The New ABC: Teaching Computers
How to Read Legal Documents

Allison Stanfield
eLaw Australia Pty Ltd
astanfield@elawaust.com.au

Delivered at the 2nd AustLII Conference on Computerisation of Law via the Internet , Australasian Legal Information Institute ( AustLII ), University of Technology, Sydney, Australia, 21-23 July 1999.

1. Introduction

Technology should be used to make a lawyer's practice easier, not more difficult. From a barrister's or a judge's perspective, paper is an easy medium - it is easy to 'flick' through and thereby to skim read. However, typically, the life of a law clerk and solicitor is made more difficult because information has to be presented, particularly in court, in such a way that it is easy to locate and easy to read, especially once a matter goes to trial. Information must be sorted and indexed and has to be easily locatable and retrievable.

In the litigation support business, a lot of time is spent taking hard copy materials and converting them into a format which a computer can understand. In turn, this helps lawyers sort, index and present material in an easy-to-locate and presentable format more quickly and more efficiently.

Is it possible to make these tasks much easier? Yes, it is and there is one simple solution: that documents used by lawyers should be created in a format that can be understood by computers.

Some may say that most documents these days are created on computers - otherwise, why do law firms spend millions on their legal precedent banks every year? This is true, but once completed, they are not in a format that the computer can understand, that is, the computer cannot identify the salient features of a document such as the date of the document, the author, what type of document it is, who it is addressed to and so on. The reader can ascertain this information, but all the computer sees is text.

Imagine a world where all documents were produced with their key bibliographical information already 'marked'. For example, a letter written six months ago could easily be found by simply typing in a search on 'author' = 'Smith' and 'date' = 'January 1999'. Alternatively, a brief could be easily generated by selecting those electronic documents that are to be included in the brief and 'zipping' them up electronically and emailing them to counsel, or simply by printing them out, if required.

So, how can a document be created in a way that it is legible to a computer? There are a number of ways in which this can be done and if set up properly in the first place can take no longer than typing the document in the normal way.

2. Why Capture Information in a Computer-Friendly way?

2.1 The Courts say so

It is becoming more common that documents are exchanged in electronic format during discovery. Two Australian courts have so far issued Practice Notes dealing with the use of information technology during civil litigation:

• Practice Note No. 105 made on 15 March 1999 by the Chief Justice of New South Wales (the ' NSW Practice Note '); and
   

• Practice Note No. 3 made on 29 April 1999 by the Chief Justice of Victoria (the ' Victorian Practice Note ').

Those Practice Notes encourage parties involved in civil litigation to:

• use electronic data (or databases) to create lists of their discoverable documents;
   

• undertake discovery by exchanging electronic data (or databases) created in accordance with an agreed protocol;
   

• exchange electronic versions of documents such as pleadings and statements;
   

• arrange for inspection of discovered material by way of images if appropriate; and
   

• the Victorian Practice Note also encourages parties to consider the use of information technology during trial.

It is understood that the Federal Court may soon issue a similar Practice Note.

This is great news! However, before exchange can be done in electronic format, all that information in hard copy needs to be converted into electronic format. This is a laborious process, and the information probably exists somewhere in electronic format anyway. Although utilising information electronically does save practitioners' time, particularly during trial, there is an enormous amount of work that needs to be done before one can even get to that point!

2.2 The World's Going Online

The Internet is here to stay. The legal profession, however, does not use the Internet as much as other areas of business. This is due largely to the fact that the law itself does not yet handle electronic information well. Some court rules still provide for information to be filed and stored in hard copy and things like 'personal signatures' are still found in practice directions and rules on dusty shelves. A contract for land is still required to be in writing and 'signed by the parties to be charged'. This is so notwithstanding that many Titles Offices store Certificates of Title electronically. Digital signature legislation is being worked upon that will overcome this, but it may yet be a while before it is implemented.

How many law firms actually use the Internet? Research in Australia is difficult to come by, however, in the USA, the American Bar Association (ABA) has conducted surveys since 1988 to provide a comprehensive 'snapshot' of technology use by small law firms across the USA. In 1998, the ABA also conducted a survey of technology used within large law firms.

