Message-Id: <199708180141.UAA18403@library.wustl.edu> Date: Mon, 18 Aug 1997 11:48:26 +1000 From: Tony Barry <mailto:tony@NINGAUI.ANU.EDU.AU> Subject: Document delivery from the library catalogue To: Multiple recipients of list WEBCAT-L <mailto:WEBCAT-L@WUVMD.WUSTL.EDU>
Apologies for cross posting but this may be of interest to a number of lists.The web enabled library catalogue raises a wide range of new opportunities. One of them is ability to enhance online access to the content of journals by linking from a catalogue entry to a document delivery services for that journal. Naturally you would do that for titles that the library did not own rather than those that it did. This essay explores the issues involved in this idea and how this might be done.
We are used to the situation where a library either owns material or the material is obtained from another library. In the case of journals this is in the form of photocopies. The advent of the network has made some changes in this approach. For instance by the use of the original Ariel system to use tftp, then ftp for transmission and now mime based email. New opportunities are offered by services which give article level delivery via web access and a variety of means of delivery (mainly http). Such services are Uncover, the Blackwell's Navigator service and Ebscohost and those from individual publishers such as Academic press.
So far none of these services have impinged greatly on what is being done with the catalogue. With the arrival web based OPACS this is changing. Tentative first steps have been made to add URL links in the OPAC to full text versions of journals via the 856 MARC tag. Users of the catalogue can then use a web browser to search , find the title, and then link to the full text. In this way the delivery of the text is integrated with the searching for the title. While the library then does no need to hold a hard copy version of the title concerned we have see that for many publishers access to the electronic version is conditional upon a print subscription. This can only be a transition solution. At some stage all publishers will need to look much harder at solely electronic access to their serial publications.
OPAC links to document delivery services
It is possible to go further than this. A link from the record could be made not to the publisher, but to a document delivery services. This would be particularly valuable if the link was to a full table of contents service. The use of the catalogue could then link from the catalogue entry to a menu of the table of contents of the journal on the suppliers web server. There the required article could be located and transferred either direct to the browser, via email or through some other means. In this way a library would be able to provide an access service to journals that it did not hold or held a subscription.
There is the vexed question of payment. The user my pay or the library may have arranged some form of subsidy. Uncover already provides a service where a library can subsidise access for users from their domain so precedents exist for such a service. The ideal would be for a link to a table of contents service for the journal as described above. Alternatevely a searching mechanism for papers published in the journal would be a possibility .
This can only be done if the service offers a stable URL to each title. Unfortunately existing services all appear to offer HTTP POST based services where the search arguments are not encoded in the URL or state maintaining transaction variables, which change each session, are used. A change in the way document delivery services are currently offered over the web might be needed. Ideally you would be able to access such a service via a URL of the form -
http://document_delivery_service.com/ISSN/123456789
where "123456789" would be the ISSN of the title.
Is the library catalogue only about holdings? =============================================
A philosophical objection might be raised by some about whether it is appropriate to use the library catalogue to describe material not held by the library. The resolution to this is held in deciding whether the purpose of the catalogue is to deliver information or to describe information artifacts held by a library. An allied question relates to the use of the 856 MARC field. This is intended to indicate the location of an electronic document. The way I have described its use, it is being addressed as a mechanism one step removed from the location of the document. Definitional purists might feel that additional tags are required in the MARC record to allow for such alternate usage.
How could such a service could be offered? ==========================================
It is all very well to postulate such a service but the labour of adding serial catalogue records is a significant cost to cataloguing operations. Is there an easier way?
A possible approach is described below.
* The library would supply the vendor with a list of their current serial holdings
* The vendor would match this against the titles they offer and supply the library with a list of titles the library did not hold.
* The library would then select the titles which interested them indicating, if appropriate, differential subsidy options. That is, whether they wish to subsidies access to some titles for users from their institution or all of them.
* The vendor would then supply MARC records for the titles chosen with URLs to their service in the 856 field which the library would load into its web enabled OPAC.
The end result is that the catalogue would contain records for additional serial titles of interest to the library's patrons to which it did not to subscribe. There may be no subscription either because of the staff costs of checkin, binding, claiming etc or due to high subscription price or both. These entries in the catalogue would then provide direct user access to the contents of the journals.
The end users get easier access and hopefully cheaper access to titles which the library could not provide other than through laborious inter-library loans services
An updated version of this paper will be maintained at - http://ningaui.anu.edu.au/e-says/docdel.opac.shtml
Tony Barry Mon, 18 Aug 1997
_____________________________________________________________ mailto:mailto:tony@ningaui.anu.edu.au |+61 6 249 5688 http://purl.oclc.org/NET/Tony.Barry |+61 6 288 0959
Ningaui Pty Ltd, GPO Box 1680, Canberra City, ACT 2601
Visiting Fellow, Department of Computer Science, Faculty of Engineering and Information Technology. Australian National University, ACT 0200 AUSTRALIA
Apologise for the stolen generation - <http://apology.west.net.au/>