Copyright (c) 2000 The Kansas Journal of Law & Public
Policy
Kansas Journal of Law & Public Policy
Spring, 2000
9 Kan. J.L. & Pub. Pol'y 425
LENGTH: 11126 words
Special Issues Involving
Electronic Discovery
John H. Jessen
SUMMARY:
... In my presentation this
morning I want to provide you with an update of what's happening in the area of
electronic discovery: What I see the courts doing, what I see
companies doing and what I see plaintiffs doing. ... Electronic data is no
longer an afterthought in discovery. ... The fact was, they had taken the data
off of the computer, stored it on a magnetic tape and hung it on the wall. ... A
3480 tape cartridge is the most standard backup tape ever used in the history of
computing. ... Where is the electronic data? Whose computer is holding copies of
this company's faxes? The telephone company. ... If you have a lot of data,
you're more vulnerable to the problem of electronic discovery.
... The printer driver with those data files, that's electronic
discovery, using that kind of information to show what really is taking
place behind the scenes. ... Increasingly, electronic data sets will be the only
source of discovery information, which means electronic
discovery is going to be increasingly important. ... It's very
important and can save time and money and make things more fair if an
organization can evaluate its electronic data processing systems and data sets,
current, legacy and future, and take some steps to make electronic
discovery more accurate and more rational. ...
TEXT:
[*425]
In my presentation
this morning I want to provide you with an update of what's happening in the
area of electronic discovery: What I see the courts doing, what
I see companies doing and what I see plaintiffs doing. I also want to discuss
the various types of data I see being brought into discovery and ways that I
think discovery can or should be controlled by decision-makers.
The most important point driving this whole discussion this
morning is that electronic data is now being aggressively targeted by attorneys,
law enforcement agencies, private investigators and regulatory agencies of all
kinds. It is no longer a matter of "IF" a given electronic data set is going to
be pursued, but rather "WHEN" it will happen and "HOW MUCH" it will cost to
respond in a proper, timely and economic fashion.
The two most
important words here are "aggressively" and "targeted." Electronic data sets of
all kinds are now being pursued directly and aggressively in a directed, focused
manner. Electronic data is no longer an afterthought in discovery. It is being
considered right from the beginning as an important information source. In fact,
in many cases, electronic data is seen as the most important type of information
that can be pursued.
One major issue in this area involves the
various requests and orders for electronic data that are being made. These
orders and requests are having a profound impact on discovery. We find that many
decision-makers do not necessarily understand electronic data processing
operations and the impact that results from their decisions. We see orders being
granted forcing organizations to turn over all of their electronic mail or to
freeze all of their back-up tapes or to allow the adverse party or a third party
access to their data sets. While perhaps valid in some matters, these types of
decisions are overly broad in most cases. We have found that in most cases these
decisions are based on either an incomplete understanding of what the impact
really was going to be or what alternative options existed for getting to the
data.
In one case, a plaintiff asked for a copy of every
electronic database in a large, multi-national corporation. You've got a couple
of problems with a request like this. First, what's a database? A database can
be defined in any number of ways. Second, you couldn't dream up a litigation
that would legitimately require a copy of all of the databases of any size
company, let alone a large company like this. The judge said, "I don't really
understand this issue. Come talk to me about it." Everyone shows up.
[*426] As the hearing began, the judge leaned forward and said,
"Listen, folks, I want to wrap this thing up quickly. I'm leaning toward
granting this. This weekend my niece showed me what a database was on her little
computer, and it doesn't seem like it's that big a deal." That was the frame of
reference being brought to this decision, a decision that would have lead to the
settling of the case if the court had approved the request.
There have been cases where decision-makers have relied upon
very weak sources of information in order to make decisions. In one case, it was
rumored that the judge was relying on a then-teenager working at a local Radio
Shack store as his electronic expert in a case. n1 So where the information
comes from that is being used to make the decisions can vary across the board. I
have run across decision-makers and special masters and magistrates who have a
very good background in dealing with electronic data issues. For the most part,
however, this is not the case. This is a new area and a lot is still left to be
learned. Today there is a wide variety of knowledge being brought to this arena,
which means there's a wide range of decisions and orders being granted.
Here are some things that my company has seen: We saw a law firm
sanctioned for failing to distinguish between purged and deleted files. They
came back, basically, and said, "We can't provide the data being asked for. It
doesn't exist anymore. It's been deleted from the system." A very simple
30(b)(6) n2 deposition, within twenty minutes of the start of the deposition,
found all the data. When the attorney had met with the client, he asked the
computer folks, "Can I get on this computer and find this data?"
He was told, "No, you cannot. It's impossible to get it off this
computer."
He said, "Fine," came back and told this story. But
the real story was that the attorney didn't know enough to continue asking the
right questions of his client's computer staff, nor did these computer people
volunteer the full answer, which is very often the case. The fact was, they had
taken the data off of the computer, stored it on a magnetic tape and hung it on
the wall. Subsequently, when I met with the same computer folks, we met in the
same room in which they had met with the lawyer. That room had a big window
overlooking the computer center. Straight across the room were the shelves
holding the tapes that were holding the data. The attorney did not know what to
ask and was not able to properly determine whether or not the data really did
still exist. In addition, the computer folks had no real incentive to volunteer
information, nor did they particularly understand the importance of this issue
to the case.
