Introduction

The contributors to the earlier parts of this Special Supplement have identified a variety of ways in which the number of documents to be produced by the parties to an arbitration may be controlled and refined. In some cases, however, and despite the best efforts of both the arbitration tribunal and the parties, the production, or at the least the analysis, of substantial numbers of documents may simply be unavoidable. Typical examples are cases which involve significant accounting, technical or engineering issues, or cases with a complex factual background spanning a lengthy period such as construction disputes. The purpose of this chapter is to examine the assistance that technological solutions can provide in alleviating the burden upon the party producing the documents, the party receiving the documents and upon the arbitrators themselves.

Technology can in fact assist throughout the entire document-handling process, from gathering the documents in from clients and relevant custodians, through making the documents requested available to one's opponent and managing the documents during the pre-hearing phases, to producing the documents for the evidential hearings and presenting them effectively to the tribunal. Indeed, technology also has a part to play once the hearings are over, in providing a full, yet portable, record of the proceedings for the arbitrators.

Collecting documents for disclosure

One of the most important elements in ensuring that the production of documents is carried out in as efficient and timely a fashion as possible is the gathering in of potentially relevant material from the client, whether as part of counsel's investigation of the dispute or following a legitimate and targeted request for documents from opposing counsel. Here innovations in technology are both part of the problem and part of the solution.

The electronic archive

The problem stems from the enormous growth in email traffic and other electronic data. Email in particular has become ubiquitous. Some companies now operate a virtually paperless office, with 80% or more of their 'documents' being held in this way. The material contained in this electronic archive is of course potentially every bit as relevant or material as the traditional hard copy archive was. [Page102:]

The sheer ubiquity of email can make the gathering-in of and review of documents far more onerous than it once was. Rather than target a relatively limited number of individuals who are likely to have custody of relevant documents, it has become necessary to search far more widely. The distribution lists on emails are frequently extensive because transmitting the message to a large number of people is easy to do, and entirely free. The number of potential responses thereby increases, which themselves generate replies, and so on.

Perhaps because of its relative informality, there appears to be a tendency on the part of some individuals to respond to email in a less guarded fashion. This, in turn, leads to broader requests for the disclosure of email correspondence in the hope, on the part of the party requesting the documents, that an unguarded comment will be made in an email which would not, perhaps, be committed to paper.

Nor is the spread of electronic data confined simply to email. A company's IT system may contain financial or other critical data on Excel spreadsheets or in other proprietary software formats, to which a party to an arbitration may quite legitimately wish to have access. Diary information is often kept on office personal computers, on laptops, or even on personal digital assistants or PDAs. Potentially each of these categories of data may be relevant to a dispute.

The electronic archive is, as a result, significantly larger than the traditional documentary one ever was. Furthermore, it has the potential to be more widely dispersed, with access possibly required to company's servers, to back-up computer tapes, to individual hard drives and even to individual PDAs.1

Gathering this material in can, of itself, be a complex exercise and involve some difficult issues. Counsel has a responsibility to ensure that the disclosure exercise is properly carried out by the party for which he or she acts. In most cases it may be sufficient to entrust this task to the client's own personnel, particularly where the client is a sophisticated organization with its own IT function. Very occasionally, however, it may be necessary to enlist the assistance of individuals with specific forensic expertise.

Searching and refining the archive electronically

Fortunately, the essential characteristics of emails, Excel spreadsheets and similar electronic information make it possible to sort and to collate the material speedily and efficiently. Once the relevant custodians of information are identified, it is an easy task to transfer the contents of their email in-boxes, out-boxes and any relevant 'public folders' on to a CD, DVD or separate hard drive (rather than print off the hard copies). The data collected from all relevant custodians can then very simply be uploaded into a computerized database. Programmes are available which then run through the entire database and 'de-duplicate' it. This involves the automatic sifting and identification of duplicate emails. This is particularly important if an email has been sent to numerous addressees and appears numerous times in the database. With traditional hard copy documents, the sorting of copies was sometimes a contentious issue as individual custodians might have made some annotations on their copies which might or might not be of some relevance or importance to the matters at issue in the case. There was, therefore, frequently a desire to have more than one copy of the same document in a bundle. With emails, the ability and temptation to make marginal notations of any sort is minimal and the ability to discard genuine duplicates is correspondingly [Page103:] higher. The de-duplication tool is particularly valuable in saving cost and time. In the writer's own experience, it has served to reduce the population of potentially relevant documents on occasion by up to 75%.

