staff.
  • The most significant factors identified by respondents that limit their authority work to the local systems are:
    • staff resources
    • library priorities
    • avoiding duplication of work (academic sector)
    • lack of staff skills (Public sector ).
    • many other factors, including cataloguing client issues, lack of uploading and downloading facilities, and the need to use local or specialist thesauri. (These are detailed in Appendices 3 and 4)
  • Most respondents cite their authority sources as Library of Congress (LC) Subject Headings and Library of Congress Name Authorities (from LC, OCLC, or Validator), and Kinetica for Australian headings. There are some local practices and thesauri in place, especially in public libraries (eg local headings for local history), special libraries school libraries (fiction headings) and library service providers (SCIS).
  • Some respondents were unaware of the facility to acquire authority data from the NBD via Z39.50 or via ftp/email from Stored records in KineticaWeb.
  • Most respondents see themselves carrying out the same amount of authority work on their local systems or the NBD/CJK systems in the future as they do now.
  • Thirty-one respondents predict that the level of authority work on their local system will increase, and two predicted that their level of authority work on the NBD/CJK systems would decrease.
  • Only a small proportion of respondents (eleven) have purchased authority records from Kinetica / ABN previously, and the same number state that they may purchase these files in future. Ten sites (half of them academic libraries) have purchased authority files from other sources, mainly from the Library of Congress
  • A much greater proportion (thirty-three sites) have used Z39.50 or copied and pasted individual records from various sources. Once again the predominant source is Library of Congress.

Issues arising from the survey

The issues relating to authority work can be broken down into the following categories:

  • Non-contribution of data to the NBD, and the loss of consequent efficiencies in the ability to share and reuse data;
  • The changing environment, including local system issues, other sources of data, changes in libraries;
  • Input and output issues, including client issues, downloading and/or uploading facilities, and products;
  • Data quality; including adherence to standards, currency of data, Library of Congress versus Kinetica headings, and the importance of an up-to-date and accurate Australian authority file, especially for Australian names.

Non-contribution of data to the NBD

From the survey responses it would appear that academic libraries and library service providers are the most prolific sectors when it comes to the creation and editing of authority data on their local systems. This is not surprising, given the number of respondents from academic libraries, and the fact that several have added large files of authority data (from LC) to their local systems, and given that library service providers undertake work on behalf of a considerable number of other libraries. Data from public and special libraries would appear to be unreliable; comments indicate that some sites actually counted the number of titles acquired and catalogued onto their local system as constituting new title authority records. Most of these sites did not count name, subject and title authorities separately.

The more telling factor is the comparison of the statistics (3) r the creation of authority records on local systems as compared to those for the creation of authority records on the NBD, with 2,933,784 new authority records added to local systems (4) over the last year, compared with 3,647 added to the NBD. The statistics for changes to authority records are also disparate, with 45,0941 (5) on local systems, and 2,400 on the NBD. A summary of responses can be found in Appendix 2. It should be noted that there is a discrepancy between the figures from the survey and those extracted from the NBD (6) . Whichever figures are considered, there is demonstrably a significant amount of authority control happening in local systems, which is lost to the NBD, and to other Kinetica users.

Authority Work on Local System, NBD and CJK
Figure 2: Authority Work on Local System, NBD and CJK

Changing environment and resourcing issues

The survey responses highlight a number of factors behind the decision of many libraries to carry out authority work on their local system, instead of on the NBD. These factors include:

  • the improved functionality of local systems to support authority data;
  • the pressure on staff resources in libraries;
  • the need to be more efficient and integrate work processes;
  • the need to avoid duplication of effort by doing work on different systems; and
  • use of other authority data sources being perceived to be more cost-effective for libraries than having scarce staff resources creating authorities on-line.

Factors limiting authority work to local system
Figure 3: Factors limiting authority work to local system

The result is that a significant proportion of authority data is not shared with the Australian library community, because it resides on local systems, instead of in the NBD, where it could be used by all. While carrying out cataloguing and all associated work on local systems may be seen by many to be the most efficient way of ensuring that the data is available on the local system quickly, the benefits of resource sharing, including the availability of sharable authority data, would appear to have a lower priority. Most sites appear to appreciate the benefits of sharing bibliographic data; contribution of this data, especially via batch load, continues to grow. This growth is important to ensure that adequate bibliographic and locations data is available to underpin interlibrary loan and other resource sharing activity. For authority data, however, both the number of contributors, and the level of contribution of authority data continue to shrink, with only approximately 40 libraries contributing or editing authority data in 2002-2003.

Input and output issues

The output of Kinetica authority records to local systems is not a problem. Records can be copied from Kinetica using Z39.50, or via the ftp or mailing of saved records from KineticaWeb. Files of records can be requested via the Kinetica Authority Records Service. Some changes have been suggested to both the output of authority records (eg an RMR facility like that for MARC bibliographic records, which triggers downloading of the record via TPS*DLL or MARCLink). Some suggestions have been forthcoming on the Kinetica Authority Record Service, and the need to be more flexible in providing files that meet the requirements of users (eg Australian name headings only, only records with notes and references). The Kinetica Authority Records Service already provides files of Australian authority headings only, including Name authorities only. It is timely to both advertise the products we already offer, and review the range of existing products to determine what use is currently made of the service, and whether the availablility of alternative products would increase the use of the service.

While many libraries have opted to contribute bibliographic and holdings data via Batch*Link, this option is not available for authority data. The Kinetica Cataloguing Client is the only means of contributing authority data to the NBD, and people are reluctant to use this interface. As stated above, the use of local systems in preference to the Cataloguing Client is very much a workflow efficiency measure in libraries, and a proportion of the libraries currently using their local system for authority work have stated their willingness to add this data to the NBD, should a means of batch loading become available.

Batch loading of authority data is a complex issue. Bibliographic records have a number of match points, some more reliable than others; yet batch loading of bibliographic data via Batch Link has resulted in an increase of duplicate records on the NBD, and a significant increase in the number of records requiring manual review. Authority data has very little in the way of reliable match points. If the data has been downloaded from the NBD, edited, and then uploaded, it will have an Amicus number, but new records from local systems will not have an Amicus number. Incoming authority records will be matched against headings, not authority records. Minor differences in punctuation or coding (eg a period or space in the wrong place, or an incorrect subfield code), could result in a duplicate record in the Authorities database. Accurate matching of incoming authority data would also require comparison of the bibliographic record(s) linked to the database heading or authority, and those linked to the incoming authority records, and there will be no incoming linked bibliographic records linked to the authority records being loaded. If batch loading of authority data is considered, the incoming data will require to be matched against an Australian authority file that is not linked to NBD bibliographic records and headings. This means the creation of a "stand alone" authority file.

However, if the NBD is to stay relevant as a reliable source of Australian authority data, some investigation of alternative input options should take place. In addition, Kinetica needs to stress the importance of resource-sharing, not just in an "it's the right thing to do" framework, but in the context that contribution of high quality data enriches the NBD, making it more up to date and relevant, and making authority data easily accessible to all (and subsequently more cost effective than the current situation, ie each library duplicating the work being done by others, each on their own local system.

Data quality

A number of respondents questioned the quality and currency of authority records on the NBD. Criticism of the currency of data is probably well-founded, given the small number of libraries that contribute to and maintain authority data on the NBD for others to use. In the words of one survey respondent:

"Many libraries, like us, are not contributing as much as we should, hence the records are not as current as they need to be."

Criticism of data quality, on the other hand, is not so well founded. Some respondents to the 2003 survey remarked on the need for Kinetica to "follow LC" and "follow LCSH practice". Some responses also refer to inconsistencies within headings in bibliographic records on the NBD.

Kinetica does follow LC practice for authority headings. Kinetica practice is described on the Kinetica Website at http://www.nla.gov.au/librariesaustralia/standards.html . Variations to "LC practice" occurs in the case of Australian Name and Subject headings, where an Australian extension of LC headings is maintained. In the case of subject headings, every effort is made to minimise conflict with LC headings. As stated on the Kinetica Website:

The future development of the Australian extension to LCSH should take account of the changing environment of Australian cataloguing. Australian libraries are obtaining their cataloguing data from a variety of sources (Kinetica, overseas databases, library supplier databases) and they need to contain the cost of maintaining the authority files in their local systems. Accordingly, the development of the Australian extension to LCSH should follow these guidelines:
  1. Where a term already exists in LCSH, an equivalent Australian term may be adopted as a reference to the LCSH term. Australian terminology should not be substituted for synonymous LCSH terms.
  2. Australian subject headings which have previously been substituted for existing LCSH terms, should be reviewed in order to reduce the variations from LCSH. The review should take into account the cost of converting back to standard LCSH, which may outweigh the savings of being consistent with LCSH.
  3. The Australian subject headings which have already been adopted, should be continuously monitored for opportunities to reduce the variation from LCSH.

In the case of Name headings, the following statement applies: Kinetica will follow the form of heading established in the Library of Congress Name Authorities (LCNA). LCNA will take precedence over name, title and series headings which have already been verified as authorities in ABN or Kinetica, and these headings may be retrospectively changed to conform with LCNA.
An exception will be made for Australian name, title and series headings, because authority records for Australian names, titles and series will usually be established earlier and more accurately by Australian libraries. Therefore, authorities for Australian names, titles and series, which conform to AACR and the Library of Congress rule interpretations, should not be changed retrospectively simply to conform with LCNA. In order to improve the quality of Australian name, title and series headings in LCNA, the National Library of Australia will offer to the Library of Congress a file of Australian name authorities. To enable Australian headings to be identified within Kinetica, they should be coded as follows: USMARC field 008, character position 10, code 'z' This is a valid USMARC code signifying 'other descriptive cataloguing rules', but will be converted to 'c' on output from Kinetica, as the headings will conform with AACR2.

Data Contribution

The following table details contributors of authority data to the NBD between July 2002 and June 2003. By far the greatest contributors are the National and State Libraries, with 88% of authority record additions, changes and deletions being carried out by the National Library, State Library of New South Wales, State Library of Tasmania and the State Library of Victoria. The next largest group of contributors is the academic library sector, with 4.4% of additions, changes and deletes contributed by thirteen universities and the South Australian TAFE network (7) .

Library Type
National/State Academic Government/Special Public Kinetica Administration Other (8) Total
New 1160 180 48 4 26 187 1605
New % 72.3 11.2 3 0.2 1.6 11.7 100
Changed 8450 158 15 5 28 290 8946
Changed % 94.5 1.8         96.3
Deleted 224 154 2 1 235 0 616
Deleted % 36.4 25     38.1   99.5
Total 9834 492 65 10 289 477 11167
Total % 88 4.4     2.6   95

Figure 4: Contribution to authority work on NBD by library type

The contribution of authority data has been an ongoing issue, both for Kinetica and for its predecessor, the Australian Bibliographic Network (ABN). A report written by Warwick Cathro and tabled at the ABN Network in February 1989 forshadowed that, with the increasing availability of local library systems, by 1995:

"…software support for cataloguing transactions would be decentralised onto local systems. ….It will be quite difficult to maintain a national authority file in a decentralised manner. Authority control work involves the resolution of conflicting records, the merging of records and the updating of records, more than it involves the creation of new records. It would be much easier to manage this by reference to a centralised authority file. I envisage a small group of major libraries, perhaps 20 to 30 in all participating in this process…" (9)

This prediction would appear to have been correct. An examination of the growth of authority data from 1990 to the present shows a steady decline in contribution of authority data, with the decline being more significant from 1994.

It is also worth noting that during the early 1990s a considerable amount of authority work was outsourced to a number of contractors, particularly authority work on government names (National, State and local government) (10)

Growth in authority records
Figure 5: Growth in authority records

The decline in authority work on the NBD (11) cannot therefore simply be attributed to a reluctance to use the Kinetica Cataloguing Client, which only came into use in 1999. In ABN days, authority work was cumbersome, with transactions passing through a "Nitely" process to load it to the database. When complex authority work was in progress (eg merging duplicate authority records), it could take many days, waiting for Nitely to process each change before the next step could be undertaken. This process was much more cumbersome and labour-intensive than using the Kinetica Cataloguing Client.

It would appear that the decline in authority work on the NBD is, to a much greater extent, due to other factors outlined above. Decreasing resources, lack of skills, the availability of authority data from other sources, and shrinking library budgets make it appear to be more cost-effective, at face value, to buy in whatever data is available, and to either carry out other authority work on the system, or to avoid it altogether

Conclusion

While the 2003 survey has failed to deliver reliable and consistent results in some areas, it can still be taken as being indicative of the high proportion of authority work being undertaken in local systems as opposed to the NBD. The continuing downward trend constitutes a significant risk to the long tradition of resource sharing in the Australian library environment, as far as authority data is concerned, and casts doubt on the credibility of the NBD authority files as the definitive source of Australian authority data.