It's very important today to make parties be very
clear about the terminology they are using. If they say these are the only tapes
that exist, fine, but it had better be true. You shouldn't be able to go to an
outside storage center or a closet or someone's desk drawer and find another set
of tapes.
The Fen-Phen litigation right now has that issue where
tapes came up late in the case, and now the jury is going to be told about this.
There are many, many cases so far where stuff shows up after the fact because
somebody just happened to run across it or it was identified in a deposition.
Parties need to be very clear about what does exist and in what format it
exists.
[*427]
Another situation we see are
companies being ordered to turn over all of their electronic mail due to their
failure to produce a small selected set.
Common answers to
requests for data are often "it's just too hard to find", "it's too costly", "we
don't have that old tape drive anymore", "the programmers aren't here who set up
the system", and "no one knows how to use it", and the like. The government once
responded to a request for data by saying, "We can't provide the electronic data
you're asking for because our computer system is too complex." That was it. That
was the answer.
They were told, "That's too bad. It's your
computer system. You set it up. You decided what hardware, what software, what
level of training was needed. You made every decision there was to make about
this computer. You could have set it up in a way that made it easy to do this
sort of search and response. You chose some other business model. That's your
problem. Now go and make it happen."
Companies today,
organizations of all types, have got to take into account the fact that their
electronic data is going to be sought in discovery and put mechanisms in place
to identify it, to locate it, to retrieve it, to review it, to preserve it, and
I'll talk specifically about preservation, and then to produce it in a proper
format. It's becoming such a mainstream thing today that it's just incumbent
upon them to do so.
In one particular case we got about 11.5
million electronic mail messages turned over because the other side said that
they couldn't do the search because they did not understand their own electronic
mail system. During the search we located about 10,000 relevant messages. Most
importantly, there were two in that 10,000 that were very interesting that I'll
show you later. Participants in a discovery action today have got to know what's
in their e-mail before they lose control of it and it goes over to the other
side. If a party comes in and says, "We can't read our e-mail system," sorry,
that doesn't let them out of the game.
If the other side is
prepared to deal with this type of response, they raise their hand and say,
"We'll deal with it. Give us the tapes. We'll read them. We'll pay for it. We'll
take care of it if you can't do it."
We get hundreds of data
sets annually by being in the position of saying, "We'll take it." It's very
important for parties to know what they have and how to get at it. If they
don't, they may lose control of their own data.
The numbers you
have to deal with today are interesting. Eleven and one half million electronic
mail messages. If each of those messages were printed on a sheet of paper, it
would constitute pretty substantial discovery. In an electronic environment,
11.5 million things, in this case e-mail, is not that impressive of a data set,
quite frankly. One of the major issues facing discovery in the future is just
the sheer volume of data that will have to be dealt with. How do you manage
millions and millions of electronic items?
I was meeting with a
prospective client based in Chicago a few weeks back and we were talking about
electronic mail and the volume that usually exists. The president of the company
was there, and when he heard about the 11.5 million electronic mail
[*428] messages he said, "Well - that's ridiculous. No company can
stay in business with that much stuff." He was thinking paper and the attendant
infrastructure you would need to manage all of it. He didn't realize that
electronic data does not require the same levels of infrastructure that paper
does.
I said, "No, you have over a hundred thousand employees.
You've got a lot more mail than that." He sent his chief information officer out
to find out what their e- mail traffic was. The guy came back and said, "Sorry
to tell you, chief, it's 22 million," and the president just went ballistic. He
said, "That's no way to run this business. We better start doing something about
it," blah, blah, blah. When he settled down, the CIO looked at him and added,
"Per week. Twenty-two million new unique e-mail messages are created on a weekly
basis."
We know historically that up to forty percent of a
corporate electronic mail set is non-business communication, or kittens-for-sale
mail. In this company's case, that's millions and millions and millions and
millions of things being created monthly that have nothing to do with the
business, people speculating, guessing about things, talking about things,
talking about other people, et cetera. We know that about eighteen percent of a
data set that size will have attachments, or other files attached to the e-mail,
such as spreadsheets, word processing files, databases, graphics, whatever, each
one of them becoming a new discoverable animal. Eight percent of that data set
will be broadcast mail which goes to more than one person.
We
have had about half a dozen cases now where the total number of electronic
things brought into play-not that were available in a global set, but which were
available after a reasonable initial review of the set-went over one billion. A
billion pieces of discovery material. Those kinds of numbers introduce a whole
host of issues about scope and management. How do you manage a billion things
like that in an evidentiary way? Each one is very simple. Each mail, each file
may be very simple in its own right to work with, but a lot of small numbers
start to get you. You start multiplying a second or two by a billion, you've
gotsome serious problems. So now every search for a specific person's name, for
example, has a material impact in terms of time and cost. How do you manage that
and deal with it? These are all questions that are being worked on as these
quantities of data start to flow into the discovery process.
A
company may be sanctioned for failing to include electronic data in their
production even though they had not been asked for it. n3 You don't have to be
asked specifically for electronic data. It's a document. It is defined as a
document everywhere. It's something that should be considered by the parties up
front and be identified, located, retrieved, reviewed, preserved, and
ultimately, if necessary, produced.
Plaintiff firms are becoming
increasingly sophisticated in this area. I'll show you a couple of areas where I
see traps being laid all the time. One is not asking for data specifically and
then coming back downstream and making electronic data requests and then using
any absence of data to claim destruction or spoilation.
