Introduction

Digital Libraries (DLs) are the focus of professional movements in the medical, legal, and scholarly communities. In this document, I discuss an ongoing study which contributes to DL research by examining some new DL design concepts within the legal domain. DLs are being used increasingly nationwide in the legal profession for research, computer-integrated court reporting, and for trial presentations of multimedia documents and evidence photos.

Court practices in the 1970's and early 1980's involved court reporters typing highly codified notes on a stenographer machine and then translating them later into text which a lawyer could interpret. In the mid-1980's, computer translations of this codified text became available and court reporters have embraced this technology to more quickly provide translated text to attorneys and judges. Recently, court reporting technology is enabling court reporters to type in specialized codes during court proceedings and to quickly produce highly intelligible language for real-time display in the courtroom, and for later access as a text file. This transition has provided a more powerful medium and has altered the way court reporters and legal professionals do their work.

Recently, this high technology in the courtroom has been widely televised in the US during the O.J. Simpson trial (Newton and Ford, April 22, 1995). There, crucial evidence has been interactively presented with computer technology, while attorneys and Judge Ito are able to review court transcripts in real-time. However, glitches can readily occur with the use of this technology as evidenced during the Simpson trial when the jurors were recently interviewed by Judge Ito. In response to juror complaints of problems with the bailiffs and among themselves, Judge Ito conducted personal, private interviews with only the court reporter present. While using the computerized translation machine during the course of one juror's interview, the court reporter failed to block the text feed to the media. Consequently, what was intended as personal and discreet became public knowledge of this juror's displeasure. Although only a few lines of text leaked to the press before the problem was fixed, the privacy of a courtroom transaction should be safeguarded.

This is but one example of how computer systems can be designed for improved usability but weaken the ability of an organization to function effectively. "Organizational usability" is a new concept that refers to the extent to which computer systems can be effectively integrated into the work practices of diverse members of specific organizations without undermining the vitality of the organization. While the Human-Computer-Interaction (HCI) field has made progress in improving interface usability, organizational usability is less well understood. Issues of organizational design have been identified and faced by researchers and practitioners in neighboring fields, such as Information Systems and Computer Supported Cooperative Work. This research addresses organizational usability in the design of DLs.

I am conducting field work to study the design and use of research DLs in the legal profession. Lawyers and judges are constantly researching statutes and appellate decisions to guide them in their daily work. Increasingly, members of the legal profession are purchasing online services or CD-ROMs to augment the use of research books with DLs. The objective of this ongoing study is to understand what aspects of organizational usability facilitate and detract from effective DL usage in the courts. This research also addresses the institutional influences which affect DL usage in the legal profession. To date, this qualitative study has consisted of a series of approximately 20 semi-structured interviews with attorneys in a civil defense firm, district attorneys, public defenders, judges and various court personnel; studying of relevant documents; and participant observation.

Six dimensions of organizational usability are being used to assess the DL technology used by judges and attorneys in the California Courts. These dimensions characterize the factors necessary to design systems which "fit" into particular work situations.

1. Accessibility -- Ease with which people can locate specific DL systems, gain physical access and electronic access to their electronic corpuses. This dimension refers to both physical proximity and administrative/social restrictions on using specific systems.
2. Compatibility -- Level of compatibility of file transfers between DLs and other systems where people want to use text and pictures to store, edit and manipulate print.
3. Integrability into work practices -- How smoothly the system fits into a person's or group's work practices, including access to complementary computing resources (i.e., communication lines, scanners, printers) to facilitate work-flows -- how people actually prefer to work.
4. Social-organizational expertise -- The extent to which people can obtain training and consulting to learn to use DL systems in combination with existing computer systems and can find help with problems in usage.
5. Reliability -- Level of reliability of DL systems. Systems should provide easy recovery from common problems (with the possible assistance of support staff).
6. Affordability -- Extent of economical feasibility when purchasing DL systems. Systems should be affordable within the domain of the purchasing organization including training, usage fees for online searches, support staff costs, and maintenance fees.