Almost 75% of lawyers in private practice work in sole practitioner and small firm practices. The ABA's surveys indicate that in 1996, 38% of small firm survey respondents reported using the Internet. In 1998, this usage increased to 80%. In addition, 37% of small law firms without Internet access today reported that they plan to provide it in the coming year.

The ABA also surveys the 500 largest law firms in the USA. It is believed that because these firms are much larger than most of the country's law firms, trends established in these firms can influence the technology adopted by other legal professionals.

Of those surveyed, 58% of large law firms had web pages. Of those who do not have web pages, 17% were in the process of developing a website.

3. How to make the Change

The change is not a technical one, rather, it is a cultural one. Wired magazine's, Kaitlin Quistgaard recently investigated the technology giant's plan to assault the legal profession ('Order in the Court' by Kaitlin Quistgaard, Wired, February 1999). The technology involved is not complicated. She said the real challenge lies in changing a culture which has evolved over centuries:

4. How to Capture Information in a Computer-Friendly way

For information to be exchanged online, it is important that it be captured, and exchanged, in a consistent format.

4.1 Word Processors?

A lot of information is captured consistently now - usually in one of two formats - Word or WordPerfect. WordPerfect is still the preferred word processor in a legal office because of its feature which allows the user to 'Reveal Codes', that is, to see what codes have been put into the document so the user can understand why the document is doing something strange.

Also, many law firms have vast precedent banks that allow legal documents to be produced in a consistent way time and time again. This is great from a reader's perspective - they see information on the same place on a page each time a similar document is produced. However, this does very little for the computer. Unless the computer is 'told' what is where on a page, it cannot understand the document. With a little work at the beginning, much work can be saved in the long run to ensure that electronic documents can be read by a computer.

Formatting is the biggest problem with documents created by a word processor. As soon as the user wants to convert it to another format, such as HTML for publication on the Internet, or to use in a litigation support tool, automatic numbering and other formatting is lost, thereby creating a lot more work for someone.

Do lawyers need a word processor? The majority of sophisticated tools now available on a word processor would not be used by the majority of legal offices. Most law firms require functions such as bolding, underlining, tables and numbering. Those with precedent banks require macros. Today, this functionality is available in not only word processing packages but also HTML editing tools. Using an HTML editor, the HTML code can be viewed and with very little knowledge of HTML, can understand what the document is doing.

To allow documents to be 'read' by a computer, the user needs to ensure salient information is appropriately marked in the document. A word processor can do this, but realistically, it will be difficult to convince lawyers that they now have to ensure information is electronically tagged. For example, creating a stylesheet in MS Word can mark up the headings in a document and these can later be gathered to create a Table of Contents. In the same way, a stylesheet can be used to mark up the salient parts of a document such as date, author and so on.

However, how many lawyers are going to be bothered doing all of this, particularly when the existing system works just fine? Also, they've probably spent a fortune just getting all of their precedents right, so why start over again? There must be a simple way to do this and one which does not impose an extra layer of work on lawyers.

4.2 Structured Mark-up Languages

Tools do exist which allow information within documents to be marked up. HTML is an example of this. When a web page is downloaded and the user sees different font sizes, tables, paragraph numbers etc, this is all possible because the HTML code has certain tags in it to tell the web browser what to do with the information. Likewise, some search engines allow searches to be made on certain fields of information. Here, the user will have a selection of fields to search from a drop-down menu. This enables searching to be refined.

Another mark-up languages is SGML (Standardised General Markup Language). SGML is another language which allows tags to be inserted beside pertinent information in a document. Indeed, HTML is a subset of SGML. Some legal publishers already use SGML to markup judgments and legislation in a consistent way and to make searching on their databases easier - searching by field such as judge, date, party and so on, is much easier than trying to conduct a full text search across an ocean of caselaw.

A benefit of using SGML is that it can be converted to HTML as required.

4.3 The New Frontier: XML

So, is HTML or SGML the best to use? It is suggested that neither is quite right, and instead, it is better to keep an eye on a third mark-up language being developed which is known as XML - eXtensible Mark-up Language.