A
company may be sanctioned for failing to produce the electronic data to the
[*429] receiving party in a reasonably usable form. n4 This is
another recommendation I'll make at the end. Make the parties agree on what
constitutes a reasonable form. Once they've identified, located, retrieved,
reviewed, and preserved, once it's time to produce, there's an obligation to
produce the data to the receiving party in a reasonably usable form. What does
that mean? It could be a floppy. It could be a mainframe. Who knows what it
means? It's not defined, and it can't be defined in generic terms. It must be
defined on a per case basis.
I know one case where a company was
sued for predatory pricing by a small competitor, and they asked for a copy of
the large company's pricing database. This company went and made a copy. They
ran a special backup and turned 120 backup tapes over to the other side. The
tapes were IBM standard 3480 tape cartridges. A 3480 tape cartridge is the most
standard backup tape ever used in the history of computing. It's the IBM
standard. If you want to give data to the National Archives of this country, it
has to be on 3480 magnetic tapes. That's their standard. It's an international
data exchange standard. It's been around for years, and they'll be around for
another who knows how long. The company made a backup onto this tape and gave it
to the other side. The little company comes back and says, "Well, this is no
good. We're just a little company. We don't know what these are. We don't have a
3480 tape drive. The tape drive costs about $ 18,000. We don't have a mainframe
to plug it into. We don't have DB2, the big mainframe database system that would
run this stuff. We can't use it. You're just playing games. This was not a
reasonable way to give us this data."
The judge agreed and said,
"It's obvious you were trying to preclude them from using this information.
Because you were doing that, I'm going to preclude you from using this
information." They were not allowed to use any of their pricing information at
trial even though that's what the case was about, pricing. An extremely large
verdict came back in that case. The company's general counsel later said that he
wished he had just bought them a mainframe computer. It would have been cheaper.
A company may be sanctioned because backup tapes showed up after
the discovery cutoff. That's becoming more and more common today. Again, a lot
of what's happening is trying to catch people and companies not having an idea
of what their data sets look like, not having an idea of where they are, how
recoverable the data is, or how they can make use of it.
Federal and state discovery rules are changing to adapt
to this new environment. Rule 26(a) n5 changed back in 1993, December of 1993,
to specifically address electronic data. We've seen that used now many times.
The states that are adopting state rules, for the most part, are modeling their
state rules after 26(a). California was different. Take Senate Bill 1034 n6 back
in 1995, for instance. It passed in the House and then got held up in the
Senate. Bill 1034 was all about electronic discovery. It said
that computers "have become so commonplace that most lawsuits involved in
discovery have some type of computer stored information and that the principle
embodied in California's discovery statutes is that information that is stored,
used, or transmitted in new forms, including [*430] computer data,
should be available through discovery with the same openness as traditional
forms." That's not a bad definition, actually. In fact, that's an unwritten
definition everywhere. That's exactly how it already sits everywhere.
What 1034 tried to do that was particularly nasty was that it
tried to create a whole new level of discovery called an "information phase,"
where the parties would have to exchange information about their hardware,
software, policies and procedures, and data sets so that they could learn enough
about each other's computers to then come back and do what 1034 called "proper"
discovery. Basically, you had to turn over information about your system to the
other side so that they could learn about your systems in order to then conduct
the "proper" depositions and ask the "proper" interrogatories and
requests-for-production. Every rule that we've seen changed, every proposed
change that we've encountered opens the door to more and more information
sharing or discovery about hardware, software, infrastructure, policies, and the
data sets themselves.
Another thing that's happening is that
users of computers are using them in a way that creates risks. Several surveys
have been done of computer users, and there are four beliefs that continually
rise to the surface. Most computer users believe they have a privacy right
inherent in their use of the system. This is not true under the Electronic
Communications Privacy Act. n7 That's been litigated and held, and yet most
people feel they have a privacy right and, therefore, do things on the computer
that they might not otherwise do.
They believe that delete means
delete. It does not. It means delete less and less every day. User friendliness
is the overriding theory in software and operating system development today.
Being user friendly means making it easy for the user to do things, including
getting back deleted data. If the user accidentally deletes something, let's
make it easy to get it back. If you do that, then you have a loyal, happy
client. The theory of user friendliness also resulted in safe formatting. If you
reformat a drive, a floppy disk for example, the default setting is a safe
format, which means you can recover all the data underneath that format. You
have to actually tell the software otherwise for anything else to happen. It's
actually much more difficult today to really delete something. The systems are
just set up in ways that encourage the retention of data.
Third,
most users know what backups are and most have an understanding of the concept
of a backup. What they don't know is who's doing it, what data is being backed
up, where it is, or for how long those backups stay in existence.
Fourth, they just don't believe anybody has the right or the
desire to come after their electronic data. That's just not true today.
Regulatory agencies and litigators are going crazy about electronic data. The
SEC is practically betting their future on electronic data. The EPA out of
Denver, their enforcement division, has a group that goes out ex parte and
seizes computer systems. The IRS has had a training center for the seizure of
computers for about nine years now. The DOJ, their antitrust group, is working
on an electronic data acquisition policy. Everybody wants this stuff today: law
enforcement and investigatory agencies of all types.
