E-Discovery in practice – where are the Emails part 1

Recently I became aware of some confusion in the legal community about where to search for emails in the context of e-discovery.  This is an important topic because if you don’t understand where relevant documents and emails can reside you will not be able to adequately focus your e-discovery requests, and consequently may miss finding documents important to your case.  Some of this confusion results from a fuzzy understanding of important technical concepts in IT, in particular ‘server’ and ‘client-server’, and the associated software technologies like POP3 and IMAP that are used to for the delivery of emails.

Maybe these terms sound unrelated to an e-discovery request for emails from the opposing party, but not so.  Only when you have a competent understanding of the possible locations of emails, based on a knowledge of the way current technology works, will you be in a position to make an informed e-discovery request.  There’s no point requesting ‘all the emails on the Server’, when in-fact they almost certainly wont be there!

So I thought I would devote a couple of posts to address this confusion.

Lets begin with discussing the meaning of ‘client’ and ‘server’.

We’ve heard the terms often enough, and they come with obvious connotations, that we can imagine what they represent.  A server is something that servers information to a client that has made a requested.  Although the client and server entities are actually computer programs, it is common to think of them in terms of their underlying hardware.  A central large computer is the Server, and it responds to requests from many smaller Client computers for specific pieces of information.

The server and the client can be separate computers each running the specific server and client software, or these programs can reside on the same computer.

What’s a simple example of the server and client?  Well, when you log on to your user account on your computer (technically called a ‘session’), your account is a software program that acts as a client to the Operating System (e.g. Windows 10 or Mac OS) which is acting as a server.  Your local session requests information like access to files from the operating system which locates the data on your hard drive.  In this case, the server and client reside on the same physical hardware.

Another example is browsing the internet.  Your browser (such as Internet Explorer, or Firefox or Safari) sends out a request to display the contents of a web page.  The server of the hosting service for that site then sends back the files and images which the client then reconstructs and displays on your screen.  This is a simplification as I’ve ignored the complex processes in the process of transferring of information over the internet, but as an outline is adequate for our purposes)

A third example is with email.  A client program on your computer, such as Outlook, or Thunderbird or Mac Mail, requests downloads of your emails from the designated server.  This email server may be your internet provider such as Netscape or TPG, or a web-mail service such as Gmail.  Note that your browser can also act as an email client, though the process is somewhat different.  We will examine the different kinds of email clients in a later post.

This paradigm of partitioning a workload between a centralised responsive data container and multiple clients requesting information is called the ‘client-server relationship’.  Many advantages the accrue from this model.  One is that data is stored in a central location which make data-management vastly easier.  Another is that the system is more secure against data loss since backup processes are more comprehensive.

Because of the IT technology and the centralised nature of the Server it was for many years a physically monolithic high-performance device, while clients were typically small personal computers.  Higher performance is now achieved by linking together many small computers into a cluster.  So our conception of a server needs to be abstracted a step  to the idea of many, perhaps thousands, of individual machines working in concert and appearing to the outside as a single unit.

In its latest form, we have to abstract this even further.  A server is no longer a fixed collection of hardware.  Many servers can run on a single cluster of computers, each server program sharing an ever changing subset of the resources of the cluster.  A servers doesn’t even have to reside in one physical location.

There is one further point to remember.   Since a server is technically a software program, a computer can be both a server and client simultaneously that is, it may have server and client software running at the same time.  For example, MS Outlook may run on your central computer at work and acts as a client downloading emails from the email provider.  That computer then acts as the server to your computer in your office. When you want to access your work emails, you, as the client then access them

OK, that’s a lot of information for one post.  In the next post we will look at how email works, and where you need to look in order to find as much email as possible.