There are some major issues to be considered in determining the future of the NBD as the definitive source of reliable and up-to-date Australian authority data. The first of these is whether NBD authority data is still relevant to, and valued by the Australian library community, and indeed valued as a component of an international authority file, such as the Virtual International Authority File, or the Anglo-American Authority File. The use of Kinetica as a source of Australian authority data by most survey respondents would appear to indicate that the NBD is still valued.

If the NBD authority file is still seen to be relevant and valued, there are two further issues; the first being to examine what can be done to encourage Australian libraries to contribute to the NBD authority files, and the second to determine what can be done to acquire and load the wealth of Australian authority data currently residing only on local systems.

This Report assumes that the NBD authority file is still valued, and recommends the consideration of a range of factors that may motivate Australian libraries to become active partners in resource sharing of Australian authority data by contributing their authority data to the NBD. Consideration should be given to alternative methods of enabling this contribution to occur, as well as the possibility of some form of recognition and/or incentive for the contribution of data. Since the overwhelming proportion of authority data is contributed to the NBD by the National and some state libraries, it would appear logical to target this group of libraries in considering encouraging or rewarding contributions.

If the quality of the NBD is improved by increased contributions of Australian authority data, especially by the state libraries currently not contributing, the NBD will provide a unique and valued source of data to onsell to Australian libraries. The costs of any incentives provided to contributing libraries could be offset by the vigorous marketing of a revised range of products to Australian libraries, which would be relevant to their needs (as outlined in the 2003 survey responses), and could form part of the proposed Australian National Bibliography service.

In addition, the marketing campaign should be aimed at convincing libraries of the value of the NBD as an immensely important shared asset, and fostering a sense of ownership and responsibility for the NBD as a whole, and not just as a unique repository of Australian bibliographic and holdings data.


Appendices

  1. Respondents by Library Type
  2. Amount of authority control - local systems and NBD/CJK
  3. Limiting factors in carrying out authority control only on local systems
  4. Limiting factors - Summary of comments
  5. Future intentions for authority work
  6. Future intentions - Summary of comments
  7. Use of Kinetica Authority Records Service
  8. Kinetica Authority Records Service - Summary of comments
  9. Other sources of authority data
  10. General comments
  11. NBD authority activity 2002/2003
  12. Survey Instrument

Appendix 1: Respondents by Library Type

Library Type Number of respondents
Academic 27
Government 9
Public 17
School 1
Special 8
State/National 6
Library Service Provider 6
Total 74

Appendix 2: Amount of authority control - local systems and NBD/CJK

Library Type
Academic Government Public School Special State/National Library Service Provider Total
Number of respondents by library type 27 9 17 1 8 6 6 74
Authority control on local system 24 4 16 1 6 4 5 60
ILMS New Total 2847783 2227 25681 3300 160 13275 41358 2933784
ILMS Amend Total 254324 2830 3422   2552 172465 15348 450941
Authority control on NBD 8 1 1 0 0 4 1 15
NBD New Total 437         3210 50 3647
NBD Amended Total 1958 23 10     299 110 2400
Number who carry out authority control on CJK 1 0 0 0 0 0   1
Authority Work on ILMS and NBD/CJK 6 2 2 1   2   13

Appendix 3: Limiting factors in carrying out authority control on local systems only

Library Type
Academic Government Public School Special State/National Library Service Provider
Respondents by library type 27 9 17 1 8 6 6
Authority control on local system 24 4 16 1 6 4 5
Limiting factors-
Staff resources
15 3 13   4 2 2
Limiting factors-
Staff skills
5   10   2 1  
Limiting factors-
Library priorities
14 2 12 1 4 3 2
Limiting factors-
NBD search costs
7 3 4   2 2 1
Limiting factors-
Duplication of work
15   3   1 1 3
Limiting factors-
Local system issues
3   4        
Limiting factors-
Local system practices
3   6 1 3 3 2
Limiting factors-
Other
13 1 2 1 2 2 3
Other -
Too expensive to use contract cataloguer
  1          
Other-
No way to contribute without Cat. Client
        1    
Other -
No facility to merge auth. Records both ways
          1  
Other-
Have not considered. Plan to upload bib & holdings
        1    

Appendix 4: Limiting factors - Summary of comments

(Some comments have been combined, and names of libraries and library systems removed. Otherwise comments from respondents have not been edited)

  1. Academic libraries
    • Cataloguing Client: awkward; slow response times, limited number of licenses; difficulty of working with name-title headings; difficulty of resolving variants and duplicate forms (a number of responses) Local system: ability to import authorities will mean extra cost and generate duplicate headings (use and maintenance issue) We concentrate on headings where we have specialist information and/or information to resolve conflicts etc.
    • As part of a larger system, we follow their policy and practice. If we were to do authority work, it would be within our local system. Unlikely to do it because of lack of staff resources, priorities within the library, and duplication of work.
    • Authorities created on Kinetica cannot be downloaded.
    • NBD is not used for cataloguing so there's no reason to use it for authority work, also our local system offers considerable advantages for cataloguing and authority work.
    • We follow US NAF for non-Australian headings and RLG Z39.50 is so fast; could not justify time to duplicate this work onthe NBD.
    • Not a contributing member for records, just for holdings. Authority work usually involves changing subject headings listed in the Cataloging service bulletin.
    • Some work is done on the NBD, especially for Australian headings and conflicts or incorrect headings that we have imported with bibs. Most work is done by importing LC or Kinetica authority records
    • Where work is done for non Australian headings, it is mostly done on the local system only. Incorrect authorities are reported to LC for non Australian headings. we do not use non standard headings or local notes. All is done to standards and worth sharing. It would seem more efficient to do our own authority work (several responses)
    • In the case of the CJK database, our staff who have the language knowledge to work in the system have no authority work training or experience.
    • Inability to conveniently upload to the NBD the work done in the local system (using headings from Validator).
    • We do not follow Kinetic Authorities
    • We have always batch loaded our data to the NBD and there has never been a facility to merge authority records both ways. We have never worked online with either ABN or Kinetica. Our adherence to LC standards has precluded our contribution to some extent.
  2. Government libraries
    • Cleanups of old authority data on system are sometimes needed to maintain consistency of local system.
    • Issues as identified plus too expensive to use contract cataloguer PLUS No financial incentive for the Library.
  3. Public libraries
    • We only duplicate authority work on the NBD for authorities related to our Local Studies Collection. Staff costs/time prohibit any other authority work on the NBD
    • Slowness of Client software. Numerous steps involved to make changes. Software is not user-friendly.
    • Do not contribute cataloguing to the NBD, download records from Kinetica only.
    • we contribute our holdings, but not cataloguing records. Eg, if we purchase a title not held in NBD, we note it and search again later for someone else's record of the item.
  4. School/Special libraries
    • We are a private institution so do not wish to have our holdings listed. We have many resources dating pre WW1 or have only copies in Australia. These would be requested for inter-loan. We do not have the staff to do this. Reciprocal rights would be of limited benefit to us.
    • We do not use the Cataloguing Client so have no way of performing authority work on the NBD
    • Speed of doing work on our local system as opposed to doing the same work on the NBD, i.e. time taken to create an average authority record.
  5. State/National libraries
    • The slowness of the connection. Correcting one authority would take about 10-15 minutes.
    • We do undertake some authority work on the NBD, but it is limited by the above factors.
    • As we plan to contribute our bibliographic and holdings data via BatchLink (or some other utility), we have not considered contributing authority records onto the NBD.
  6. Library service providers
    • We have always batch loaded our data to the NBD and there has never been a facility to merge authority records both ways. We have never worked online with either ABN or Kinetica. Our adherence to LC standards has precluded our contribution to some extent.
    • Time required to carry out Kinetica input: this is not the same issue as the costs of searching but it works out as a cost issue also. It is not the staff time so much as the system response time which makes this out of the question. Time/cost of obtaining Kinetica authority records
    • [Site] add authority records to its own database which is specifically designed for the needs of school libraries.

Appendix 5: Future intentions for authority work

Library Type
Academic Government Public School Special State/National Library Service Provider Total
Number of respondents by library type 27 9 17 1 8 6 6 74
Authority control on local system 24 4 16 1 6 4 5 60
Authority control on NBD 8 1 1 0 0 4 1 15
Number who carry out authority control on CJK 1 0 0 0 0 0   1
Local system -
Anticipate decrease
  1 1   1   1 4
Local system -
Anticipate increase
8 4 9   3 6 1 31
Local system -
Anticipate same
19 4 7 1 5   4 40
NBD/CJK -
Anticipate decrease
  1       1   2
NBD/CJK -
Anticipate increase
3   2   1 3   9
NBD/CJK -
Anticipate same
24 8 15 1 7 2 6 63

Appendix 6: Future intentions - Summary of comments

(Some comments have been combined, and names of libraries and library systems removed. Otherwise comments from respondents have not been edited)

  1. Academic libraries
    • We see local authority work as taking precedence over NBD, and cataloguers' time is limited
    • Feedback from clients and reader services when there is conflict or inconsistent name headings.
    • Current library automation system does not allow for downloading of NBD authorities. Due to staffing constraints minimal work is done on local database. We rely on the correct format for headings on the Kinetica records.
    • Following the EAGAD report, University of XX considered the implications of the report for the University library and made recommendations for improvements to the delivery of records. Investigations into the feasibility of these recommendations are currently taking place.
    • This question is almost impossible to answer and is dependent on technical developments and resources available In third quarter of 2002 we loaded the full retrospective file of Library of Congress Subject Headings to our local system. We have made decisions to use LC headings and not Kinetica variations. Our resourcing priorities must be to our own system.
    • New local system is allowing us to identify authority record gaps more easily, and reports with new system are better for following up problems. This combined with retrospective effort on serial records, and training/permission now given to copy cataloguers to bring in standard authority records where they exist should mean that additional authority records are utilised as part of normal workflows. N/A. We believe we do sufficient for our purposes
    • With implementation of new functionality in ILS, the ease of loading authority files should enable an increase in authority work
    • We want to "clean up" our name headings, some of which are abbreviated and inadequate, having migrated from an earlier inventory file·
    • Improvement to local ILMS
    • Our current practices are perhaps not ideal but they do work and we are not anticipating any major changes in the foreseeable future
  2. Government libraries
    • Unlikely that any staff losses will be replaced and authority work would be of low priority with reduced staff.
    • We have only begun creating authorities within the past year and aim to have our entire database under authority control within two years. This is a major project for our Cataloguing team.
    • Training in authority work, and on the Kinetica database
    • I hope to be able allocate staff time to this job.
    • An upgrade to our system will make authority work easier. Currently, local system issues inhibit our doing authority work.
  3. Public libraries
    • The implementation of a new library management system will facilitate authority work.
    • Hopefully staff time will be available to raise the priority.
    • The authority work on our local database has not been done for years, needs to be cleaned up.
    • Upgrade of local system allowing easier use of authority control programs. Change to windows based system requiring more authority work.
    • Delivery of catalogue over the Web
    • Access to global change facility in new system. Currently this is planned for the next year.
    • Current project to clean up authority files prior to and following a system upgrade. Names complete, started on subjects. Following this, general maintenance only expected as part of normal cataloguing task.
    • Cataloguing backlog has gone - more time to do authority work.
    • New Library management system and changed work practices.
    • Still cleaning up the database after merging 3 libraries catalogues 7 years ago.
    • need to serve public well, and knowledge that current local authorities are in a parlous state
  4. School/Special libraries
    • URL cataloguing
    • We're in the process of converting our catalogue from XX systemto YY system. Many (many) errors have been found and are being cleaned up as part of this (slow) process.·
    • We implemented a new LMS 15 months ago and it is only now that the system is properly bedded down that we have the time available to devote to tasks such as authority work. During 2003 this work was a lower priority than other tasks hence the relatively small number of records created. We have already increased this considerably so far this year and would expect that to continue from now on.·
    • We have done a number of retrospective conversions over the past two years and much authority work has resulted from these. I expect the number of new authorities to decrease once this work is completed.
  5. State/National libraries
    • Possible outsourcing of some technical functions freeing up staff to undertake other duties.
    • If the XX transfers its cataloguing operations to [new ILMS], the performance of authority control is likely to also be transferred to [new ILMS].
    • Difficult to assess for a number of reasons. In the last financial year a lot of editing has been due to the migration and data clean-up. To the end of 2001 we loaded weekly LCSH tapes and ran reports. To date this has not been possible with the new system. When we are able to do we anticipate another peak in authority work. Another peak is temporary increased funding for library materials this financial year and the next 2 financial years due to a large extension to the building and increased funding for additional library materials
    • We are currently setting up the authority record formats we require on our local system, after which we will commence creating authority records inhouse as required, as a test phase. Later, we will probably load authority files.
  6. Library service providers
    • Batch contribution would be a major factor that would affect our level of contribution to the NBD.