[*431]
I was on a panel recently with an attorney who made an interesting
comment. He said that one of his greatest fears for his clients was that, before
long, anybody who ever sold a used computer was going to be out trying to advise
attorneys on going after electronic data, and, indeed, in some locations, we're
starting to see that very type of thing happen. It's critical that you look at
credentials, to look at what knowledge is being brought to the acquisition and
the retrieval and the review of the data, because it varies greatly depending on
who's doing it. Without the appropriate skill set data can be misinterpreted. A
perception of nefarious activity, such as data destruction, can be made when it
just wasn't the case. One note, we have found over the years that an academic
skill set is often the worst skill to bring to this arena. The way in which
computers act in the real workplace and the way in which humans really interact
with computers is far removed from the classroom. A Ph.D. in computer science is
often a guarantee of misinterpretation of data unless tempered by real work with
real computers in the real world.
I pulled out some sample
e-mail from some cases just to give you an idea of how people are using
electronic mail. The first one: "Yes, I know we shipped 100 barrels of
[redacted] but, on our end, steps have been taken to ensure that no record
exists; therefore, it doesn't exist, if you know what I mean. Remember, you owe
me a golf game the next time I'm in town." This came from the shipping computer
of a company involved in a big toxic tort litigation that had claimed that they
had never worked with the particular chemical in question.
Here's my favorite case: Our client was a big national retail
chain who went to one of their vendors and said, "Give us 2500 free units of
your product. We're going to go out and do a test marketing study for our
stores, and, if it goes well, we'll place an order for 100,000 units." Well, the
vendor said fantastic. Who wouldn't? The vendor gave our client 2500 units of
their product, and the market test went extremely well. Our client bought
100,000 of these things, put them on the shelves, at which point they didn't
sell at all, singularly one of their worst products in over three decades of
retailing. They went back to the vendor and asked, "Did you change the
packaging?" They tried to figure out what was wrong. Eventually litigation
ensued.
We found this e-mail in the laptop of the salesman who
had put the deal together. It had been deleted, or so he thought. He lays out
the deal to his boss here. He says, "I recommend we provide the free units for
the following reasons: 2500 free units is cheap compared to the profit from an
order of 100,000." Good legitimate business. "They will provide the names and
addresses of the stores involved in the study so that we can drop ship the
product." That's fine. "Since we will know where the product will be, I will
send someone around to buy all of them, thereby ensuring a successful test!"
Uh-oh - not so good. That's creative marketing, right? His boss writes back, "If
you can pull this one off, I'll buy you a new car; you name the brand."
Another example: "Did you see what Dr. So-and-So did today? If
that patient survives, it will be a miracle!" This hospital chain, in lieu of a
Christmas bonus a few [*432] years back, bought every doctor a
laptop computer and told them to start recording their practice notes in the
word processor and start using e-mail to communicate with others. It probably
was very productive, but a risk was created by not providing any education about
protocol, retention, deletion, and so forth.
The last two
examples come from that set of 11.5 million e-mail messages we talked about
earlier. We found 10,000 relevant e-mails, two of which were very interesting.
The first read: "Hi, David, please destroy the evidence on [the name of the
case] that you and I talked about today. Thanks, Laura." Well, that was good.
That probably would have been enough, but we decided to keep looking. We wanted
to know if anybody in this company was dumber than Laura. We weren't
disappointed. David writes back, "Evidence destroyed," in the subject line. A
little note here, anytime the subject line of an e-mail says "evidence
destroyed", you want to look at that one. "Hi, Laura. Acknowledge your message
and taken care of. Aloha, David." David hopes there's a federal penitentiary on
Maui for destroying the evidence here and then documenting it so well in the
e-mail system.
All of these examples had been deleted. Deleted
means very little. It makes it harder to get the data. It makes it more costly
to get the data. But deleted data can often be recovered or retrieved from
related or support systems, such as backup tapes.
What is data?
We keep talking about electronic data. Let's try to define it a little bit. It
seems like a silly question. What is data? What isn't? While this seems like a
silly answer, the truth is that virtually everything is electronic today. There
is more paper in use today than ever. There is more paper consumption than ever
before, but, in the past, there were multiple sources of paper: it was
mimeographed, it was typed on, it was handwritten, whatever. Today there's
virtually only one source for the paper documents that you see: the print key on
a computer. This means that the true source for virtually every paper document
today is an electronic file sitting on an electronic device somewhere. Thepaper
is a facsimile. It's a representation of the true source sitting on that
computer. There's an interesting issue with electronic data and paper. Without
exception, an electronic data file contains more information than can ever be
printed to paper. Whether that information is evidentiary to a given proceeding,
who knows, but, without exception, an electronicdata file contains more
information than you could ever print to paper. It's impossible to put in a
paper format - or any physical format, for that matter, microfilm, microfiche,
whatever - all of the contents of an electronic data file, which means that the
electronic source file is the source that needs to be looked at to get to the
real answer. There are things in electronic data files that might give the
answer. If you're looking for dates and chronology, temporal matters, a lot of
what may be helpful is embedded in files in ways that you can never print out.
You've got to look to the file. How a file is physically stored, how and where
it's stored on a hard drive in relation to other files around it, how pieces of
it are commingled and mixed with other pieces will tell you how old it is; or at
least it will answer the question "Is it X years old?" If you can date things,
you can tell a lot of information. I can look at a file on a hard drive and tell
[*433] you if that file was created on that drive or if it had been
copied to it at some point, a tremendous amount of information comingjust from
that drive infrastructure. That's evidence, depending on what you're trying to
prove.