Further refinements of the electronic information are then possible. It is likely that the request for documents will be targeted to specific periods, specific topics and, perhaps, specific individuals. Simply by inputting the relevant parameters or keywords into the database, a search can be conducted which will analyse all of the fields in an email message (including the sender, the addressee, any copied parties, the subject line and the text of the message itself) or in the relevant electronic media. This will then automatically identify documents falling within the relevant parameters and permit the non-relevant material to be discarded without the need for extensive and costly review by individual lawyers (although clearly some form of cross-checking is critical to ensure that the task has been completed properly). The lawyers' analysis can instead be focused clearly and specifically on the documents that remain, assessing them for relevance, materiality and privilege.

Nor is this technique and technology suitable only for emails. It can be used, as indicated above, for Excel documents, Word documents and electronic diary entries, albeit not with the same degree of accuracy, given the different properties of those documents.

Searches of this sort can, if necessary, be carried out directly on a client's own systems. This can be dangerous, however, as it could interfere with business processes or otherwise affect 'original' electronic evidence. Downloading the source material to a separate database and then running the searches is a safer option.

The further advantage of this technology is that it is capable of automatically generating a schedule of the documents which remain after the sifting process has been carried out, by reference to date, sender, recipient and subject field.

'Traditional' documents

None of this helps of course with documents in a more traditional format. Such documents still need to be gathered in from the client and other relevant custodians and collated in the conventional manner. There are software programmes available that purport to 'read' these documents by means of optical character recognition. The documents are scanned into a database and the software attempts to create a schedule of the documents, which is then searchable, by identifying certain key characteristics. One such programme works by searching for the word 'Dear'. The word that appears after that will be listed as the addressee. Similarly, the name that appears after 'Yours sincerely' or 'Very truly yours' will be treated as the sender. This technique works well when the letter is addressed individually. It is less useful if it is addressed to 'Dear Sir', or 'Dear Colleague'.

Whilst such programmes clearly have value as a means of carrying out a preliminary sorting-out of material, they are insufficiently advanced at this stage to justify total reliance upon them as a means of producing an entirely accurate compilation of the relevant documents for production. It will still be necessary, in most cases, for the lawyers to review the traditional documents, to analyse them against the other side's legitimate requests and to assess them for relevance and privilege. [Page104:]

Once the material documents have been identified, it is sensible in large cases to code them and to enter their details too into a database. This is useful both in order to identify documents responsive to a particular document request, and for case management purposes (to identify documents which might assist in evidencing one's own case). The format of such a database is very much a matter of personal preference. Typically, however, it would identify as a minimum the nature of the document, its source, the sender and the recipient, any parties copied on it and the date. The formal subject of the document should also be recorded, together with a note of any issues or topics arising in the case to which it is particularly relevant. It is also sensible for a core list of key topics to be identified and established by the legal team at an early stage and for the documents retrieved from the client, as well as those disclosed by the other side, to be coded against that list. This makes their deployment and retrieval at a later stage far more efficient.

Disclosure of documents to the other party

Disclosing documents in electronic form

It will be apparent from what has been said already that the exercise of actually disclosing relevant, responsive documents to an opponent is greatly simplified by the use of the various facilities described in the preceding section. The material responsive to a particular document request can be identified simply by searching the databases established. Material can be pulled down automatically from the database of electronic information and either printed off or, more usefully, loaded onto a CD or DVD which can then be handed over. The details of pertinent hard copy documents can be identified from the coded database and those documents can then either be photocopied or, if they have been scanned-in electronically, printed off from the database itself, or downloaded onto a CD or DVD. The generation of a list of documents being disclosed-if the tribunal considers one to be necessary and useful-is also a simple, and relatively automatic, exercise. It may require the removal of certain columns of information from the report-for example if the database has recorded key topics or comments by counsel-but this too is easily achieved.

Similarly, if as a matter progresses a legitimate request for the production of further documents is made, a search of the databases can promptly identify whether any responsive documents have already been collected, thereby expediting the response.