Appendix 7: Use of the Kinetica Authority Records Service

Library Type
Academic Government Public School Special State/National Library Service Provider Total
Number of respondents by library type 27 9 17 1 8 6 6 74
Purchased Kinetica authorities previously 6 1 2     2   11
Likely to purchase Kinetica authorities 5   3     3   11

Appendix 8: Kinetica Authority Records Service - Summary of comments

(Some comments have been combined, and names of libraries and library systems removed. Otherwise comments from respondents have not been edited)

  1. Academic libraries
    • Easy download
      Free searches in support of authority work (incl. to select correct authorities
    • Keeping up with LCNA + LCSH is impossible.
      If *only* Australian entities had authorities on Kinetica, and they were constructed to NACO standards (names, titles) or LCSH (subjects), we could rely on them and it would be worth making the effort. While standard, maintenance and content of Kinetica authority records falls short of these standards, the NBD can't be taken seriously as a source of Australian authorities.
    • Would use the Kinetica authority record service if it could be easily downloaded into our local system at reasonable cost.
    • Purchasing the authority records is a Management issue.
    • A ready made record would make it easier and less time consuming for us.
      For new Australian subject headings there is discussion group where by consensus is sought on particular headings. This cannot be done by individual libraries. A central system would be better.
    • If Kinetica authorities were maintained and kept up to date.
      We did purchase authority files from ABN which formed the basis of the authorities in our catalogue, and now we maintain them individually.
    • Direct downloading of headings with see refs into a local system. An accurate file. Kinetica needs a massive cleanup and all duplicate and incorrect headings should be removed.
    • Full compliance with LC name and subject authority files
    • Ability to downline load individual authority records from Kinetica as required.
    • Cost of authority tapes is expensive, in addition there is the cost of setting up a separate load table. When authorities are obtained in these large batches, if any problems occur with the load, the volume of manual corrections can be overwhelming
    • We have purchased records in the past we had concerns about their currency and we found some headings purchased as Australian were not in fact Australian. This resulted in considerable unnecessary work fo us.
    • Appropriate references must be included, eg see and see also headings ; downloading authorities in real-time would be an advantage
    • Flagging of authority records for Australian authors to discriminte between these and overseas authors.
    • We chose the LC files rather than Kinetica because of perceived lack of completeness of the files and accuracy of data.
    • Since we can now download Kinetica authorities by Z39.50 we see no reason to use this service in the future.
    • Have not really investigated their use. We check our headings against the free Library of Congress authorities online.
    • The ability to easily identify Australian authorities and immediately capture them. To be able to capture them through the client when we make changes.
      The ability to get updated files of Australian headings that are new or have been worked on between last file load against holdings of our library at regular intervals, at a reasonable cost.
    • In general both inputs and outputs are not as easy as they could be.
      Name/titles are extremely difficult to input or update, and some of the owrk done on these, especially in the music area is flawed. Many libraries, like us, are not contributing as much as we should, hecne the records are not as current as they need to be.
    • Make it more reliable (i.e. clean-up obsolete and duplicate headings and authorities)
      Improve authorities by encouraging the use of 670 field for citation as in LC practice. Don't change the code for Australian names in the 008 field, position 10 from "z"(other) to "a" (AACR2) when output to file.
    • We have "kicked started" our authorities database by loading approx. 400,000 records in 2000. We have deleted some and replaced others by LC authorities.
      Although we now import most of our LC authority records from RLIN, we still see Kinetica as the preferred source but only for Australian authorites. Maybe. Quality source of records and ease of loading records onto local system.
    • The availability of individual auth. records for downloading into our III system via Z39.50 searching.
      A listing of Australian name/title authorities limited to those containing notes and references - from which we could select particular records to be included in a customized file.
    • The possible development of an 'RMR' type functionality for authority records from Kinetica to the local system
    • I'm not sure we can really comment as I don't think we are familiar enough with the service
    • The ability to download individual authority records (as exists for bibliographic records) and the ability to select a group of related authority records using Boolean search criteria.
    • Complete conformity with LCSH, apart from distinctly Austn.headings. Preferably have the latter incorporated into LCSH.
    • Try to blend with usage in the main international system, LCSH. As LCMARC was adopted in place of AUSMARC, so go full-on for conformity with the big int'l system for authorities. Many efficiencies flow from this.
  2. Government libraries
    • Higher quality records, simpler browse interface enabling fast identification of authorities and ability to mark for download in batch from hit list or browse display.
    • We have had great difficulty locating upper level authorities (i.e. Police). Those we have found have not been detailed enough for our purposes.
    • A training course on authority work complete with an interactive hands course on the Kinetica database
    • Cost. When our system is upgraded, we would like to do more authority work and it may be more effective to purchase a file.
  3. Public libraries
    • The ability to flag and download individual authorities in a similar manner to adding holdings to bibliographic records; cost would be an important factor
    • To be able to include selected authorities from the record which is being downloaded
    • XX is a fairly small public library service that has a large multicultural community. For this reason our local database does not require the full level of cataloguing for names and subject headings. If possible it would be great if there was an option on Kinetica that would enable us to retrieve records that are aimed or focused for public libraries. Ideally these authorities/headings would just include the "basic" information. This would then result in reduced staff time in editing/amending records on our local database.
      Cost.
      Ability to load updates (not the whole file)
      Ability to import complete reference structure with the heading
    • Knowledge of the service.
      Need
      Cost issues
    • If you could download them and if they were more consistant
    • Could be better
    • Ability to match to Local Authority File.
    • If Authorities could be downloaded with catalogue records (we only add our ANF holdings to Kinetica, so couldn't select holdings option for batch authorities).
      For general searching through Kinetica, if searching display & speed was improved, I would use it more as part of the cataloguing process.
    • Currently not a user friendly display & is slow, sometimes doubtful quality & inconssistencies often occur with Kinetica catalogue records downloaded. I prefer to use LC authorities for their speed, quality, extra detail, and user friendly display.
    • Accuracy of the authorities is an important issue
    • Reduced costs.
    • Our current library system is not capable of loading Kinetica authority data.
    • free of cost, like Library of Congress would be excellent incentive
  4. School/Special libraries
    • is good now - we just need to clean up our database before we can acquire it
    • Records may have been purchased prior to 2000, but I haven't been able to find any records saying so.
    • Lower cost
    • More information in the records, e.g. references, history notes. Many Kinetica authorities seem to lack these although the information often readily exists elsewhere.
    • We use a custom made subject headings list on philanthropy, which we created and maintain inhouse. We are looking to undertake a project on extending it to become a 'common language' within the philanthropy sector, but it is not based on any authorities used by Kinetica, but a mixture of local and overseas thesauri/controlled vocabulary lists.
    • Perhaps I'm having a blonde day but what is a "file from the Kinetica Authoity Record Service"?
  5. State/National libraries
    • Note: although we create authority records on Kinetica at present, they are not imported to our ILMS due to a system problem. The move to cataloguing on Voyager will involve a load of authority records from Kinetica.
    • More information in the name authorities so that a reasonable judgement can be made as to whether the name relates to the heading in our local database (e.g. 670 field - source of info., sources consulted for info.) as well as an indication whether the heading has been evaluated - both of these features are inherent in LC headings. Also the fact that LC authorities are available free of charge makes it less likely that we would consider purchasing Kinetica records when LC headings usually contain more information
    • Ability to purchase files of authority records limited to those having references or notes.
      Bibliographic change files should include records which have changes to headings (otherwise when we make corrections on NBD that affect dozens of records, each one has to be downloaded individually!)
      The ability to use either RTPS/DLL or Marc*Linc to download MARC authority records from Kinetica would be a wonderful enhancement. eg add RMR function for authority records
    • We have received excellent service from Kinetica staff each time we have purchased authority files. They have been extremely helpful and delightful to deal with.
    • Ability to purchase customised files, at least to those with references and scope notes only, if that matches our future policy. Perhaps NBD authority work could attract a discount for the cost of these files.
    • It would be convenient if the Ncould provide a one stop shop for the acquisition of authority records, so the continuation of this Service would AV28be useful.
  6. Library service providers
    • Selective downloading of records.
    • Could be better
    • Many more Australian headings for corporate bodies and series. These are our chief need from Kinetica and we are finding fewer and fewer of them Ability to request headings by this sort of category (eg "All Australian corporate bodies added in the last year" Very competitive pricing

Appendix 9: Other sources of authority data

Library Type
Academic Government Public School Special State/National Library Service Provider Total
Respondents by library type 27 9 17 1 8 6 6 74
Batch load from other sources 5 3     1 1   10
Batch Source - LC 2 2     1     5
Batch Source - OCLC 1             1
Batch Source - LTI 2             2
Batch Source - Blackwell's 1             1
Copy/Paste individual records 18 2 6   3 1 3 33
Copy/Paste Source - LC 11 2 6   3 1 3 26
Copy/Paste Source - OCLC 3         1   4
Copy/Paste Source - Classification Web 1             1
Copy/Paste Source - DRA 2             2
Copy/Paste Source - RLIN via Z39.50 8             8
Copy/Paste Source - Kinetica 5           1 6
Copy/Paste Source - Validator 5             5

Appendix 10: General comments

(Some comments have been combined, and names of libraries and library systems removed. Otherwise comments from respondents have not been edited)