Typical electronic things that we see pursued today: Word
processing files are interesting, very interesting, actually. It used to be word
processing files were simple little text files. Today they're extremely
sophisticated data processing. You can embed other files inside of a word
processing file, you can link them out to the Internet or whatever else you
want. They are tremendously sophisticated files that hold a lot of information.
Financial and accounting information has been automated for decades, and is very
popular in discovery. Personnel records are one of the hottest areas because
most personnel records today are automated, i.e., payroll and commission
databases. When you get in trouble, attorneys are coming in and asking, "Yes, I
want the paper personnel files but I also want any electronic files." That's
when they ask for the e-mail. "I want any e-mail that mentions my client's name
regardless of sender or receiver."
Here's a piece of software
that showed up in a matter. This software replaces old-fashioned employee
reviews. When it's time to do the employee's performance review, you key it in,
and, when you're done, you hit a button and it reads what you just wrote and
then rewrites it using legally and politically acceptable language. But it keeps
both versions. It has this big dictionary of bad words and corresponding
replacement words. Now, that's a staff meeting you didn't want to miss, right,
developing that list? It has a little editor, so, if you use certain words that
aren't already in there, you can put in your word and replace it; "moron" with
"challenged", for example. That's an interesting list to read, your own list, in
fact, it was used in this case to see what they keyed in as their own kind of
bias, what were they thinking that they didn't want to say and put in there.
There's all kinds of software like that, hundreds of different kinds of packages
that capture information, store it away, make it available and can be produced.
Databases are being sought. A database is basically defined as a
logical collection of related information, so, if that logical set of
information coincides with the topic of discovery, there you go. There's a
tremendous amount of information stored in there for people to start going
through.
Policies and procedures are being targeted today, going
in and looking at the policies of how data is handled, especially surrounding
backup tapes. People are coming in and taking depositions of backup
administrators and finding out what their backup policy is and then asking for
inspectors to come in and count the number of tapes. A backup policy is a
mathematical formula, we'll make X number of backups per week, we'll save the
fifth one, we'll keep that three years, we'll keep one of those and save it ten
years. You can extrapolate that out and find out exactly how many tapes should
be in existence at any one point in time. If the physical count doesn't match
that, very interesting. The adverse side will say, "Why did you destroy those
tapes?" Then keep using that theme, "Why did you destroy those tapes?"
"We didn't destroy them. We lost them," or, "We didn't make
them," or [*434] something. All of a sudden, one side is on the
defensive for a very logical, practical real world environment reason, and yet
it's very easy to show that discrepancy and say, "Well, you've got lots of
excuses, but your policy says you should have this, here's what you have." I
have seen that kind of tactic used many times.
Backup and
archive tapes are of interest because backups do hold differential data. The
backups hold data that are not available anywhere else, so they are a legitimate
target; but the questions become which ones, and how many? A favored, and I'll
just say, generically, "plaintiff,"-meaning someone who really wants to get the
other side-tactic is to ask for all the backup tapes. The producing party then
has to read all their tapes, because every single backup set contains something
different. So, again, we see situations where tremendous volumes of data are
brought into play because of that reason. But there are better ways to do it. If
you made a full backup, if you back up every file in a computer onto a tape
every night, and you do that every night for a year, you will have 365 tapes. If
you look at January 2nd's tape and look at every file on it and compare that to
the day before, would you expect to see a big difference in the amount of data?
No, it's negligible. We call the difference the "data differential." The data
differential is very small when you start shrinking down a time frame. A
day-to-day data differential is very small, so the likelihood of unique data
residing on tapes a day apart is very small. When you compound that with
relevancy, the likelihood that there is unique data on day-to-day tapes that's
relevant to a particular matter approaches zero. As you spread the differential
out, the data differential gets bigger. If you take a weekly jump between tapes,
there is going to be more unique data between those two sets than daily,
obviously, and monthly, et cetera.
Historically we have found,
that when dealing with literally hundreds and hundreds of thousands of tapes
over the years, a quarterly tape review often works well. You have to have some
parameters about having a common set of data from a common system; you can't
commingle data from all over the company and do this - but you can take a set
coming from the same system over time. For instance, quarterly tape reads
generally yield over ninety-five percent of the data. So, instead of reading 365
tapes, if you pick one every three months, read four tapes, you'll get
ninety-five percent of the data. Assume that it costs $ 1000 per tape to analyze
it. Four thousand dollars versus $ 365,000; is the cost different enough to
force the producing party to spend $ 361,000 more to get five percent of the
data? Who knows? It might be. It depends on the case. That's where we see cost
sharing coming into play a little bit. Usually the cost burden is put on the
owner of the data, and probably rightly so, because, again, how the computer is
set up does impact discovery in this case. You can set up your systems, you can
prepare your systems in a way that makes this stuff a lot easier and a lot
cheaper. If you choose not to do that, then perhaps the burden should be on you.
But should you have to read every tape? So this whole idea of data differential
analysis I think is going to play an important role in the future to make it
more fair.
Lets look at an example. In one case, there were
about 1800 tapes that had to be [*435] read. They went back over a
twenty-year period, lots of different computers, and lots of different data. The
company went out and got three bids to read the data. Four million dollars was
the cheapest. We conducted a data differential study and found that sixty-four
tapes out of that set would yield over ninety-eight percent of the data, 98.4
percent of the data, actually. Should the producing party have to spend four
million dollars rather than perhaps a couple of hundred thousand dollars for
just 1.6 percent of the data? The receiving party doesn't have to be totally
precluded from the data. This is where cost-sharing comes in. The producing
party should pay for the 98.4 percent of the data. That's fair. But, if the
receiving party still wants the remaining 1.6 percent of the data, they should
have to pay for that processing.