Technology Studied

The judges whom I interviewed have access from their chambers to Lexis/Nexis, an online legal research service from a personal computer (PC). The district attorneys are authorized to dial into Westlaw, an online legal research service, from a PC in the law library. In the same location, the district attorneys have access to two CD-ROM PC work stations containing LawDesk, a CD- ROM legal research corpus, and Shepard's {1} California Citations and United States Supreme Court citations on CD-ROM. Out-of-state citations on CD-ROM are maintained at their Appellate Division's library. The public defenders and civil lawyers also have access to LawDesk. and Shepard's Citations on CD- ROM. The online research services provide the following legal information: California case (or statute) legal research, California and United States citations, federal case law research, Witkin {2} treatises, out-of-state legal research, newspaper (or other news) research, and corporate (Dun and Bradstreet) research. The LawDesk {3} CD-ROMs contain the California cases and statutes (full text), the United States Supreme Court Reports, Lawyers Ed. 2d, and the Witkin treatises (full text).

Discussion

The most compelling result found during my field work has been the need for user training that goes beyond a vendor's brief tutorial upon installation and an 800 "help" line. Even brief in-house training classes haven't proven successful in achieving widespread usage. Although in-house training sessions are available to both district attorneys and public defenders, most never take these classes. These busy professionals do not want to take the time out of their daily work to drive the two hour round-trip to the training center and attend the two hour training sessions. For district attorneys, a librarian offers personal training sessions but finds it difficult to switch people from using books to using CD-ROMs for legal research. My observation is that training which involves direct involvement in an attorney's work flow seems to work best.

For example, in the civil defense law firm, attorney A, who is most familiar with LawDesk and Shepard's Citations , gives ongoing training and support to other attorneys who want to use the systems. While all attorneys attended a LawDesk vendor's one hour tutorial upon installation, only attorney A (with previous computer experience) was able to use the system afterwards and consequently trained the others. No attorneys in this law firm have ever read the user's manuals. If questions arise, attorney A calls an 800 number to obtain help.

Another area of organizational usability which presented stark evidence of its import in DL design or configuration is the "integrability into work practices". Judge A does not like Lexis/Nexis very much. He especially finds the use of Lexis/Nexis impractical for the "shepardizing" procedure where many volumes of books are used to look up current decisions which may have altered the validity of cases and statutes. Recently, the court administrators made the decision to eliminate the purchasing of the Shepard's Citations books for each judge, requiring them to use Lexis/Nexis for all current "shepardizing". This creates problems for the judge because if he wants to read something from a Shepard's Citations book, he likes to take it to the bench and he generally knows exactly which code he is referencing. It takes more time and is inconvenient for him to excuse himself in the middle of a trial, go to his chambers, dial into Lexis/Nexis , print out a citation, and take it back to the bench to be read.

In contrast, the civil attorneys praise the Shepard's CD-ROM as highly valuable to their research in both expediency and quality. Shepardizing from the books is laborious and time-consuming to them. Since they have ready access to their CD-ROM during inhouse research, it becomes an integral part of their work practices. This is an example of how a computer system can be more usable in one organization versus another depending on smooth integrability into work flow.

An overall striking finding has been that acceptability of information technology plays an important role in achieving organizational change when altering work practices with DLs. This research shows that personal beliefs and preferences, personal inflexibility to modify work patterns, and preferences for alternative computer systems can have a powerful impact on widespread acceptability of computer innovations. Future research will investigate reasons for resistance to computer usage in the legal profession and how to best achieve DL dissemination.

My interest in attending this forum is to share with the DL research community research results on the organizational usability characteristics of DL systems for legal research. I believe that my results can provide insights into elements of organizational usability that are universal to the socio-technical implications of technological implementations. By attending, I will also gain valuable feedback from the DL community regarding this research. I hope to become acquainted with DL researchers working on related problems such that we may synergistically advance our research.

Notes

1. Shepard's Citations are legal references to up-to-date rulings on cases and statutes.
2. Witkin treatises are summaries of California law.
3. The civil attorneys ' configuration of LawDesk only includes the full text of California cases.

References

Elliott, M. S. and Kling, R. Organizational Usability of Digital Libraries in the Courts. To appear in Proceedings of Hawaii International Conference on System Sciences (HICSS), Maui, Hawaii, January 3-6, 1996.

Newton, J and Ford, A. Transfer of deputies provokes boycott by 13 Simpson jurors. Los Angeles Times 114 (Sat., April 22, 1995), A1, col. 1, 56 col. in.