  1. Academic libraries
    • I often send corrections and additional information to CPSO at LC, and it's usually accepted.
      The lack of any Australian participation in NACO is a serious gap in global authority work.
      I compile the quarterly extract of new & changed LCSH headings for the XX Association, so I'm well aware of the nature and volume of subject authority work in [subject area]. Likewise, my colleague here is chief editor of [publication] and is equally aware of such issues and of the Australian angle.
    • Authority record service is extremely important, as this improves the quality of our catalogues by allowing control of headings. As many of our bib. records have links to journal databases and other catalogues in Australia and overseas it is important to get consistency on headings. This improves quality of retrieval
    • We use Kinetica as the authority for Australian headings only.
    • The library is currently investigating the possibility of importing authority records, via z39.50, from LC on the web.
    • We also occasionally download individual authority records from the NBD.
    • Current staffing resources make it almost impossible to keep up with Authority Control in the NBD ; Authority control is done in the local catalogue, and records are changed in the NBD as our library receives a monthly bibliographic changes file - not doing authority work on the NBD would negate our efforts in correcting headings in our local catalogue ; training is another issue (it is very difficult to do authority work on name/title headings on the NBD)
    • Minimising duplication of work should be a priority. Uploading from local systems would allow this and also ensure greater timeliness. Kinetica should consider following LC for non-Australian authors.
    • This has not been a priority area for our Library. We think most users depend on Keyword searches which are not really dependent on authority control. We follow LCSH practice, so some of our headings vary from Kinetica headings.
    • My staff are really pleased that the Authority Workshop is to be held with the KAUM.
      Uncertainty has crept in with lack of practice with some cataloguers. A refresher is very welcome
    • Currently the only way in which we increase the size of the authorities database is by importing records. In 2002 5420 records were imported reported from RLIN and about 300 from Kinetica via Z39.50
    • The idea of copy authority work (in a similar way to copy cataloguing of bib. records) is an attractive idea. Local system limitations need to be addressed to make this as automated as possible.
    • We do not do this on a routine basis, but could if needed (happens only very occasionally)
    • Although authority control and authority work are considered critically important for consistency and quality of both the NBD and our local database, with staffing/funding constraints authority work cannot be regarded as a priority area of activity. There are few demonstrable links to levels of user satisfaction and catalogue functionality. Our levels of authority work are thus likely to remain minimal.
    • We are oriented to strong support for the NBD and would like to cooperate more. To do this we require facilities which avoid duplication of effort and which provide efficient loading mechanisms.
  2. Government libraries
    • We are currently testing Z39.50 real time capture of bib and authority records into our system from other library catalogues (Library of Congress). We intend to subscribe to OCLC World Cat next financial year for our authority work.
    • We have additional concerns about: 1) the number of incorrect tags and subfields; 2) problems with the "skip search" option in the Client - changing tags and subfields to agree with the first heading in the authority list.
    • We make every effort to ascertain the correct form for any entry using the "Library of Congress Authorities" website and the authorized form of headings on Kinetica Client but we do not do the authority work for new entries.
    • At present, only essential authority work is done. If it was a quicker process, our library would probably do more
  3. Public libraries
    • We have issues relating to blind authorities - subject authority in use with subdivisions only. Could cause problems with importing authority records
    • Because of the reasons above authority work is very random onto our local system - new authorities lists are printed each day but dealt with only when time permits. We do respond to reports of difficulties in searches on our catalogue & add references to assist. We have no statistics on the volume of work performed.
    • LC authority records are available online and no cost involved, are well maintained and up-to-date.
    • Sorry difficult to give you much more information than this. But I think it is important to at least advise you
    • Shortly upgrading our library software, so importing of authority records will be looked at once we find out the system's capabilities
    • Recently added feature with new library management system
  4. School/Special libraries
    • Our records were loaded onto Kinetica (batch load) in 2001. As Kinetica records with XX holdings (& generated by XX) are identified as being sub-standard and a better record for the same item exists, Kinetica staff are contacted to see if they can remove the record and 'clean up' the Kinetica database. As we get more organised in our cataloguing processes and download records from Kinetica, there will hopefully be little need for this to continue. I've inherited an author and subject authority file, but I think that's just from our own catalogue. The majority of our catalogue's records were at one stage downloaded from ABN. Some records were very ordinary but it doesn't seem that the previous librarians amended them. As our library catalogue has grown, the problem has greatly increased, hence the clean up before we commence on a new and hopefully better path. Sorry I couldn't help more.
    • This is a fantastic resource and once the ability to import authority records directly is functioning it will be even better.
    • very negative responses, but I thought it would still be worth responding, from a small special library point of view.
    • We piggy-back onto the [State Library] system so use their authorities. For new authorities we will look at Kinetica and LC or follow form and create a new entry. [State Library] however follow up and accept/reject our headings
  5. State/National libraries
    • In the interests of a high-standard database and efficient access to the NBD, the [State Library] is committed to ensuring authorities are correct. The workload associated with the maintenance of authorities on our local library system is taking greater staff resources than anticipated when the system was implemented and this has reduced our capacity to make a higher contribution to authority maintenance on Kinetica. In addition, the Kinetica pricing model does not provide the incentives for authority work as did ABN. We are currently investigating batch uploads of bibliographic records to Kinetica but have been advised that uploading of our authorities is not at present possible. [State Library] does undertake authority work on Kinetica, particularly for [State] names, however the complexities and slowness of the cataloguing client relegate this task to a low priority.
  6. Library service providers
    • Above totals include transfer of records from, and subsequent deletion of, incorrect forms of headings. This process is not very streamlined on Kinetica and would be more attractive if it were less unwieldy and time-consuming. In the past year the number of our staff has been drastically reduced, and with this and the amount of time and effort required to undertake authority work, it has tended to become a lower priority amongst our duties.
    • Most authority records are created as original records but very occasionally we will copy/paste or import name headings from LC.

Appendix 11: NBD authority activity 2002/2003

New Changed Deleted Total
National Library 511 7835 155 8501
State Library of New South Wales 250 257 23 530
State Library of Tasmania 194 186 380
Kinetica Administration 26 28 235 289
State Library of South Australia 127 92 34 253
State Library of Victoria 78 79 11 168
Joint Theological College 41 46 20 107
University of Adelaide 23 17 38 78
University of Newcastle 25 25 23 73
University of NSW 16 13 27 56
Macquarie University 17 17 5 39
Flinders University 24 1 13 38
Australian Department of Treasury 30 5 35
Victoria University of Technology 10 11 8 29
Australian Museum Research Library 10 8 1 19
University of New England 6 7 13
Edith Cowan University 3 2 8 13
University of Sydney 3 4 5 12
SA Institutes of TAFE 4 6 2 12
Murdoch University 4 3 4 11
Deakin University 3 5 1 9
Australian Institute of Aboriginal and Torres Strait Islander Studies Library 6 6
Sutherland Libraries & Information Service 3 3 6
City of Botany Bay Library Service 1 2 1 4
Supreme Court of Victoria 1 1 1 3
Swinburne University Library 1 1 2
Supreme Court of NSW, Joint Law Courts Library 1 1
Repatriation General Hospital, Daw Park, SA 1 1
State Library of Western Australia 1 1
Other (12) 187 290 479
1606 8945 615 11166

Appendix 12: Survey Instrument

SURVEY ON AUTHORITY WORK UNDERTAKEN BY AUSTRALIAN LIBRARIES

Name of library *
Type of library StateAcademicSpecialGovernmentPublicSchool*
Other type, please state
Name *
Telephone number
Email *
Does your library have a policy on authority control? Yes No *
If yes, please outline the key elements of your policy
Do you currently undertake authority work on your local library management system?
Yes No *
Name
Subject
Title
Total
Name
Subject
Title
Total
Do you currently undertake authority work on the National Bibliographic Database (NBD)? Yes No *
Name
Subject
Title
Total
Name
Subject
Title
Total
Do you currently undertake authority work on the Australian CJK database? Yes No *
Name
Subject
Title
Total
Name
Subject
Title
Total
Do you currently replicate authority work on your local system and the NBD or Australian CJK database? Yes No *
If you undertake work on your local system but not the NBD or Australian CJK database, what are the factors limiting your authority work on the NBD or Australian CJK database? (You may chose more than one reason) Lack of staff resources available Lack of staff skills in this area Priorities in the Library Costs of NBD searching Duplication of work Local system issues Local system practices e.g. ability to create "non-standard" or institution specific resources or notes relevant to users Other - please specify
Please specify
If you are not currently undertaking authority work on the NBD, what changes would need to occur to enable your library to undertake authority work on the NBD?
Do you anticipate that over the next 2-3 years your level of authority work undertaken on your local library system will increase decrease remain the same *
Comments on the reasons for any future changes in your level of authority work on your local system
Do you anticipate that over the next 2-3 years your level of authority work undertaken on the NBD or Australian CJK Database will increase decrease remain the same *
Comments on the reasons for any future changes in your level of authority work on the NBD or Australian CJK Database
Have you purchased authorities from Kinetica (the Kinetica Authority Record Service)? Yes No
Are you likely to purchase a file from the Kinetica Authority Record Service in the future Yes No
What features would make the Kinetica Authorities Record Service more attractive or relevant to your library
Any comments on the Kinetica Authority Record Service
Have you batch loaded authority records from any other source/service to your library management system (e.g. OCLC, LC) Yes No
If you answered yes, from which source
Do you copy/paste/import individual authority records from any other source/service (e.g. LC authorities on the web, RLG via Z39.50) Yes No
If you answered yes, from which source
Further comments


Footnotes

  1. Seventy-three responses were received in 2002, and seventy-four in 2003.
  2. One school library responded to the survey. This library carried out extensive authority work, including subject headings for fiction. However, this site, , being the sole respondent, cannot be considered indicative of school libraries.
  3. Statistics taken from survey responses.
  4. This figure is unreliable. It would appear to include number of headings authomatically generated by the addition of new bibliographic records, or the number of new titles added to local systems.
  5. Likewise unreliable, although the figure is inflated by large data loads in some academic sites, and the counting of new and amended records together in a number of other sites.
  6. Statistics derived from the NBD indicate that a total of 1721 authority records were added to the NBD in 2002, or 1605 from July 2002 to June 2003.
  7. These figures are derived from NBD statistics for 2002/2003
  8. "Other" consists of contributors who cannot be identified, mainly because records have been imported from other sources (eg LC, RLG), and the correct NUC of the importing library has not been included in the record.
  9. ABN Network Committee Agenda Papers, February 1989. (ABN/NC/1989/1/15)
  10. ABN Network Committee Agenda Papers, March 1993. (ABN/NC/1993/1/6)
  11. Pre-Kinetica Implementation statistics (June 1989 - Jan 1998) are taken from the Service Reports in the Network Committee Agenda Papers.
    Post- Kinetica Implementation statistics (Jan. 2000 - June 2002) are taken from NBD statistics kept locally.

    The decline in authority record growth from 1994 is possibly due to the increasing availability of local systems, which have been increasingly used over the last decade for authority work. The rise in authority activity in 1998-9 is due to database cleanup in preparation for migration from ABN. The significant decrease in 2000 is due to the unavailability of an authority input process in the early days of the Kinetica Service implementation, together with lack of training provided at the time.
  12. "Other" consists of contributors who cannot be identified, mainly because records have been imported from other sources (eg LC, RLG), and the correct NUC of the importing library has not been included in the record.

Survey on Authority Work
National Library of Australia
Kinetica Kinetica
This page is an archive page and will no longer be updated.

Survey on Authority Work

Margaret Kennedy
August 2003

Contents

Executive summary and recommendations
Background
Respondents
Summary of responses to the 2003 survey
Issues arising from the survey
Non-contribution of data to the NBD
Changing environment and resourcing issues
Input and output issues
Data quality
Conclusion
Appendices

Executive Summary

In August 2002 the final report of the Expert Advisory Group on Authority Record Delivery was tabled and discussed at the Kinetica Advisory Committee meeting. One of the ensuing recommendations of the Committee was that a follow-up survey be conducted to ascertain the level of authority work being undertaken in local library systems, but not included in the National Bibliographic Database (NBD).

This second survey was conducted in June 2003. The findings of the survey were very similar to those of the previous survey. Both clearly illustrated the significant quantity of authority work carried out on local systems only, and therefore lost to the NBD and to other Australian libraries. The survey results are also consistent with statistics derived from the Kinetica Amicus database.

The 2003 Survey results indicate that:

  • extensive authority work is carried out, especially in state and national libraries and academic libraries, and by library service providers;
  • the overwhelming proportion of authority work is carried out on local systems, the major factors being staff resources, library policies and priorities, and in the academic sector, avoiding duplication of work;
  • use of the Kinetica Cataloguing Client and lack of a batch uploading facility were also seen to be limiting factors in non-contribution of authority data to the NBD;
  • currency of data in the NBD is seen to be a problem;
  • many libraries source authority data from elsewhere, mainly from the Library of Congress (LC); and
  • some responses indicated a level of confusion about what authority data is, and failed to distinguish between the number of titles added to their databases, or the number of headings automatically generated in their local systems when new bibliographic records were added to the system, and the number of authority records added to the system.

The decline in contribution of authority data to the NBD is of considerable concern. In the year 2002/2003 NBD statistics indicate that only 1605 new authority headings were added to the NBD. Of these 1160 (or over 72%) were added by the National Library and four state libraries, and 180 (or 11%) by a small number of academic libraries. A comparison of these figures with the estimated number of new authority headings created on local systems, 2,933,784 (sourced from survey responses; not to be considered reliable - see last dot-point above), is indicative of the extent of the problem.

This Report recommends that a number of possible solutions be examined to improve the level of contributions. Such an examination should include a cost-benefit analysis of each option in providing a high quality and relevant service to Kinetica customers. Options may include the following:

  • focusing on Australian authority data. Non-Australian authority data is readily available from other sources (including Library of Congress Subject Headings and Name Authorities via KineticaWeb from RLG);
  • entering into an agreement with the National and state libraries (who contribute more than 90% of authority data to the NBD) to facilitate the acquisition of national or state-specific data from these libraries for the NBD.
  • launching a publicity and promotion campaign to educate Australian libraries about the benefits of an Australian authority file, Kinetica authority record standards and the types of services Kinetica can offer. Promote existing Kinetica support for authority data - searching via Client, KineticaWeb and Z39.50; downloading via Z39.50 or email/ftp to a local file;
  • offering a range of products more tailored to the provision of Australian authority data to Kinetica customers (eg only records with references or notes);
  • considering the development of an alternative means of online input of authority records (eg Web update);
  • considering the development of an unlinked authority file which supports batch loading and matching of authority records.

Some of the options detailed above are considered problematic. The option, of providing an alternative means of authority data input, would be of limited value. As stated in the report, the reluctance to use the Kinetica Cataloguing Client is not considered to be the primary factor in the decision to use local systems for authority work. The major factors are identified in responses as resourcing and workflow issues and library priorities.