Spreadsheets contain a lot of
information. CAD, or computer aided design, files help in the products liability
environment. CAD files include computer aided design, computer aided
manufacturing, computer aided modeling, computer aided engineering, computer
aided statistical quality control, and computer aided sampling. Every aspect of
design today is automated. Over eighty percent of CAD files never get printed to
paper because they are what-ifs. It's an engineer theorizing about a document,
changing a little thing and seeing how that works, well, that didn't work
changing something else and seeing how that works. Those get saved. All those
incremental drawings get saved. The ones that get printed are usually the final
sets, the production sets. What if you come in and find all of them? Let's say
the fortieth version of the drawings was the production drawing. The expert
comes in and says, "Well, I looked at all these incremental drafts, and in the
thirty-sixth version, see that extra screw here, see how it doesn't appear in
the next one, in my opinion, that screw would have prevented this accident. I
guess they just didn't want to pay the cost of an extra screw on each wheel."
Using a very small change in drawings to then come back and make an argument;
but, if the automated system is capturing every single version, it's very easy
to go back and grab that. Is that fair? Should the company even be keeping all
of those what-if drawings from the start? Again, there's a whole host of issues
to be raised today. In the product liability environment, the CAD stuff is
extremely popular.
I have a client whose primary CAD design
group, if you look at them from a business model standpoint, is overstaffed by
thirty percent. The reason being, at any given moment up to thirty percent of
that group is either involved in discovery, being deposed, or being prepped for
a deposition. That's how invasive their CAD files have now become in their line
of litigation. It's a tremendous cost, and yet the plaintiff firms who
specialize in their industry have discovered that it's very fruitful to go in
and ask for these things.
UNIDENTIFIED SPEAKER: How much would
have to be at stake for someone to even initiate an electronic data search
effectively?
MR. JESSEN: It depends on how knowledgeable the
attorney was about taking some of [*436] the early depositions. If
an attorney has some basic knowledge of how to initiate the process and just
needs some assistance technically, it might only be a couple thousand bucks to
properly invoke electronic discovery in a personnel matter. On
the bigger product kind of cases, it may require significant dollars.
We are seeing firms, plaintiff firms especially, ramping up for
electronic discovery. In the professional plaintiff
environment, we are seeing that the plaintiffs are way ahead in this area, and
the corporate defense bar is way behind. They have not put in the thought, or
the resources, to counter this yet, because the incentive lies with the
plaintiff today in this area. So we're seeing professional firms put in hard
dollars. We're seeing organizations like NELA, the National Employment Lawyers,
and ATLA, having workshops. They both have workshops on how to conduct a
30(b)(6) n8 deposition of an e-mail administrator. There are publications, white
papers and things floating around on how to do it. I have lectured in law
schools over the years and I have seen a dramatic change in the classroom. Years
ago, honestly, it was like "what's a computer and why do we care?" That was the
kind of line of questioning. Then time and billing for a while became big. Then
demonstrative evidence was the huge thing four or five years ago. How can we
make a great graph that will impress everybody? Now it's electronic
discovery. I know a couple of schools that want to develop a curriculum
around electronic discovery and have a track for electronic
evidence. We're seeing a growth in that area. So, at a school level, at the
organization level, at the firm level, we're seeing hard dollars being poured in
so they can develop a lot of electronic discovery skills
internally.
E-mail we've talked about, just the volumes of the
mail involved make e-mail a staggering discovery event. Voice mails, are
interesting. This is an example of a traditional technology that's changed over
time. It used to be voice mail wasn't a big risk in discovery because it resided
on third-party proprietary platforms. You couldn't get to it, it got overwritten
effectively, and it was hard to deal with. Today every major voice mail system,
every one, is a computer program running on a regular old computer, which means
the voice mail messages are digital files sitting on computers, which make all
of the rules of discovery, of recoverability, of review, et cetera, fall right
into place. We routinely find thousands, tens of thousands, in one case, a
quarter million, voice mails showing up from various backups and computer
platforms. You can search those. You can make use of those. Here is a technology
that's changed over time. It wasn't a problem. Now it is. It makes it incumbent
on all of us to understand how technology, past, existing and future, is going
to play a role in and impact discovery.
Here is an example of a
new technology. There is no reason a caller into a fax machine should ever
receive a busy signal. The telephone can grab the in-coming fax, save it on a
computer, and send it to the fax machine when it's free. That is an interesting
activity. It's a very productive thing for a business, right? But, let's look at
it from a discovery standpoint. Where is the electronic data? Whose computer is
holding copies of this company's faxes? The telephone company. Do they delete
it? No idea. Do they [*437] backup their computer? You can always be
certain people backup computers. How long does that backup stay in existence
before it goes away? We have no way of knowing. We know two things. The
company's data is on a third-party system that gets backed up. We don't know if
it ever gets deleted or how long the backups stay around. Can you go to a third
party and get a subpoena and get that information? Yes. I've heard that Yahoo!,
one of the big Internet service providers, gets so many subpoenas today for
e-mail they have a separate fax line set up just for subpoenas. This type of
activity is becoming mainstream discovery very quickly.
UNIDENTIFIED SPEAKER: Why would a company, like a telephone
company, design a system where they hold faxes indefinitely, not delete it after
sending it to the receiving fax machine?
