January 1, 2004

The main issues in e-disclosure concern volume, searching, metadata and data format.


With both the increased use of computers and the comparatively low cost of data storage, there is no effective pressure on organisations to embark upon aggressive document management policies. For example, it has been calculated that a PC with a 20Gb hard drive can contain enough paper if printed to reach the height of Mount Snowdon. If one assumes that each page takes a minute to read, then one person working eight hours a day, 300 days a year, would take 29 years to review the contents of that PC. According to information supplied to me by Phil Sealey of Deloitte and Touche, the figure would rise to 88 years with a 60Gb hard drive and 438 years for a 300Gb hard drive.

Volume, however, is not the only factor in favour of storing information electronically. Electronic data is easy to produce, to disseminate and to store. Indeed, the absence of physical constraints means that, in theory at least, the use of electronic data could continue to grow exponentially into the distant future.

As a litigator, when one is faced with a data mountain, as will increasingly be the case, how does one comply with the requirement under the Civil Procedure Rules 1998 to deal with a case in ways which are proportionate? What is certain is that the volume of electronic data will be such that it will no longer be possible to print out each piece of data, check for relevance or privilege and then formulate a client’s case. The traditional approach will simply be time and cost prohibitive.

Searching for electronic documents

Despite the increased volume of documents, it is far easier to interrogate the electronic data by using search terms than it ever was when dealing with paper. However, if one extends the search to the many available sources of electronic data, the starting point from which a litigator needs to conduct a standard disclosure exercise is more burdensome than it would otherwise have been. Is it not a clear abuse of the disclosure process to give wider disclosure than that strictly required simply because it is easier and cheaper for the disclosing party to undertake electronic disclosure since that short-cut will result in increased cost for the other party? How does one define a reasonable search for the purposes of making disclosure when the volume of data that now exists is so vast and the CPR offers no assistance? Should a search be made only of data directly accessible on a computer? Can one restrict the search to key word searches? Given limited technical knowledge amongst the judiciary, how likely is it that an application to the court will assist?


Contrary to the position with hard copy documents, there are a number of different dimensions of relevance to electronic data. In particular, quite apart from what the record says, there is the embedded metadata information for each document which details how, when and by whom the document was created and amended. That information may be of particular relevance in the context of an argument about the provenance and authenticity of documents. Hence, in appropriate circumstances, it is just as important to make proper disclosure of metadata as it is of the information contained in a hard-copy document itself. Furthermore, if one accepts documents in hard-copy form that were previously in electronic form (eg any word processed document or e-mail), can the metadata be requested too?

Data format

Electronic data comes in four different forms; active, replicant, back-up and residual data. Active data is what is held on the hard drive of a computer and is accessible simply by turning on the computer. Replicant data is copies of data automatically created by the computer when, for instance, one opens an e-mail attachment. Back-up data speaks for itself and is held on magnetic tapes. Finally, residual data comprises data which has been deleted from active use and is relegated to the nether regions of the hard drive.

The problem here is that, although all of these categories may contain disclosure information, each of these other than active data may require appropriate technical input to put the data into a form in which it can be read. This can be hugely expensive. In Zubulake v UBS Warburg LLC 2003 U.S. Dist. LEXIS 7939 (S.D.N.Y., May 13, 2003), UBS informed Zubulake that it would cost approximately $300,000 to restore and produce e-mails held on back-up tapes.

The Way Forward

Until litigators, either with or without technical support, understand the issues surrounding electronic disclosure and attempt to embrace them in the process of making proper disclosure in all cases, both parties and their lawyers will inevitably be exposed to criticism from the courts with the risk of serious cost sanctions.

This means that there has to be a serious raising of awareness within the litigation fraternity as a whole, including the judiciary. This is precisely what the CLF is attempting to achieve.

To help achieve this consensus, a questionnaire has been produced the purpose of which is to identify the general nature of the electronic documents held by each party and the architecture of electronic storage systems. A copy of that questionnaire is reproduced below and comments can be submitted via the CLF Web site.

It has been suggested that parties would not be willing to disclose to each other the architecture of their electronic data storage systems. Indeed, in some respects, this information could be just as damaging or commercially sensitive as the electronic data itself. The fact is that it should not be necessary to disclose the architecture of the electronic data systems in order to satisfy normal disclosure requirements. It should not be necessary to provide the other side with the completed questionnaire. It can be used purely as an internal aide memoire to enable a consensus to be achieved with the other parties regarding acceptable disclosure parameters.

Any consensus reached between the parties should include the fixing of technical criteria for the format of any database to be used for storing the electronic data.

In addition, it is submitted that, in the interests of proportionality, other than in the most exceptional case, a reasonable search for electronic data can be limited to data which is retained in a readily usable form (eg it is accessible on the hard drive of a computer) or which can easily be recovered. Guidance on this issue has been provided by the US Courts (see below). An example of an exceptional case where this approach might not be sufficient is a fraud case where, for obvious reasons, it may be necessary to venture beyond the active data stored in the electronic data system.