There has in the past been some reluctance amongst counsel to be too open to their opponents about the existence of a database. This appears to stem from the belief that it offers some form of tactical advantage. The usefulness of this technology is beyond dispute. No useful purpose can surely be served however by keeping it a secret. Indeed, national court rules in England & Wales require parties to a dispute to liaise with each other about the use of technology for the purposes of document management and disclosure, and to be prepared to discuss it at a case management conference.2 The aim is to minimize disputes, streamline the procedure and potentially to share the burden of costs. The same approach could and should be extended to arbitration. [Page105:]

Producing documents to an opponent in electronic format has several advantages. It is, first and foremost, a speedy and cost-effective method of making the documents available. Conventional photocopying of a large number of documents is both time-consuming and extremely costly. The delivery of documents on CD-ROMs is both easier and cheaper. The receiving party can then, if it wishes, review the disclosed documents on screen. This can be particularly useful in reviewing financial documents and spreadsheets which seldom print off in a reader-friendly manner. The receiving party can also print off any documents it wishes to scrutinize on paper. It may also be possible for the recipient itself to conduct word searches, particularly of the electronic documents, thereby simplifying their own review of the material disclosed.

Metadata

One particularly sensitive issue arises in relation to the disclosure of data and documents generated by computer. This is the issue of metadata. As many readers will know, metadata is the information which is effectively embedded in a document. A letter or memorandum produced by a computer word-processing programme will usually contain certain additional pieces of information which are not apparent to the reviewer unless certain properties are enabled on the reviewing computer. This will include details of the changes made to a document, the identity of the individual who created the document and the identities of anyone who amended it as well as the dates upon which the changes were made. With Excel files, the metadata will reveal the workings behind individual spreadsheets. Such information can occasionally be highly useful to opposing counsel and fertile ground for the questioning of the other side's witnesses.

The metadata, being simply the background data behind the creation of the document, can rarely, if ever, be strictly relevant and disclosable of itself.3 If this is so, care needs to be taken in disclosing documents which will have metadata embedded in them. It is possible to strip metadata from documents (either singly or in batches) using specialist tools. An alternative solution is not to transfer across the documents in their native electronic format, but rather to convert them directly to *tif or *pdf images which can then be read on-screen, but not analysed for the embedded data. This can now be done automatically, without the cost and time required to print off the documents and converting them through a scanning process into the alternative format, although care is still needed.

The use of 'sampling'

From time to time an arbitrator may be asked to preside over a case where there is a dispute over the operation of a business process which is repeated a multitude of times in a given day or week. If the dispute is as to whether that process has been properly carried out, there may be a legitimate interest on the part of one of the parties to the arbitration in testing the accuracy of the process in question over a given period. On the other hand, disclosure of all of the underlying information for the relevant period could be extremely onerous, particularly if the process involves a multitude of separate steps, each one of which generates a separate document and computer entry. A similar problem may arise in an accounting context where claims are made for, say, a failure to pass on certain costs contained in hundreds of invoices submitted on a daily basis. Disclosure of the whole documentary population in relation to that issue would be exceedingly onerous. A balance clearly needs to be drawn in this situation between the legitimate need of [Page106:] one party to test the evidence, and the cost and disruption that would be caused by requiring full disclosure of very extensive, repetitive documentation.

One possible solution in such circumstances is for the arbitrators to consider inviting the requesting party to examine a sample of the transactions or repetitive actions comprising the disputed process. This approach is particularly suitable for accounting disputes. It requires the party in control of the documents to generate a spreadsheet or list giving details of all of the individual transactions which make up the population of the disputed process. Clearly, if this can demonstrably be done from a computerized ledger, its credibility will be enhanced. A sample of the individual transactions can then be selected. It will usually be important that this sample is selected on a random basis and that the selection be carried out by the party requesting the documents, in order to avoid any appearance of impropriety. There are various software programmes available which can be used to generate the sample, thereby minimizing any risk of suggestion that the sample has been 'cherry-picked' or is not in some way representative of the population as a whole.

Once the sample has been selected, the disclosing party can then collate and produce the documentary evidence for the individual steps in the process. Clearly, it is helpful if the parties can agree in advance on the identity of those steps, and the description of the documents to be produced. If disputes remain, the tribunal might consider inviting the party-appointed experts, or an expert specifically appointed by the tribunal, to advise on the scope of disclosure legitimately required in such a case.

Access to a counterparty's business systems

Arbitrators may also wish to consider whether it is appropriate, in a given case, to allow a requesting party or its expert, or perhaps an expert appointed by the tribunal, to have access to and to conduct tests on the producing party's accounting systems and databases. The major accounting firms have, for example, developed software tools that can be used to analyse computerized ledgers and to search for and identify errors, anomalies and differences, greatly reducing the time and cost required both to produce, and to analyse, accounting materials. Even if the dispute is not of an accounting nature, the ability of an expert to run tests, either individually or on a joint basis, on a party's database or systems may obviate the need for the production of significant quantities of hard copy documents.