MR. JESSEN: First, when
these systems were initially designed electronic discovery was
not an issue. People were not coming after this information so it was never a
criterion in the design spec. Secondly, the issue of user friendliness comes
into play again. Isn't it nice when you lose one of your faxes and you can get a
copy from the vendor? Doesn't that make you happy about paying them for their
service? Should these companies start addressing this issue? Absolutely. They
could, and they should, and that's one thing in the risk reduction area that we
are seeing companies starting to address. What do we have, how long do we really
need it, applying traditional record retention rules and the like.
Today things like backup tapes are being addressed. When I first
started with computers, my first job was making backups of the mainframe. I had
sixty-five tape drives I was responsible for. It took all night using these old
nine track tapes to make one backup. You didn't have many backups. It just took
too long. You couldn't physically do them. So backups, when they were originally
envisioned, were short- term disaster discovery devices. Over time, because it
got cheaper and faster and easier to make backups, they became long-term, ad hoc
record retention tools, and so we've gotten lazy, basically. It's very easy to
just keep a lot of data today.
Video mail is starting to show
up. Put a little camera at everyone's desk so they can do ad hoc telephone
conferences and video mail messages. Twenty percent of the laptops on the market
today have a camera built in, so when you flip open the lid, there is a camera
staring at you. You've seen keyboard mail. You've all heard voice mail you
didn't like. What are people going to do when they have a camera at their desk
and they think they have privacy and they think delete means delete? Everything
you're thinking and I guarantee you a few things you've never thought of.
Once people understand these risks, aren't they going to stop
doing it? The answer is, some will. Some will not. Others have to be continually
reminded. There have been serious news stories in the popular press about these
risks for six years now, and it still continues all the time. I often speak for
clients at auditoriums full of [*438] employees, 500, 600, or more,
on the horrors of e-mail, kind of a fire and brimstone anti-e-mail kind of talk.
We look at their e-mail traffic before that talk, and then we look at what
happens afterward; and, on the average, after they hear this, there is a sixteen
percent drop in e-mail traffic. Remember, forty percent of an average corporate
e-mail set is non-business communication, so that sixteen percent that goes
away, first of all, is not business related. Over about a three-month period, it
starts creeping back up. You've got to keep reminding them. Humans are humans,
basically, is the answer, and some will do it, some won't, others will have to
be told all the time.
UNIDENTIFIED SPEAKER: Typically, when
there's a question or a request for discovery under e-mail in a court, state
court, and the request is granted, when you go back and you retrieve the
information, then what or how does the judge make a distinction between work
product and information that may actually be discoverable? Particularly if it's
under e-mail; not under a file or in a cabinet or anything, but if it's just
under e-mail.
MR. JESSEN: It depends a lot on how the parties
discussed it in advance. If the owner of the e-mail says, "Well, we're not going
to do it, it's too costly, it's too hard, we don't know how to do it," which is
probably half the time, and the other side says, "We'll do it," most of the
time, they're given the whole e-mail set.
UNIDENTIFIED SPEAKER:
Based on that?
MR. JESSEN: Yes. And they will sign a
confidentiality agreement saying they won't use anything that isn't relevant.
Then you see the whole set turned over.
UNIDENTIFIED SPEAKER:
Judges are granting the right to retrieve that?
MR. JESSEN: Yes.
It's called "grazing", to go grazing through the set and, basically, run
searches of their design. There are ways to protect privilege if these searches
are done properly. You can run a search to exclude or to preclude privileged,
trade-secret, and private items. You can run the search in a way that it moves
the "hit" files to a new drive. Then those can be further reviewed for
privilege. There are ways to run searches against any kind of data that will not
allow humans to see the content. This is a good way to protect privileged
information. If a search is properly designed, and many times you have to run a
number of searches from various angles, it can identify the relevant set of data
in a timely, cost-effective and proper manner.
UNIDENTIFIED
SPEAKER: I don't have any problems with those things that can be indicated as
files, because that may speak for itself, but I'm talking about the straight
files or messages that are sent just by e-mail which may just be half joking.
[*439]
UNIDENTIFIED SPEAKER: That's life.
MR. JESSEN: That is what they're doing. I mean, e-mail is a
communication system. It's a viable part of a business today. They are
conducting business on e- mail.
UNIDENTIFIED SPEAKER: Do you
have any cases in which a discovering party hoping to discover a smoking gun
memo apparently doesn't discover it and, in the absence of it, fraudulently
creates one in the system? If so, how would the victim party exclude that from
occurring?
MR. JESSEN: I have not seen that at the end of a
process where a party is given a set, because, for example, we never give
anything to anybody without first creating what we call a "virgin set" and
placing it in escrow. So, if we ever have to turn over a client data set to the
other side, we have an evidentiary copy of the data made and immediately sealed
and put into escrow. Then, if the other side ever finds something, then we'll
have a special master appointed to replicate the finding. That takes care of
that. There are some cases where employees fabricated mail in advance as a
strategy and then sued the company and pointed their attorney in the direction
of those particular mails. I can tell you that, barring e-mail, it's usually
fairly straightforward to prove that those were fabricated. E-mail is more
difficult. If you have an e-mail system and you leave your mail logged in and
you run out to go get a cup of coffee and someone walks in your office and types
mail and runs out and isn't caught, every single piece of evidence, electronic
evidence, will prove that mail came from your terminal.