The option of batch loading of authority data is problematic as it is much more complex than batch loading bibliographic data. Authority records have very few reliable match points. If the Authority file is to be linked to the NBD bibliographic file then accurate matching also requires a comparison of the bibliographic record(s) linked to the database heading (or authority) and those linked to the incoming authority records, and of course the incoming authority records will not be accompanied by bibliographic records. This scenario may also require matching of an incoming authority record with only a heading (not an authority record) on the NBD. If batch loading of authority data from Kinetica customer libraries is to be attempted then an unlinked authority file may be the only practical option. Irrespective of the approach taken, batch loading of authority data could impact on the degree of duplication in the NBD, and would result in a significant increase in resources required for database cleanup and manual review.

A combination of options is recommended. Our major contributors of authority data are the national and state libraries. Working closely with the these libraries and acknowledging the contribution they make to the NBD, even to the extent of providing monetary incentives for the contribution of authority data, would reward those libraries who currently contribute in excess of 90% of authority data to the NBD, and would encourage those who do not currently contribute. The contributions of these libraries would provide a unique, comprehensive, reliable and up to date source of Australian authority data for libraries, and costs could be passed on with the availability of a range of Australian authority products that would be of relevance to them. This is a market that is not adequately served by other sources of authority data. Meanwhile, we can concentrate on encouraging contribution of bibliographic and holdings data from all Kinetica customers, both online and via Batch*Link, in the interests of resource sharing.

Background

In 2002 the Kinetica Expert Advisory Group on Authority Record Delivery (EAGAD) conducted a survey of Kinetica users and reported on a range of issues relating to authorities and the National Bibliographic Database (NBD). The final report of the group is available online at http://www.nla.gov.au/librariesaustralia/eag_ad.html. One of the EAGAD recommendations was "That a more extensive survey be conducted to identify the volume of authority records created on local systems, and to determine the extent of the market for NBD authority records".

The 2003 survey has been carried out to gather information on the extent of authority work being undertaken in Australian libraries on their local systems. The results of the survey provide a valuable base for the National Library and Kinetica Advisory Committee to consider the scope and range of issues relating to authority work that needs to be addressed in the future. This Report outlines the results of the survey.

Respondents

Seventy-four respondents completed the 2003 survey. Of these, twenty-seven responses were received from academic libraries, nine from government libraries, seventeen from public libraries, one from a school library, eight from special libraries, six from state and national libraries, and six library service providers.

Respondents by Library Type
Figure 1: Respondents by Library Type

Some respondents did not comfortably fit into any of the categories listed on the form; these have been separated into the additional category of Library Service Providers.

Summary of responses to the 2003 Survey

  • There is a strong correlation between the results of the 2002 and 2003 surveys. Both drew responses from more than seventy Kinetica users (1) , representing a range of library types, and both surveys highlighted similar issues regarding the contribution of authority data to the NBD, and the use of the NBD as a source of authority data.
  • Most of the respondents from State/National libraries (five of six) and Library Service Providers have authority policies (five of five). Five of seventeen public libraries, eight of twenty-seven academic libraries, five of seventeen public libraries and three of eight special and school libraries also had policies (2). Only one government library respondent (of nine) stated that the library had an authority policy. Some respondents stated that they realised the need for an authority policy, and were developing the policy at present.
  • There was clearly a lack of understanding of what authority control entails. Consequently there were inconsistencies in responses to a number of questions relating to authority control. Only thirty-seven of the seventy-four respondents gave a positive response to the question of whether they carried out authority control. However, sixty respondents claimed to carry out authority control on their local system, and fifteen claimed to carry out authority control on the NBD. From some of the comments on authority control, it appears that some respondents considered that the automatic generation of headings when new records were created on or downloaded to the local system constituted authority control.
  • Library service providers would appear to carry out the most rigorous level of authority work on behalf of their clients. Authority work in academic libraries is thorough and far-reaching, with many of the respondents checking all new headings and carrying out necessary authority work on them. In addition, on-going maintenance work is carried out to resolve conflicts. While authority work in academic libraries is considered desirable, even essential, several respondents cite a lack of resources as the reason for limiting their authority work to conflict resolution only.
  • The situation is similar in state/national libraries. In addition, there appears to be little authority work carried out on headings created as a result of copy cataloguing, because of lack of professional staff.
  • The most significant factors identified by respondents that limit their authority work to the local systems are:
    • staff resources
    • library priorities
    • avoiding duplication of work (academic sector)
    • lack of staff skills (Public sector ).
    • many other factors, including cataloguing client issues, lack of uploading and downloading facilities, and the need to use local or specialist thesauri. (These are detailed in Appendices 3 and 4)
  • Most respondents cite their authority sources as Library of Congress (LC) Subject Headings and Library of Congress Name Authorities (from LC, OCLC, or Validator), and Kinetica for Australian headings. There are some local practices and thesauri in place, especially in public libraries (eg local headings for local history), special libraries school libraries (fiction headings) and library service providers (SCIS).
  • Some respondents were unaware of the facility to acquire authority data from the NBD via Z39.50 or via ftp/email from Stored records in KineticaWeb.
  • Most respondents see themselves carrying out the same amount of authority work on their local systems or the NBD/CJK systems in the future as they do now.
  • Thirty-one respondents predict that the level of authority work on their local system will increase, and two predicted that their level of authority work on the NBD/CJK systems would decrease.
  • Only a small proportion of respondents (eleven) have purchased authority records from Kinetica / ABN previously, and the same number state that they may purchase these files in future. Ten sites (half of them academic libraries) have purchased authority files from other sources, mainly from the Library of Congress
  • A much greater proportion (thirty-three sites) have used Z39.50 or copied and pasted individual records from various sources. Once again the predominant source is Library of Congress.

Issues arising from the survey

The issues relating to authority work can be broken down into the following categories:

  • Non-contribution of data to the NBD, and the loss of consequent efficiencies in the ability to share and reuse data;
  • The changing environment, including local system issues, other sources of data, changes in libraries;
  • Input and output issues, including client issues, downloading and/or uploading facilities, and products;
  • Data quality; including adherence to standards, currency of data, Library of Congress versus Kinetica headings, and the importance of an up-to-date and accurate Australian authority file, especially for Australian names.

Non-contribution of data to the NBD

From the survey responses it would appear that academic libraries and library service providers are the most prolific sectors when it comes to the creation and editing of authority data on their local systems. This is not surprising, given the number of respondents from academic libraries, and the fact that several have added large files of authority data (from LC) to their local systems, and given that library service providers undertake work on behalf of a considerable number of other libraries. Data from public and special libraries would appear to be unreliable; comments indicate that some sites actually counted the number of titles acquired and catalogued onto their local system as constituting new title authority records. Most of these sites did not count name, subject and title authorities separately.

The more telling factor is the comparison of the statistics (3) r the creation of authority records on local systems as compared to those for the creation of authority records on the NBD, with 2,933,784 new authority records added to local systems (4) over the last year, compared with 3,647 added to the NBD. The statistics for changes to authority records are also disparate, with 45,0941 (5) on local systems, and 2,400 on the NBD. A summary of responses can be found in Appendix 2. It should be noted that there is a discrepancy between the figures from the survey and those extracted from the NBD (6) . Whichever figures are considered, there is demonstrably a significant amount of authority control happening in local systems, which is lost to the NBD, and to other Kinetica users.

Authority Work on Local System, NBD and CJK
Figure 2: Authority Work on Local System, NBD and CJK

Changing environment and resourcing issues

The survey responses highlight a number of factors behind the decision of many libraries to carry out authority work on their local system, instead of on the NBD. These factors include:

  • the improved functionality of local systems to support authority data;
  • the pressure on staff resources in libraries;
  • the need to be more efficient and integrate work processes;
  • the need to avoid duplication of effort by doing work on different systems; and
  • use of other authority data sources being perceived to be more cost-effective for libraries than having scarce staff resources creating authorities on-line.

Factors limiting authority work to local system
Figure 3: Factors limiting authority work to local system

The result is that a significant proportion of authority data is not shared with the Australian library community, because it resides on local systems, instead of in the NBD, where it could be used by all. While carrying out cataloguing and all associated work on local systems may be seen by many to be the most efficient way of ensuring that the data is available on the local system quickly, the benefits of resource sharing, including the availability of sharable authority data, would appear to have a lower priority. Most sites appear to appreciate the benefits of sharing bibliographic data; contribution of this data, especially via batch load, continues to grow. This growth is important to ensure that adequate bibliographic and locations data is available to underpin interlibrary loan and other resource sharing activity. For authority data, however, both the number of contributors, and the level of contribution of authority data continue to shrink, with only approximately 40 libraries contributing or editing authority data in 2002-2003.

Input and output issues

The output of Kinetica authority records to local systems is not a problem. Records can be copied from Kinetica using Z39.50, or via the ftp or mailing of saved records from KineticaWeb. Files of records can be requested via the Kinetica Authority Records Service. Some changes have been suggested to both the output of authority records (eg an RMR facility like that for MARC bibliographic records, which triggers downloading of the record via TPS*DLL or MARCLink). Some suggestions have been forthcoming on the Kinetica Authority Record Service, and the need to be more flexible in providing files that meet the requirements of users (eg Australian name headings only, only records with notes and references). The Kinetica Authority Records Service already provides files of Australian authority headings only, including Name authorities only. It is timely to both advertise the products we already offer, and review the range of existing products to determine what use is currently made of the service, and whether the availablility of alternative products would increase the use of the service.

While many libraries have opted to contribute bibliographic and holdings data via Batch*Link, this option is not available for authority data. The Kinetica Cataloguing Client is the only means of contributing authority data to the NBD, and people are reluctant to use this interface. As stated above, the use of local systems in preference to the Cataloguing Client is very much a workflow efficiency measure in libraries, and a proportion of the libraries currently using their local system for authority work have stated their willingness to add this data to the NBD, should a means of batch loading become available.

Batch loading of authority data is a complex issue. Bibliographic records have a number of match points, some more reliable than others; yet batch loading of bibliographic data via Batch Link has resulted in an increase of duplicate records on the NBD, and a significant increase in the number of records requiring manual review. Authority data has very little in the way of reliable match points. If the data has been downloaded from the NBD, edited, and then uploaded, it will have an Amicus number, but new records from local systems will not have an Amicus number. Incoming authority records will be matched against headings, not authority records. Minor differences in punctuation or coding (eg a period or space in the wrong place, or an incorrect subfield code), could result in a duplicate record in the Authorities database. Accurate matching of incoming authority data would also require comparison of the bibliographic record(s) linked to the database heading or authority, and those linked to the incoming authority records, and there will be no incoming linked bibliographic records linked to the authority records being loaded. If batch loading of authority data is considered, the incoming data will require to be matched against an Australian authority file that is not linked to NBD bibliographic records and headings. This means the creation of a "stand alone" authority file.

However, if the NBD is to stay relevant as a reliable source of Australian authority data, some investigation of alternative input options should take place. In addition, Kinetica needs to stress the importance of resource-sharing, not just in an "it's the right thing to do" framework, but in the context that contribution of high quality data enriches the NBD, making it more up to date and relevant, and making authority data easily accessible to all (and subsequently more cost effective than the current situation, ie each library duplicating the work being done by others, each on their own local system.

Data quality

A number of respondents questioned the quality and currency of authority records on the NBD. Criticism of the currency of data is probably well-founded, given the small number of libraries that contribute to and maintain authority data on the NBD for others to use. In the words of one survey respondent:

"Many libraries, like us, are not contributing as much as we should, hence the records are not as current as they need to be."

Criticism of data quality, on the other hand, is not so well founded. Some respondents to the 2003 survey remarked on the need for Kinetica to "follow LC" and "follow LCSH practice". Some responses also refer to inconsistencies within headings in bibliographic records on the NBD.

Kinetica does follow LC practice for authority headings. Kinetica practice is described on the Kinetica Website at http://www.nla.gov.au/librariesaustralia/standards.html . Variations to "LC practice" occurs in the case of Australian Name and Subject headings, where an Australian extension of LC headings is maintained. In the case of subject headings, every effort is made to minimise conflict with LC headings. As stated on the Kinetica Website:

The future development of the Australian extension to LCSH should take account of the changing environment of Australian cataloguing. Australian libraries are obtaining their cataloguing data from a variety of sources (Kinetica, overseas databases, library supplier databases) and they need to contain the cost of maintaining the authority files in their local systems. Accordingly, the development of the Australian extension to LCSH should follow these guidelines:
  1. Where a term already exists in LCSH, an equivalent Australian term may be adopted as a reference to the LCSH term. Australian terminology should not be substituted for synonymous LCSH terms.
  2. Australian subject headings which have previously been substituted for existing LCSH terms, should be reviewed in order to reduce the variations from LCSH. The review should take into account the cost of converting back to standard LCSH, which may outweigh the savings of being consistent with LCSH.
  3. The Australian subject headings which have already been adopted, should be continuously monitored for opportunities to reduce the variation from LCSH.