Even if such tests are technically feasible, they will not always be practicable or appropriate. If access to a company's systems will cause unjustified disruption or give rise to confidentiality concerns, alternative solutions may have to be sought. At least where accounting information is concerned, this may be done by providing electronic copies of relevant ledgers or databases on CD, DVD or separate hard drive. This avoids the need to print off copies (which can, in some cases, literally take days) and still allows the requesting party's expert to run analytical software on the data.

Case management

Even after disclosure has been given, the appropriate use of technology can continue to pay dividends during the case management process. On a day-to-day basis documents can be reviewed on screen without the need for extensive [Page107:] production of hard copy documents. More importantly, if both parties' documents are available in a searchable format (either through a fully electronic database or in part through a manually coded database), it enables documents that are pertinent to a particular issue to be identified and collated with speed and efficiency. As a case develops, fresh issues may arise and a search may be conducted for relevant documents. Counsel will, in most instances, have developed an idea of the main issues and topics in a case by the time disclosure is sought and received. However, there are software programmes available which deploy what is known as a 'concept search' tool. This effectively searches through all the emails and other electronic documents available on a database, examines the text in the message and then identifies concepts and themes which run through the documents. It will produce clusters of documents around each theme, which can then be reviewed. If the theme thrown up has merit, it can be adopted. If it does not, it can be discarded.

This software can also be used in a sort of 'reverse' mode in order to highlight gaps in documents received from the other party or even one's own client. If a particular subject has been the matter of ongoing correspondence for some months and is then not mentioned for a period of days or weeks, the programme can be adjusted to pick this up. A gap of this sort may indicate that documents are missing or, perhaps, that a decision has been made not to commit further discussion to writing.

The ability to search through large amounts of documents quickly can be useful at various stages in the life of an arbitration. It can enable a party to identify documents relevant to a particular interlocutory issue. It can swiftly identify documents responsive to a further legitimate document request or to an inquiry from the tribunal. It can distil the documents prepared or seen by a particular individual and which should be discussed with that individual as a precursor to preparing a witness statement or proof of evidence (if that stage has not already been reached). By means of simple inquiry, all documents drafted by a particular individual or sent by or to that individual can be identified and printed out and/or collated. The documents can then be further reviewed in order to refine the key ones upon which discussion should be focussed. The same approach can be used at the hearing stage in pulling together documents for cross-examination if the procedure adopted in the arbitration envisages a cross-examination process.

Presentation of documents at hearings

Even if significant quantities of documents have had to be disclosed in the early stages of a case, it should be possible for parties who are cooperating substantially to reduce the number of documents that are required for any evidential hearing. On occasion, however, and even with such cooperation, the mass of documents required for a hearing can be very significant indeed. The conventional way of handling this would have been to copy the relevant documents many times and to prepare a 'hearing bundle' for the tribunal, for counsel on both sides and for the witnesses.

Hearing presentation systems

Not surprisingly, technology has a part to play in simplifying this process as well. The process of identifying the contents of the hearing bundle may be facilitated by [Page108:] judicious use of each party's database of documents. Once the contents are agreed, the bundle itself can be made available almost exclusively in electronic form. There are several possibilities. The most sophisticated products consist of full hearing presentation systems. These are most typically used when the hearing bundle would otherwise simply be so enormous that it would be unwieldy and impractical to use. Each of the key players in the hearing-counsel, the arbitrators, the witness and the stenographer-will have a computer monitor placed in front of them. There is a separate system operator whose job it is to call up, for presentation on the screens at a given time, the relevant document. Clearly, in order to do so effectively, the operator needs to know what document is required and everyone participating in the hearing will have to know how to identify the relevant documents. Properly handled, such a system can be very efficient and can virtually create a paperless hearing room.

A slightly less sophisticated alternative is for the hearing bundle to be loaded onto a CD or DVD and to be pulled up and projected on screen by counsel for the parties as they need it, usually by means of a bar-code reader. This is cheaper and easier to operate and it can be used either for the entire hearing or simply for specific witnesses or documents. Documents needed for the cross-examination of a particular witness could, for example, be loaded onto a specific CD or DVD and projected on to a screen in front of the witness. This would obviate the need for hard copy bundles to be located and put in front of the witness and can further streamline the cross-examination process. To be effective, however, it does require careful preparation.