Why not HTML? Because HTML 'comes bound with a set of semantics and does not provide arbitrary structure'.

Why not SGML? Because 'SGML provides arbitrary structure, but is too difficult to implement just for a web browser. Full SGML systems solve large, complex problems that justify their expense. Viewing structured documents sent over the web rarely carries such justification.' (Norman Walsh, What is XML?).

It is also better to store documents in a language such as SGML/XML and then to convert to Word, PDF, Excel and HTML as needed. Search indexes can also be generated, and more fancy navigational aids such as topic maps can also be generated automatically.

5. What is XML?

Applying this analogy to the legal profession, XML can allow the computer to tell the difference between a Writ, a Statement of Claim, a judgment or a piece of legislation. Further, XML will allow the computer to know what information is on the Writ such as the party who filed it, the date it was filed and the court in which it was filed.

An organisation called World Wide Web Consortium (W3C) is working on developing XML which will allow business communities to prescribe how their documents should be marked up. XML will allow a Document Type Definition (DTD) to be developed for each type of document. This is the way, a computer will be able to know the difference between a writ of summons, an Affidavit of Documents, a witness statement, correspondence and so on. Also, the computer will be able to tell what the document is about. For example, with a letter, the computer can tell who it is from, who it is addressed to, the date of the document and what it is about.

Stylesheets can also be created to work with DTDs to allow documents to be formatted as organizations wish. This means that the information in a judgment in Queensland can be 'tagged' in the same way as a judgment in New South Wales. However, the style in which a judgment is printed may vary between states. By using eXtensible Stylesheet Language (XSL), style can be applied to make a document look a particular way.

Both Netscape and MS Internet Explorer are being developed to support XML.

6. Why use XML?

Lawyers use information from documents over and over again. Let's take the humble Writ of Summons. Certain pieces of information which are created on the document, may be used in many ways. Such information can include the names of parties, the date of filing, the value of the claim, the matter number and so on. If this information is typed onto a Writ of summons in a law firm, it is then filed in hard copy at the court and a court clerk in turn copies the information from the piece of paper into the court's computer system, the file at some point goes to the judge who hears the case. If a judgment is written, the judge's secretary types the judgment and copies the information from the file onto the cover sheet of the judgment. If the matter is appealed, the whole process starts again. The same information is keyed and re-keyed. The savings in being able to transport this information electronically are obvious.

7. Legal XML?

Already a few lawyers have seen the value of moving to a standard mark-up language such as XML. The Legal XML Work Group was founded in November 1998, and 'is a non-profit organization comprised of volunteer members from private industry, non-profit organizations, government, and academia. The mission of Legal XML is to develop open, non-proprietary technical standards for legal documents and related applications.'

The work group now has several 'vertical' groups which are each considering standards for different legal document types: Court Filings, Case Law, Public Law (e.g., bills, statutes), Private Law (e.g., contracts, wills), and Publications (e.g., legal books, law journals). There are also 'horizontal' standards being considered for things such as Citations, General Vocabulary (e.g., names, addresses), and Logical Document Structure (e.g., root elements, tables, outlines, paragraphs, media, general structural methodology.

The group has already had several face-to-face meetings and is well on the way to agreeing standard legal XML Document Type Definitions for various legal documents.

8. Conclusion

Lawyer should begin to think about creating documents in a format that computers can understand. Much time will be saved and this will result in savings to lawyer's clients and more consistent and accurate use of information.

Mark-up languages such as XML really are powerful tools to give meaning to textual documents and with standards such as Legal XML being developed, it will mean that the legal profession will really move online and in line with the rest of the business world.

This is a Commentary published on 29 February 2000.

Citation: Stanfield A, 'The New ABC: Teaching Computers How to Read Legal Documents', Commentary, 2000 (1) The Journal of Information, Law and Technology (JILT). <http://elj.warwick.ac.uk/jilt/00-1/stanfield.html> .New Citation as at 1/1/04: <http://www2.warwick.ac.uk/fac/soc/law/elj/jilt/2000_1/stanfield/>