The US approach

An emerging principle in the US appears to be that if a court considers a request for electronic disclosure to be unreasonable, it will nevertheless be allowed but at the expense of the requesting party. In Zubulake, guidelines were provided for determining whether data was accessible or inaccessible and how this affected whether a cost shifting analysis would be appropriate. The case sets out a three-step analysis to assist in deciding which information should be treated as inaccessible and which party should pay the cost of producing data deemed inaccessible according to principles of proportionality. It appears that one of the central questions was, “how important is the sought after evidence in comparison to the cost of production?”

The American Bar Association has revised its electronic discovery standards dealing with preservation duties and cost shifting which they adopted in 1999. The new standards were circulated for comment in November this year. According to Gregory P. Joseph, the co-chair of the task force responsible for the new proposals, “they are designed to supplement existing rules and address practical aspects of the electronic discovery process that are not [covered] by the rules”.

The draft standards cover, amongst other things, the issues surrounding the appointment of an independent IT expert who is responsible for the extraction of electronic data (Standard 32). The Federal Rules of Civil Procedure already require that a discovery conference should occur at the outset of every case and the draft standards look at the effective use of discovery conferences to deal with electronic discovery issues (Standard 31). Standard 31 forms a checklist similar in principle to that referred to below but includes questions relating to accessibility of the electronic data, preservation of that data and appropriate search criteria to search for, inter alia, potentially relevant data. It is interesting to note that these standards do envisage parties disclosing to the court details of how the architecture of electronic storage systems are organised.

The draft standards are posted at www.abanet.org/litigation/taskforces/electronic/home.htm.

Electronic data is the largest single issue facing litigators in the 21st Century and it is inevitable that it will cause enormous grief for clients and their legal and other advisers unless lawyers become accustomed to collaborating with each other (and with the court) over the disclosure parameters and issues at an early stage, so that acceptable proportionate guidelines and benchmarks for this form of disclosure can be established in the cases they handle.

Neil Mirchandani is a litigation partner at Lovells in London. He gratefully acknowledges the assistance of Serena Cobley in the production of this article. The CLF Web site is at www.commerciallitigatorsforum.com. Neil’s fellow authors of the working paper are Mark Humphries, Head of advocacy at Linklaters, and Sanjay Bhandari, Senior Associate in Baker & McKenzie’s Dispute Resolution Department.



The purpose of this questionnaire is to assist each party to consider and explain to the court the architecture of its electronic storage systems, identifying the various electronic media which may contain information relevant to these proceedings and to enable the court to make an appropriate order for the disclosure of such electronic material.


1. Please explain, in general terms, the architecture of your electronic storage systems (eg in terms of where data may be stored centrally via servers, whether users access central servers and/or create their own files locally).

2. Please identify the networks or servers which you use.

3. Please identify where such networks/servers are physically located.

4. Please explain, in general terms, how individual users gain access to servers and/or mainframes. Do they obtain access via a network, the internet or through other means?

5. How often is information on networked systems backed up and/or archived?

6. On what medium is such backed up information stored?

7. Where are the back up tapes and/or archives stored?

8. Which back up software is used (manufacturer, package and version number(s))?

9. Do you use a data warehouse or on-line back up regime via ISDN, broadband or other dedicated transmission line?

10. If so, where is the transmitted back-up data stored and by whom?

11. Do you have a disaster-recovery back up?

12. If so:

(a) on what medium is it stored?

(b) where are such physical media located?

(c) what types of disaster recovery computers are used?

(d) which back up software is used?

13. Please list any organisations that may hold data on your behalf, eg ASP providers or scanning and coding bureaux.

14. As a result of the year 2000 upgrade, are there any systems which you have mothballed? If so, please provide details.


15. Please identify and list all individuals who you believe may have created electronic documents in your control.

16. In relation to each such individual please answer the following:

(a) Did that person create documents by accessing the network/server?

(b) Did that person create documents using a personal computer or data location [ISP location?] which may be stored on a personal computer/laptop/notebook/PDA or any other device? If so, please identify and list the relevant equipment.

(c) Who owns that computer or device?

(d) What equipment did that individual use which may have stored relevant electronic documents (e.g. PC, laptop, notebook, hard drive, PDA, CD-ROMS, DVDs, memory sticks, diskettes, zip drives, printers etc)? Please include equipment that the individual may previously have used but which he/she is not currently using. Please consider all forms of electronic storage media, whether magnetic or optical.

(e) Did that individual produce his/her own back ups? If so:

(i) how often?

(ii) in what media were such back ups stored (e.g. on floppy disc)?

(f) Are there any replaced or removed drives which are no longer in use but which may contain material relevant to these proceedings?