Nor does the display technology simply allow the straightforward presentation of documents. Even relatively simple programmes on the market enable counsel to compare documents side-by-side (say for the purpose of examining different drafts), to highlight extracts on screen with a virtual marker pen and to enlarge portions of the text (either for emphasis or simply to enable the relevant portion to be more easily read). Indeed, a presentation tool of this sort can be particularly useful in cases where there are a great many spreadsheets and other accounting documents. Spreadsheets are notoriously difficult to review in hard copy, particularly if there are a great many horizontal columns. The result is that what is notionally the first page of the spreadsheet will not print off on one sheet, and the confirmation appears many pages further on. By projecting the spreadsheet onto the screen, it can be easily navigated and reviewed without difficulty.

Properly used, these presentational tools can be very helpful. It is as well, however, to raise a couple of cautionary notes. This Special Supplement has been produced, amongst other reasons, to assist arbitrators to address delays caused by or in the production of documents. With this in mind it is important to ensure that the use of presentational software does not itself become the source of further delay. In practical and experienced hands it can enable the tribunal and counsel to move swiftly and efficiently through the hearing with minimum disruption and without constantly bringing out and replacing hard copy documents. Without experience, however, the technology can be cumbersome to use.

Furthermore, although the use of techniques such as highlighting, and expanding text, has its place, care needs to be taken to ensure that presentational gimmicks do not prevail over substance and that time is not wasted on unnecessary attempts to deploy or demonstrate counsel's full range of technological skills which do not advance the tribunal's understanding of the case.

In this regard, there is a significant advantage to employing a specialist, trained [Page109:] operator who will be able to use the system and navigate around the documents with maximum efficiency. If this is not practicable, arbitrators should encourage counsel to make sure that whoever is to be in charge of the presentational system has practised with it and can use it effectively.

A paperless hearing

One other consideration that arbitrators should bear in mind when deciding whether to encourage the parties to use a presentational system of this sort is the impact that it will have on the arbitrators' own deliberations. In its most sophisticated incarnation, a presentation system of this sort can result in a virtually paperless hearing. Is this what the tribunal wants? Some tribunal members may like to mark up particular documents. This cannot easily be done on such a system. Either a separate note has to be made about the impact of a specific document or the arbitrator will have to print it off and mark it up later.

A similar issue arises in connection with the arbitrators' review of the record for the purpose of drafting their award. Extensive marking of documents is difficult. The arbitrators will need, in these circumstances, to decide, and to let the parties know, what they wish to see. In most cases, written closing submissions exhibiting the key documents should provide the assistance the tribunal needs (and even this can be facilitated by 'interactive submissions', as to which see further below). It is in fact possible for documents that have been shown to a particular witness to be 'embedded' in the transcript by means of an electronic link. This means that when an arbitrator is reviewing the transcript on a computer screen, he or she can click on the reference to the document and the document itself will be displayed on the screen, so that the arbitrator does not need to find the document himself amongst the hearing papers or to search for it on a CD or DVD provided by the parties. The cost of this facility has reduced significantly over the last couple of years, with one supplier charging just three euros per document linked.

Post-hearing submissions

Few things can be as irritating for an arbitrator as having to make space available in his or her offices, following a hearing, for substantial quantities of evidence and transcripts. Yet these documents will be necessary for the purposes of deliberating, of drafting and of considering the award.

The solution to this dilemma may be for the tribunal to request, and the parties to agree, that any post-hearing submissions or materials be produced in an interactive format. This consists of a CD or, in cases where the documentation is particularly heavy, a DVD which has loaded upon it a party's post-hearing memorial or submissions, together with all of the supporting material upon which that party relies.

The technology is extremely versatile and allows varying levels of complexity. At one level, it is possible simply to load a catalogue of materials on to the CD or DVD for the tribunal to review and analyse in whatever order its members see fit. This could include the memorial, the transcripts, select documents, the statement of case and answer, and so on. In this way, the CD is an extremely portable library of key materials. [Page110:]

The interactive brief

Where the technology really comes into its own, however, is when it is used as a fully 'interactive brief'. In this format, the CD or DVD contains the post-closing memorial which has embedded within it links to the relevant materials upon which counsel wishes to rely. What this means is that the brief can state a fact which counsel believes should be found by the tribunal, or a proposition of law upon which counsel seeks to rely, followed by an exhibit number which is electronically linked to the relevant supporting document, portion of transcript or legal authority or commentary upon which counsel relies. By clicking on the exhibit number or link, the arbitrator is then able automatically to bring up on screen a copy of the document in question. Once it has been read, the arbitrator can simply close it and return to the original text, continuing to read the memorial until the next link is reached, and so on.