In the case of Name headings, the following statement applies: Kinetica will follow the form of heading established in the Library of Congress Name Authorities (LCNA). LCNA will take precedence over name, title and series headings which have already been verified as authorities in ABN or Kinetica, and these headings may be retrospectively changed to conform with LCNA.
An exception will be made for Australian name, title and series headings, because authority records for Australian names, titles and series will usually be established earlier and more accurately by Australian libraries. Therefore, authorities for Australian names, titles and series, which conform to AACR and the Library of Congress rule interpretations, should not be changed retrospectively simply to conform with LCNA. In order to improve the quality of Australian name, title and series headings in LCNA, the National Library of Australia will offer to the Library of Congress a file of Australian name authorities. To enable Australian headings to be identified within Kinetica, they should be coded as follows: USMARC field 008, character position 10, code 'z' This is a valid USMARC code signifying 'other descriptive cataloguing rules', but will be converted to 'c' on output from Kinetica, as the headings will conform with AACR2.

Data Contribution

The following table details contributors of authority data to the NBD between July 2002 and June 2003. By far the greatest contributors are the National and State Libraries, with 88% of authority record additions, changes and deletions being carried out by the National Library, State Library of New South Wales, State Library of Tasmania and the State Library of Victoria. The next largest group of contributors is the academic library sector, with 4.4% of additions, changes and deletes contributed by thirteen universities and the South Australian TAFE network (7) .

Library Type
National/State Academic Government/Special Public Kinetica Administration Other (8) Total
New 1160 180 48 4 26 187 1605
New % 72.3 11.2 3 0.2 1.6 11.7 100
Changed 8450 158 15 5 28 290 8946
Changed % 94.5 1.8         96.3
Deleted 224 154 2 1 235 0 616
Deleted % 36.4 25     38.1   99.5
Total 9834 492 65 10 289 477 11167
Total % 88 4.4     2.6   95

Figure 4: Contribution to authority work on NBD by library type

The contribution of authority data has been an ongoing issue, both for Kinetica and for its predecessor, the Australian Bibliographic Network (ABN). A report written by Warwick Cathro and tabled at the ABN Network in February 1989 forshadowed that, with the increasing availability of local library systems, by 1995:

"…software support for cataloguing transactions would be decentralised onto local systems. ….It will be quite difficult to maintain a national authority file in a decentralised manner. Authority control work involves the resolution of conflicting records, the merging of records and the updating of records, more than it involves the creation of new records. It would be much easier to manage this by reference to a centralised authority file. I envisage a small group of major libraries, perhaps 20 to 30 in all participating in this process…" (9)

This prediction would appear to have been correct. An examination of the growth of authority data from 1990 to the present shows a steady decline in contribution of authority data, with the decline being more significant from 1994.

It is also worth noting that during the early 1990s a considerable amount of authority work was outsourced to a number of contractors, particularly authority work on government names (National, State and local government) (10)

Growth in authority records
Figure 5: Growth in authority records

The decline in authority work on the NBD (11) cannot therefore simply be attributed to a reluctance to use the Kinetica Cataloguing Client, which only came into use in 1999. In ABN days, authority work was cumbersome, with transactions passing through a "Nitely" process to load it to the database. When complex authority work was in progress (eg merging duplicate authority records), it could take many days, waiting for Nitely to process each change before the next step could be undertaken. This process was much more cumbersome and labour-intensive than using the Kinetica Cataloguing Client.

It would appear that the decline in authority work on the NBD is, to a much greater extent, due to other factors outlined above. Decreasing resources, lack of skills, the availability of authority data from other sources, and shrinking library budgets make it appear to be more cost-effective, at face value, to buy in whatever data is available, and to either carry out other authority work on the system, or to avoid it altogether

Conclusion

While the 2003 survey has failed to deliver reliable and consistent results in some areas, it can still be taken as being indicative of the high proportion of authority work being undertaken in local systems as opposed to the NBD. The continuing downward trend constitutes a significant risk to the long tradition of resource sharing in the Australian library environment, as far as authority data is concerned, and casts doubt on the credibility of the NBD authority files as the definitive source of Australian authority data.

There are some major issues to be considered in determining the future of the NBD as the definitive source of reliable and up-to-date Australian authority data. The first of these is whether NBD authority data is still relevant to, and valued by the Australian library community, and indeed valued as a component of an international authority file, such as the Virtual International Authority File, or the Anglo-American Authority File. The use of Kinetica as a source of Australian authority data by most survey respondents would appear to indicate that the NBD is still valued.

If the NBD authority file is still seen to be relevant and valued, there are two further issues; the first being to examine what can be done to encourage Australian libraries to contribute to the NBD authority files, and the second to determine what can be done to acquire and load the wealth of Australian authority data currently residing only on local systems.

This Report assumes that the NBD authority file is still valued, and recommends the consideration of a range of factors that may motivate Australian libraries to become active partners in resource sharing of Australian authority data by contributing their authority data to the NBD. Consideration should be given to alternative methods of enabling this contribution to occur, as well as the possibility of some form of recognition and/or incentive for the contribution of data. Since the overwhelming proportion of authority data is contributed to the NBD by the National and some state libraries, it would appear logical to target this group of libraries in considering encouraging or rewarding contributions.

If the quality of the NBD is improved by increased contributions of Australian authority data, especially by the state libraries currently not contributing, the NBD will provide a unique and valued source of data to onsell to Australian libraries. The costs of any incentives provided to contributing libraries could be offset by the vigorous marketing of a revised range of products to Australian libraries, which would be relevant to their needs (as outlined in the 2003 survey responses), and could form part of the proposed Australian National Bibliography service.

In addition, the marketing campaign should be aimed at convincing libraries of the value of the NBD as an immensely important shared asset, and fostering a sense of ownership and responsibility for the NBD as a whole, and not just as a unique repository of Australian bibliographic and holdings data.


Appendices

  1. Respondents by Library Type
  2. Amount of authority control - local systems and NBD/CJK
  3. Limiting factors in carrying out authority control only on local systems
  4. Limiting factors - Summary of comments
  5. Future intentions for authority work
  6. Future intentions - Summary of comments
  7. Use of Kinetica Authority Records Service
  8. Kinetica Authority Records Service - Summary of comments
  9. Other sources of authority data
  10. General comments
  11. NBD authority activity 2002/2003
  12. Survey Instrument

Appendix 1: Respondents by Library Type

Library Type Number of respondents
Academic 27
Government 9
Public 17
School 1
Special 8
State/National 6
Library Service Provider 6
Total 74

Appendix 2: Amount of authority control - local systems and NBD/CJK

Library Type
Academic Government Public School Special State/National Library Service Provider Total
Number of respondents by library type 27 9 17 1 8 6 6 74
Authority control on local system 24 4 16 1 6 4 5 60
ILMS New Total 2847783 2227 25681 3300 160 13275 41358 2933784
ILMS Amend Total 254324 2830 3422   2552 172465 15348 450941
Authority control on NBD 8 1 1 0 0 4 1 15
NBD New Total 437         3210 50 3647
NBD Amended Total 1958 23 10     299 110 2400
Number who carry out authority control on CJK 1 0 0 0 0 0   1
Authority Work on ILMS and NBD/CJK 6 2 2 1   2   13

Appendix 3: Limiting factors in carrying out authority control on local systems only

Library Type
Academic Government Public School Special State/National Library Service Provider
Respondents by library type 27 9 17 1 8 6 6
Authority control on local system 24 4 16 1 6 4 5
Limiting factors-
Staff resources
15 3 13   4 2 2
Limiting factors-
Staff skills
5   10   2 1  
Limiting factors-
Library priorities
14 2 12 1 4 3 2
Limiting factors-
NBD search costs
7 3 4   2 2 1
Limiting factors-
Duplication of work
15   3   1 1 3
Limiting factors-
Local system issues
3   4        
Limiting factors-
Local system practices
3   6 1 3 3 2
Limiting factors-
Other
13 1 2 1 2 2 3
Other -
Too expensive to use contract cataloguer
  1          
Other-
No way to contribute without Cat. Client
        1    
Other -
No facility to merge auth. Records both ways
          1  
Other-
Have not considered. Plan to upload bib & holdings
        1    

Appendix 4: Limiting factors - Summary of comments

(Some comments have been combined, and names of libraries and library systems removed. Otherwise comments from respondents have not been edited)

  1. Academic libraries
    • Cataloguing Client: awkward; slow response times, limited number of licenses; difficulty of working with name-title headings; difficulty of resolving variants and duplicate forms (a number of responses) Local system: ability to import authorities will mean extra cost and generate duplicate headings (use and maintenance issue) We concentrate on headings where we have specialist information and/or information to resolve conflicts etc.
    • As part of a larger system, we follow their policy and practice. If we were to do authority work, it would be within our local system. Unlikely to do it because of lack of staff resources, priorities within the library, and duplication of work.
    • Authorities created on Kinetica cannot be downloaded.
    • NBD is not used for cataloguing so there's no reason to use it for authority work, also our local system offers considerable advantages for cataloguing and authority work.
    • We follow US NAF for non-Australian headings and RLG Z39.50 is so fast; could not justify time to duplicate this work onthe NBD.
    • Not a contributing member for records, just for holdings. Authority work usually involves changing subject headings listed in the Cataloging service bulletin.
    • Some work is done on the NBD, especially for Australian headings and conflicts or incorrect headings that we have imported with bibs. Most work is done by importing LC or Kinetica authority records
    • Where work is done for non Australian headings, it is mostly done on the local system only. Incorrect authorities are reported to LC for non Australian headings. we do not use non standard headings or local notes. All is done to standards and worth sharing. It would seem more efficient to do our own authority work (several responses)
    • In the case of the CJK database, our staff who have the language knowledge to work in the system have no authority work training or experience.
    • Inability to conveniently upload to the NBD the work done in the local system (using headings from Validator).
    • We do not follow Kinetic Authorities
    • We have always batch loaded our data to the NBD and there has never been a facility to merge authority records both ways. We have never worked online with either ABN or Kinetica. Our adherence to LC standards has precluded our contribution to some extent.
  2. Government libraries
    • Cleanups of old authority data on system are sometimes needed to maintain consistency of local system.
    • Issues as identified plus too expensive to use contract cataloguer PLUS No financial incentive for the Library.
  3. Public libraries
    • We only duplicate authority work on the NBD for authorities related to our Local Studies Collection. Staff costs/time prohibit any other authority work on the NBD
    • Slowness of Client software. Numerous steps involved to make changes. Software is not user-friendly.
    • Do not contribute cataloguing to the NBD, download records from Kinetica only.
    • we contribute our holdings, but not cataloguing records. Eg, if we purchase a title not held in NBD, we note it and search again later for someone else's record of the item.
  4. School/Special libraries
    • We are a private institution so do not wish to have our holdings listed. We have many resources dating pre WW1 or have only copies in Australia. These would be requested for inter-loan. We do not have the staff to do this. Reciprocal rights would be of limited benefit to us.
    • We do not use the Cataloguing Client so have no way of performing authority work on the NBD
    • Speed of doing work on our local system as opposed to doing the same work on the NBD, i.e. time taken to create an average authority record.
  5. State/National libraries
    • The slowness of the connection. Correcting one authority would take about 10-15 minutes.
    • We do undertake some authority work on the NBD, but it is limited by the above factors.
    • As we plan to contribute our bibliographic and holdings data via BatchLink (or some other utility), we have not considered contributing authority records onto the NBD.
  6. Library service providers
    • We have always batch loaded our data to the NBD and there has never been a facility to merge authority records both ways. We have never worked online with either ABN or Kinetica. Our adherence to LC standards has precluded our contribution to some extent.
    • Time required to carry out Kinetica input: this is not the same issue as the costs of searching but it works out as a cost issue also. It is not the staff time so much as the system response time which makes this out of the question. Time/cost of obtaining Kinetica authority records
    • [Site] add authority records to its own database which is specifically designed for the needs of school libraries.

Appendix 5: Future intentions for authority work

Library Type
Academic Government Public School Special State/National Library Service Provider Total
Number of respondents by library type 27 9 17 1 8 6 6 74
Authority control on local system 24 4 16 1 6 4 5 60
Authority control on NBD 8 1 1 0 0 4 1 15
Number who carry out authority control on CJK 1 0 0 0 0 0   1
Local system -
Anticipate decrease
  1 1   1   1 4
Local system -
Anticipate increase
8 4 9   3 6 1 31
Local system -
Anticipate same
19 4 7 1 5   4 40
NBD/CJK -
Anticipate decrease
  1       1   2
NBD/CJK -
Anticipate increase
3   2   1 3   9
NBD/CJK -
Anticipate same
24 8 15 1 7 2 6 63

Appendix 6: Future intentions - Summary of comments

(Some comments have been combined, and names of libraries and library systems removed. Otherwise comments from respondents have not been edited)

  1. Academic libraries
    • We see local authority work as taking precedence over NBD, and cataloguers' time is limited
    • Feedback from clients and reader services when there is conflict or inconsistent name headings.
    • Current library automation system does not allow for downloading of NBD authorities. Due to staffing constraints minimal work is done on local database. We rely on the correct format for headings on the Kinetica records.
    • Following the EAGAD report, University of XX considered the implications of the report for the University library and made recommendations for improvements to the delivery of records. Investigations into the feasibility of these recommendations are currently taking place.
    • This question is almost impossible to answer and is dependent on technical developments and resources available In third quarter of 2002 we loaded the full retrospective file of Library of Congress Subject Headings to our local system. We have made decisions to use LC headings and not Kinetica variations. Our resourcing priorities must be to our own system.
    • New local system is allowing us to identify authority record gaps more easily, and reports with new system are better for following up problems. This combined with retrospective effort on serial records, and training/permission now given to copy cataloguers to bring in standard authority records where they exist should mean that additional authority records are utilised as part of normal workflows. N/A. We believe we do sufficient for our purposes
    • With implementation of new functionality in ILS, the ease of loading authority files should enable an increase in authority work
    • We want to "clean up" our name headings, some of which are abbreviated and inadequate, having migrated from an earlier inventory file·
    • Improvement to local ILMS
    • Our current practices are perhaps not ideal but they do work and we are not anticipating any major changes in the foreseeable future
  2. Government libraries
    • Unlikely that any staff losses will be replaced and authority work would be of low priority with reduced staff.
    • We have only begun creating authorities within the past year and aim to have our entire database under authority control within two years. This is a major project for our Cataloguing team.
    • Training in authority work, and on the Kinetica database
    • I hope to be able allocate staff time to this job.
    • An upgrade to our system will make authority work easier. Currently, local system issues inhibit our doing authority work.
  3. Public libraries
    • The implementation of a new library management system will facilitate authority work.
    • Hopefully staff time will be available to raise the priority.
    • The authority work on our local database has not been done for years, needs to be cleaned up.
    • Upgrade of local system allowing easier use of authority control programs. Change to windows based system requiring more authority work.
    • Delivery of catalogue over the Web
    • Access to global change facility in new system. Currently this is planned for the next year.
    • Current project to clean up authority files prior to and following a system upgrade. Names complete, started on subjects. Following this, general maintenance only expected as part of normal cataloguing task.
    • Cataloguing backlog has gone - more time to do authority work.
    • New Library management system and changed work practices.
    • Still cleaning up the database after merging 3 libraries catalogues 7 years ago.
    • need to serve public well, and knowledge that current local authorities are in a parlous state
  4. School/Special libraries
    • URL cataloguing
    • We're in the process of converting our catalogue from XX systemto YY system. Many (many) errors have been found and are being cleaned up as part of this (slow) process.·
    • We implemented a new LMS 15 months ago and it is only now that the system is properly bedded down that we have the time available to devote to tasks such as authority work. During 2003 this work was a lower priority than other tasks hence the relatively small number of records created. We have already increased this considerably so far this year and would expect that to continue from now on.·
    • We have done a number of retrospective conversions over the past two years and much authority work has resulted from these. I expect the number of new authorities to decrease once this work is completed.
  5. State/National libraries
    • Possible outsourcing of some technical functions freeing up staff to undertake other duties.
    • If the XX transfers its cataloguing operations to [new ILMS], the performance of authority control is likely to also be transferred to [new ILMS].
    • Difficult to assess for a number of reasons. In the last financial year a lot of editing has been due to the migration and data clean-up. To the end of 2001 we loaded weekly LCSH tapes and ran reports. To date this has not been possible with the new system. When we are able to do we anticipate another peak in authority work. Another peak is temporary increased funding for library materials this financial year and the next 2 financial years due to a large extension to the building and increased funding for additional library materials
    • We are currently setting up the authority record formats we require on our local system, after which we will commence creating authority records inhouse as required, as a test phase. Later, we will probably load authority files.
  6. Library service providers
    • Batch contribution would be a major factor that would affect our level of contribution to the NBD.

Appendix 7: Use of the Kinetica Authority Records Service

Library Type
Academic Government Public School Special State/National Library Service Provider Total
Number of respondents by library type 27 9 17 1 8 6 6 74
Purchased Kinetica authorities previously 6 1 2     2   11
Likely to purchase Kinetica authorities 5   3     3   11

Appendix 8: Kinetica Authority Records Service - Summary of comments

(Some comments have been combined, and names of libraries and library systems removed. Otherwise comments from respondents have not been edited)

  1. Academic libraries
    • Easy download
      Free searches in support of authority work (incl. to select correct authorities
    • Keeping up with LCNA + LCSH is impossible.
      If *only* Australian entities had authorities on Kinetica, and they were constructed to NACO standards (names, titles) or LCSH (subjects), we could rely on them and it would be worth making the effort. While standard, maintenance and content of Kinetica authority records falls short of these standards, the NBD can't be taken seriously as a source of Australian authorities.
    • Would use the Kinetica authority record service if it could be easily downloaded into our local system at reasonable cost.
    • Purchasing the authority records is a Management issue.
    • A ready made record would make it easier and less time consuming for us.
      For new Australian subject headings there is discussion group where by consensus is sought on particular headings. This cannot be done by individual libraries. A central system would be better.
    • If Kinetica authorities were maintained and kept up to date.
      We did purchase authority files from ABN which formed the basis of the authorities in our catalogue, and now we maintain them individually.
    • Direct downloading of headings with see refs into a local system. An accurate file. Kinetica needs a massive cleanup and all duplicate and incorrect headings should be removed.
    • Full compliance with LC name and subject authority files
    • Ability to downline load individual authority records from Kinetica as required.
    • Cost of authority tapes is expensive, in addition there is the cost of setting up a separate load table. When authorities are obtained in these large batches, if any problems occur with the load, the volume of manual corrections can be overwhelming
    • We have purchased records in the past we had concerns about their currency and we found some headings purchased as Australian were not in fact Australian. This resulted in considerable unnecessary work fo us.
    • Appropriate references must be included, eg see and see also headings ; downloading authorities in real-time would be an advantage
    • Flagging of authority records for Australian authors to discriminte between these and overseas authors.
    • We chose the LC files rather than Kinetica because of perceived lack of completeness of the files and accuracy of data.
    • Since we can now download Kinetica authorities by Z39.50 we see no reason to use this service in the future.
    • Have not really investigated their use. We check our headings against the free Library of Congress authorities online.
    • The ability to easily identify Australian authorities and immediately capture them. To be able to capture them through the client when we make changes.
      The ability to get updated files of Australian headings that are new or have been worked on between last file load against holdings of our library at regular intervals, at a reasonable cost.
    • In general both inputs and outputs are not as easy as they could be.
      Name/titles are extremely difficult to input or update, and some of the owrk done on these, especially in the music area is flawed. Many libraries, like us, are not contributing as much as we should, hecne the records are not as current as they need to be.
    • Make it more reliable (i.e. clean-up obsolete and duplicate headings and authorities)
      Improve authorities by encouraging the use of 670 field for citation as in LC practice. Don't change the code for Australian names in the 008 field, position 10 from "z"(other) to "a" (AACR2) when output to file.
    • We have "kicked started" our authorities database by loading approx. 400,000 records in 2000. We have deleted some and replaced others by LC authorities.
      Although we now import most of our LC authority records from RLIN, we still see Kinetica as the preferred source but only for Australian authorites. Maybe. Quality source of records and ease of loading records onto local system.
    • The availability of individual auth. records for downloading into our III system via Z39.50 searching.
      A listing of Australian name/title authorities limited to those containing notes and references - from which we could select particular records to be included in a customized file.
    • The possible development of an 'RMR' type functionality for authority records from Kinetica to the local system
    • I'm not sure we can really comment as I don't think we are familiar enough with the service
    • The ability to download individual authority records (as exists for bibliographic records) and the ability to select a group of related authority records using Boolean search criteria.
    • Complete conformity with LCSH, apart from distinctly Austn.headings. Preferably have the latter incorporated into LCSH.
    • Try to blend with usage in the main international system, LCSH. As LCMARC was adopted in place of AUSMARC, so go full-on for conformity with the big int'l system for authorities. Many efficiencies flow from this.
  2. Government libraries
    • Higher quality records, simpler browse interface enabling fast identification of authorities and ability to mark for download in batch from hit list or browse display.
    • We have had great difficulty locating upper level authorities (i.e. Police). Those we have found have not been detailed enough for our purposes.
    • A training course on authority work complete with an interactive hands course on the Kinetica database
    • Cost. When our system is upgraded, we would like to do more authority work and it may be more effective to purchase a file.
  3. Public libraries
    • The ability to flag and download individual authorities in a similar manner to adding holdings to bibliographic records; cost would be an important factor
    • To be able to include selected authorities from the record which is being downloaded
    • XX is a fairly small public library service that has a large multicultural community. For this reason our local database does not require the full level of cataloguing for names and subject headings. If possible it would be great if there was an option on Kinetica that would enable us to retrieve records that are aimed or focused for public libraries. Ideally these authorities/headings would just include the "basic" information. This would then result in reduced staff time in editing/amending records on our local database.
      Cost.
      Ability to load updates (not the whole file)
      Ability to import complete reference structure with the heading
    • Knowledge of the service.
      Need
      Cost issues
    • If you could download them and if they were more consistant
    • Could be better
    • Ability to match to Local Authority File.
    • If Authorities could be downloaded with catalogue records (we only add our ANF holdings to Kinetica, so couldn't select holdings option for batch authorities).
      For general searching through Kinetica, if searching display & speed was improved, I would use it more as part of the cataloguing process.
    • Currently not a user friendly display & is slow, sometimes doubtful quality & inconssistencies often occur with Kinetica catalogue records downloaded. I prefer to use LC authorities for their speed, quality, extra detail, and user friendly display.
    • Accuracy of the authorities is an important issue
    • Reduced costs.
    • Our current library system is not capable of loading Kinetica authority data.
    • free of cost, like Library of Congress would be excellent incentive
  4. School/Special libraries
    • is good now - we just need to clean up our database before we can acquire it
    • Records may have been purchased prior to 2000, but I haven't been able to find any records saying so.
    • Lower cost
    • More information in the records, e.g. references, history notes. Many Kinetica authorities seem to lack these although the information often readily exists elsewhere.
    • We use a custom made subject headings list on philanthropy, which we created and maintain inhouse. We are looking to undertake a project on extending it to become a 'common language' within the philanthropy sector, but it is not based on any authorities used by Kinetica, but a mixture of local and overseas thesauri/controlled vocabulary lists.
    • Perhaps I'm having a blonde day but what is a "file from the Kinetica Authoity Record Service"?
  5. State/National libraries
    • Note: although we create authority records on Kinetica at present, they are not imported to our ILMS due to a system problem. The move to cataloguing on Voyager will involve a load of authority records from Kinetica.
    • More information in the name authorities so that a reasonable judgement can be made as to whether the name relates to the heading in our local database (e.g. 670 field - source of info., sources consulted for info.) as well as an indication whether the heading has been evaluated - both of these features are inherent in LC headings. Also the fact that LC authorities are available free of charge makes it less likely that we would consider purchasing Kinetica records when LC headings usually contain more information
    • Ability to purchase files of authority records limited to those having references or notes.
      Bibliographic change files should include records which have changes to headings (otherwise when we make corrections on NBD that affect dozens of records, each one has to be downloaded individually!)
      The ability to use either RTPS/DLL or Marc*Linc to download MARC authority records from Kinetica would be a wonderful enhancement. eg add RMR function for authority records
    • We have received excellent service from Kinetica staff each time we have purchased authority files. They have been extremely helpful and delightful to deal with.
    • Ability to purchase customised files, at least to those with references and scope notes only, if that matches our future policy. Perhaps NBD authority work could attract a discount for the cost of these files.
    • It would be convenient if the Ncould provide a one stop shop for the acquisition of authority records, so the continuation of this Service would AV28be useful.
  6. Library service providers
    • Selective downloading of records.
    • Could be better
    • Many more Australian headings for corporate bodies and series. These are our chief need from Kinetica and we are finding fewer and fewer of them Ability to request headings by this sort of category (eg "All Australian corporate bodies added in the last year" Very competitive pricing