Subject to the capacity of the CD or DVD, there is of course no restriction on the sort of exhibits that can be linked to a memorial in this way. It is possible, for example, to link the full transcript for a particular witness to a particular point or points in the memorial. The link may cause the transcript to 'open' at a particular page, but the arbitrator will then have the capacity to scroll forward or back through the transcript to assess the context or to see how the particular passage of questioning develops.

It is also worth noting that particular portions of documents or passages from the transcript can be highlighted, so that the tribunal's attention is directed immediately to the relevant passage.

The use of such interactive submissions has considerable advantages. It avoids the need for the arbitrators and counsel to have to review closing memorials surrounded by a vast array of hearing bundles and transcripts. By doing so, it enables the key materials to be highly portable, permitting busy arbitrators to deliberate and to work on an award away from their offices. Finally, it ensures that the relevant supporting documents are properly reviewed by the tribunal as they are made available to the arbitrators without the need for a prolonged search.

One practical piece of advice: it is not possible to insert the necessary links into the memorial unless and until it is virtually complete. Depending upon the number of links and documents, the process can still take several days. It is therefore imperative that the parties and the tribunal make sufficient allowance for this process within the timetable and, critically, that counsel ensures that their memorial is complete well in advance of the deadline for service, in order that the technicians have sufficient time available to them to scan in the relevant documents and to insert the links.4

Conclusion

The pace of innovation means that technological assistance with collecting, reviewing, disclosing, presenting and summarizing documentary evidence is going to become increasingly sophisticated. Where this technology can be deployed cost-effectively to expedite and render more efficient the arbitral process, it should be embraced. Care does however have to be taken. The indiscriminate use of technology may hinder, rather than help. Arbitrators and counsel, might, I suggest, usefully consider the following questions early in the life of the reference: [Page111:]

? Is the dispute in question sufficiently complex and genuinely document-intensive to justify the use of document-management technology?

? At what stages will the use of such technology be useful? Not all of the facilities described above will be appropriate in every case.

? What will the cost of deploying the technology at each stage be? Some of the technology described is expensive. Will the efficiencies to be gained justify the costs incurred?

? Can the parties and the tribunal agree a direction for the use of document-management technology at the different stages of the arbitration? Early consultation should enable decisions to be made which will further save costs and enhance efficiencies, rather than having to 'bolt on' presentational software, for example, at a later stage. Knowing that a particular presentational software programme is envisaged and/or that the tribunal will want interactive closing submissions may enable a holistic approach to be taken which will avoid the need, say, for further scanning of documents at a later stage.

? As part of such a direction, should the parties be asked to liaise and to endeavour to agree upon the technologies and systems to be deployed, upon the electronic exchange of documents, the stripping of metadata, etc., again with a view to saving both time and cost?

? Finally, and this is probably a consideration more for counsel than for the tribunal, early attention needs to be paid to the means by which data, particularly electronic data, will be collated, from whom, and by reference to which topics. Careful thought and preparation at an early stage will pay significant dividends later.

1
The generation of so much electronic data has given rise to conflicting problems. On the one hand, some companies reuse their computer space within a very short period, which gives rise to urgent questions of document preservation when tapes are 'overwritten'. An even more dramatic example is where companies operate live systems which change on a daily basis. Care needs to be taken in these circumstances to ensure that critical information is not deleted or lost. The converse problem is the ability in many cases to recreate or restore data, which has ostensibly been deleted, by forensic examination of a computer's hard drive. Is this necessary or desirable? Deleted documents may very well be relevant. Discussion of these issues, whilst undoubtedly interesting, is beyond the scope of this paper.

2
Civil Procedure Rules, Practice Direction 31.

3
The US Sedona Principles, for example, contain a presumption that metadata is not preservable or disclosable unless it is itself material to resolving the dispute.

4
There are a number of companies that offer this service. Typically, they charge by reference to the number of 'links' in the finished memorial. One well-known supplier is currently charging 0.15 euros for each link to and from a particular page of a document and 1.75 euros for each link to and from a specific paragraph or sentence.