Appendix 9: Other sources of authority data

Library Type
Academic Government Public School Special State/National Library Service Provider Total
Respondents by library type 27 9 17 1 8 6 6 74
Batch load from other sources 5 3     1 1   10
Batch Source - LC 2 2     1     5
Batch Source - OCLC 1             1
Batch Source - LTI 2             2
Batch Source - Blackwell's 1             1
Copy/Paste individual records 18 2 6   3 1 3 33
Copy/Paste Source - LC 11 2 6   3 1 3 26
Copy/Paste Source - OCLC 3         1   4
Copy/Paste Source - Classification Web 1             1
Copy/Paste Source - DRA 2             2
Copy/Paste Source - RLIN via Z39.50 8             8
Copy/Paste Source - Kinetica 5           1 6
Copy/Paste Source - Validator 5             5

Appendix 10: General comments

(Some comments have been combined, and names of libraries and library systems removed. Otherwise comments from respondents have not been edited)

  1. Academic libraries
    • I often send corrections and additional information to CPSO at LC, and it's usually accepted.
      The lack of any Australian participation in NACO is a serious gap in global authority work.
      I compile the quarterly extract of new & changed LCSH headings for the XX Association, so I'm well aware of the nature and volume of subject authority work in [subject area]. Likewise, my colleague here is chief editor of [publication] and is equally aware of such issues and of the Australian angle.
    • Authority record service is extremely important, as this improves the quality of our catalogues by allowing control of headings. As many of our bib. records have links to journal databases and other catalogues in Australia and overseas it is important to get consistency on headings. This improves quality of retrieval
    • We use Kinetica as the authority for Australian headings only.
    • The library is currently investigating the possibility of importing authority records, via z39.50, from LC on the web.
    • We also occasionally download individual authority records from the NBD.
    • Current staffing resources make it almost impossible to keep up with Authority Control in the NBD ; Authority control is done in the local catalogue, and records are changed in the NBD as our library receives a monthly bibliographic changes file - not doing authority work on the NBD would negate our efforts in correcting headings in our local catalogue ; training is another issue (it is very difficult to do authority work on name/title headings on the NBD)
    • Minimising duplication of work should be a priority. Uploading from local systems would allow this and also ensure greater timeliness. Kinetica should consider following LC for non-Australian authors.
    • This has not been a priority area for our Library. We think most users depend on Keyword searches which are not really dependent on authority control. We follow LCSH practice, so some of our headings vary from Kinetica headings.
    • My staff are really pleased that the Authority Workshop is to be held with the KAUM.
      Uncertainty has crept in with lack of practice with some cataloguers. A refresher is very welcome
    • Currently the only way in which we increase the size of the authorities database is by importing records. In 2002 5420 records were imported reported from RLIN and about 300 from Kinetica via Z39.50
    • The idea of copy authority work (in a similar way to copy cataloguing of bib. records) is an attractive idea. Local system limitations need to be addressed to make this as automated as possible.
    • We do not do this on a routine basis, but could if needed (happens only very occasionally)
    • Although authority control and authority work are considered critically important for consistency and quality of both the NBD and our local database, with staffing/funding constraints authority work cannot be regarded as a priority area of activity. There are few demonstrable links to levels of user satisfaction and catalogue functionality. Our levels of authority work are thus likely to remain minimal.
    • We are oriented to strong support for the NBD and would like to cooperate more. To do this we require facilities which avoid duplication of effort and which provide efficient loading mechanisms.
  2. Government libraries
    • We are currently testing Z39.50 real time capture of bib and authority records into our system from other library catalogues (Library of Congress). We intend to subscribe to OCLC World Cat next financial year for our authority work.
    • We have additional concerns about: 1) the number of incorrect tags and subfields; 2) problems with the "skip search" option in the Client - changing tags and subfields to agree with the first heading in the authority list.
    • We make every effort to ascertain the correct form for any entry using the "Library of Congress Authorities" website and the authorized form of headings on Kinetica Client but we do not do the authority work for new entries.
    • At present, only essential authority work is done. If it was a quicker process, our library would probably do more
  3. Public libraries
    • We have issues relating to blind authorities - subject authority in use with subdivisions only. Could cause problems with importing authority records
    • Because of the reasons above authority work is very random onto our local system - new authorities lists are printed each day but dealt with only when time permits. We do respond to reports of difficulties in searches on our catalogue & add references to assist. We have no statistics on the volume of work performed.
    • LC authority records are available online and no cost involved, are well maintained and up-to-date.
    • Sorry difficult to give you much more information than this. But I think it is important to at least advise you
    • Shortly upgrading our library software, so importing of authority records will be looked at once we find out the system's capabilities
    • Recently added feature with new library management system
  4. School/Special libraries
    • Our records were loaded onto Kinetica (batch load) in 2001. As Kinetica records with XX holdings (& generated by XX) are identified as being sub-standard and a better record for the same item exists, Kinetica staff are contacted to see if they can remove the record and 'clean up' the Kinetica database. As we get more organised in our cataloguing processes and download records from Kinetica, there will hopefully be little need for this to continue. I've inherited an author and subject authority file, but I think that's just from our own catalogue. The majority of our catalogue's records were at one stage downloaded from ABN. Some records were very ordinary but it doesn't seem that the previous librarians amended them. As our library catalogue has grown, the problem has greatly increased, hence the clean up before we commence on a new and hopefully better path. Sorry I couldn't help more.
    • This is a fantastic resource and once the ability to import authority records directly is functioning it will be even better.
    • very negative responses, but I thought it would still be worth responding, from a small special library point of view.
    • We piggy-back onto the [State Library] system so use their authorities. For new authorities we will look at Kinetica and LC or follow form and create a new entry. [State Library] however follow up and accept/reject our headings
  5. State/National libraries
    • In the interests of a high-standard database and efficient access to the NBD, the [State Library] is committed to ensuring authorities are correct. The workload associated with the maintenance of authorities on our local library system is taking greater staff resources than anticipated when the system was implemented and this has reduced our capacity to make a higher contribution to authority maintenance on Kinetica. In addition, the Kinetica pricing model does not provide the incentives for authority work as did ABN. We are currently investigating batch uploads of bibliographic records to Kinetica but have been advised that uploading of our authorities is not at present possible. [State Library] does undertake authority work on Kinetica, particularly for [State] names, however the complexities and slowness of the cataloguing client relegate this task to a low priority.
  6. Library service providers
    • Above totals include transfer of records from, and subsequent deletion of, incorrect forms of headings. This process is not very streamlined on Kinetica and would be more attractive if it were less unwieldy and time-consuming. In the past year the number of our staff has been drastically reduced, and with this and the amount of time and effort required to undertake authority work, it has tended to become a lower priority amongst our duties.
    • Most authority records are created as original records but very occasionally we will copy/paste or import name headings from LC.

Appendix 11: NBD authority activity 2002/2003

New Changed Deleted Total
National Library 511 7835 155 8501
State Library of New South Wales 250 257 23 530
State Library of Tasmania 194 186 380
Kinetica Administration 26 28 235 289
State Library of South Australia 127 92 34 253
State Library of Victoria 78 79 11 168
Joint Theological College 41 46 20 107
University of Adelaide 23 17 38 78
University of Newcastle 25 25 23 73
University of NSW 16 13 27 56
Macquarie University 17 17 5 39
Flinders University 24 1 13 38
Australian Department of Treasury 30 5 35
Victoria University of Technology 10 11 8 29
Australian Museum Research Library 10 8 1 19
University of New England 6 7 13
Edith Cowan University 3 2 8 13
University of Sydney 3 4 5 12
SA Institutes of TAFE 4 6 2 12
Murdoch University 4 3 4 11
Deakin University 3 5 1 9
Australian Institute of Aboriginal and Torres Strait Islander Studies Library 6 6
Sutherland Libraries & Information Service 3 3 6
City of Botany Bay Library Service 1 2 1 4
Supreme Court of Victoria 1 1 1 3
Swinburne University Library 1 1 2
Supreme Court of NSW, Joint Law Courts Library 1 1
Repatriation General Hospital, Daw Park, SA 1 1
State Library of Western Australia 1 1
Other (12) 187 290 479
1606 8945 615 11166

Appendix 12: Survey Instrument

SURVEY ON AUTHORITY WORK UNDERTAKEN BY AUSTRALIAN LIBRARIES

Name of library *
Type of library StateAcademicSpecialGovernmentPublicSchool*
Other type, please state
Name *
Telephone number
Email *
Does your library have a policy on authority control? Yes No *
If yes, please outline the key elements of your policy
Do you currently undertake authority work on your local library management system?
Yes No *
Name
Subject
Title
Total
Name
Subject
Title
Total
Do you currently undertake authority work on the National Bibliographic Database (NBD)? Yes No *
Name
Subject
Title
Total
Name
Subject
Title
Total
Do you currently undertake authority work on the Australian CJK database? Yes No *
Name
Subject
Title
Total
Name
Subject
Title
Total
Do you currently replicate authority work on your local system and the NBD or Australian CJK database? Yes No *
If you undertake work on your local system but not the NBD or Australian CJK database, what are the factors limiting your authority work on the NBD or Australian CJK database? (You may chose more than one reason) Lack of staff resources available Lack of staff skills in this area Priorities in the Library Costs of NBD searching Duplication of work Local system issues Local system practices e.g. ability to create "non-standard" or institution specific resources or notes relevant to users Other - please specify
Please specify
If you are not currently undertaking authority work on the NBD, what changes would need to occur to enable your library to undertake authority work on the NBD?
Do you anticipate that over the next 2-3 years your level of authority work undertaken on your local library system will increase decrease remain the same *
Comments on the reasons for any future changes in your level of authority work on your local system
Do you anticipate that over the next 2-3 years your level of authority work undertaken on the NBD or Australian CJK Database will increase decrease remain the same *
Comments on the reasons for any future changes in your level of authority work on the NBD or Australian CJK Database
Have you purchased authorities from Kinetica (the Kinetica Authority Record Service)? Yes No
Are you likely to purchase a file from the Kinetica Authority Record Service in the future Yes No
What features would make the Kinetica Authorities Record Service more attractive or relevant to your library
Any comments on the Kinetica Authority Record Service
Have you batch loaded authority records from any other source/service to your library management system (e.g. OCLC, LC) Yes No
If you answered yes, from which source
Do you copy/paste/import individual authority records from any other source/service (e.g. LC authorities on the web, RLG via Z39.50) Yes No
If you answered yes, from which source
Further comments


Footnotes

  1. Seventy-three responses were received in 2002, and seventy-four in 2003.
  2. One school library responded to the survey. This library carried out extensive authority work, including subject headings for fiction. However, this site, , being the sole respondent, cannot be considered indicative of school libraries.
  3. Statistics taken from survey responses.
  4. This figure is unreliable. It would appear to include number of headings authomatically generated by the addition of new bibliographic records, or the number of new titles added to local systems.
  5. Likewise unreliable, although the figure is inflated by large data loads in some academic sites, and the counting of new and amended records together in a number of other sites.
  6. Statistics derived from the NBD indicate that a total of 1721 authority records were added to the NBD in 2002, or 1605 from July 2002 to June 2003.
  7. These figures are derived from NBD statistics for 2002/2003
  8. "Other" consists of contributors who cannot be identified, mainly because records have been imported from other sources (eg LC, RLG), and the correct NUC of the importing library has not been included in the record.
  9. ABN Network Committee Agenda Papers, February 1989. (ABN/NC/1989/1/15)
  10. ABN Network Committee Agenda Papers, March 1993. (ABN/NC/1993/1/6)
  11. Pre-Kinetica Implementation statistics (June 1989 - Jan 1998) are taken from the Service Reports in the Network Committee Agenda Papers.
    Post- Kinetica Implementation statistics (Jan. 2000 - June 2002) are taken from NBD statistics kept locally.

    The decline in authority record growth from 1994 is possibly due to the increasing availability of local systems, which have been increasingly used over the last decade for authority work. The rise in authority activity in 1998-9 is due to database cleanup in preparation for migration from ABN. The significant decrease in 2000 is due to the unavailability of an authority input process in the early days of the Kinetica Service implementation, together with lack of training provided at the time.
  12. "Other" consists of contributors who cannot be identified, mainly because records have been imported from other sources (eg LC, RLG), and the correct NUC of the importing library has not been included in the record.