Now, we
have a program that does a psycho-linguistic analysis of the e-mail. We can read
the e-mail and look at things like verb usage and grammar, and can profile the
author's writing style. You then compare that profile to a known set of e-mail
belonging to the purported author. You can compare and see if that mail doesn't
fit that person's profile of writing, and that's actually been used several
times to prove somebody was not the author. It's a lot of work to go through
this process, but it can catch the fabrication of evidence.
UNIDENTIFIED SPEAKER: I hate to ask this question, but are you
aware of any cases in which judges' e-mails have been subject to discovery on
recusal motions?
MR. JESSEN: No.
UNIDENTIFIED
SPEAKER: Good.
UNIDENTIFIED SPEAKER: Is there software being
developed, and, if not, why not, which would be sensitive to these discovery
problems?
[*440]
MR. JESSEN: There are software
systems being developed, we do it for example, that will make this whole process
much faster, cheaper and more defensible. The primary issue, however, is
managing risk from an overall viewpoint. If you have a lot of data, you're more
vulnerable to the problem of electronic discovery. We are
seeing more attention being paid to these issues at corporate levels, to
retention, to recoverability, to searchability of data; and there are tools
starting to enter the marketplace that do make it easier.
To go
through a couple more examples. The Internet is just a huge way for data to go
in and out of companies. Trade secret theft is way up because the Internet makes
it easy. Hostile work environment. A NELA lawyer once said at a conference that
he thinks the Internet is very simple because it only does two things; trade
secrets go out, pornography comes in. That's about right. The Internet creates
problems, as do Intranets where a company has set up these databases internally
where stuff just gets piled on and never gets deleted. Extranets, where you have
outside organizations talking together, are starting to become very popular.
Counsel use Extranets to share communications.
Groupware, like
Lotus Notes, is not just e-mail but, basically, you design a system, let's say
for forms processing. You never want to fill out a PO again, so you buy a
groupware system and you put your forms on it. You fill them out electronically,
they are evaluated electronically, the order is placed electronically, and the
invoice comes back electronically. It's a way to tie together a whole business
application electronically, and they are becoming more and more popular. They
save tremendous amounts of time.
Deleted data isn't deleted, we
talked about that. What next? Who knows what's next. I mean, in Manhattan right
now, there's a pilot study going on with a shipping company where they put
sensors on all the doors of the delivery trucks and a satellite sensor on the
top and they are tracking in real time the movement of the vehicles, and every
time a door is opened, so that if someone is trying to steal stuff, they know.
All that is captured, and they can model that entire transaction. I mean, just
unbelievable kinds of data being captured today by people and events and things.
Who knows what the next one is? This is by no means an exhaustive listing of the
kinds of data that can be used in discovery. These are popular ones we see
increasingly, but all kinds of those come up. It's very important to not just
think of electronic discovery as e-mail. That's an important
part of it, but it's not everything. It's a log showing when people logged in.
Home use of computers is becoming very important. I know one attorney that's
been able to get forty ex parte discovery orders to go into people's homes and
take their computers away in trademark and IP infringement things. It's
everything. It's all kinds of uses.
Electronic evidence is fast
becoming the primary source of discovery. If you read the legal press and
things, there have been two kinds of milestones indicated. At some point in the
future, in the majority of litigation in which one is involved, the other side
will understand electronic discovery issues; they'll have the
tools and the know-how to properly execute it. That's one to three years away
before the majority of litigation involves some type of properly executed
electronic discovery. Right now most electronic
[*441] discovery is not properly done. It's sophomoric and
overly broad. It's, "Give us all the mail, give us all the backup tapes." It
needs to be more explicit than that, but at some point, they will catch on, and
proper requests will be made.
The second milestone after that is
a quantitative one. It's probably five to seven years out, and that is when the
majority of discovery in the majority of litigation will be electronic and paper
will take a second seat after that. There are several reasons why
electronic discovery is taking over.
First,
virtually all information is produced and retained electronically. Huge
repositories of data exist in lots of formats, lots of backups and things.
People just don't like to get rid of stuff. It's a cut and paste effect. You
save every file because, at some point in the future, you may want to cut a
sentence out and paste it in another document. Tremendous volumes of things get
saved away.
Second, electronic data is the most authentic
evidence because it has things in it that can never be printed to paper. Printer
drivers, for example. There was a case where some employees left our client. It
was believed that they took CAD files, computer aided design files, with them.
They claimed, no, that they recreated the files based on their own knowledge and
expertise. Well, our client had a printer. It was a very exclusive, very
expensive printer. The printer driver, the instruction set that tells a file how
to print properly on a printer, for that particular kind of printer, had to be
custom written and part of the custom written driver was the serial number of
the printer. The unique serial number of the printer had to be embedded in the
driver. So one unique aspect of a CAD file from that company was that it had
this unique printer driver embedded in it, so when we went to do discovery
against the other side, all we looked for was that. Even if they had the same
printer and even if there was a one in a trillion chance they coded the printer
driver identically, because the serial numbers are never the same on printers,
they could not be identical, and we found thousands of CAD files with our
client's exact printer driver. The case settled immediately. That was the
evidence. The printer driver with those data files, that's electronic
discovery, using that kind of information to show what really is taking
place behind the scenes. If people go in and try to modify that, the computer is
keeping a record, and it can show that the edits were made. There is a
tremendous amount of technology available today to show what really is happening
with these things.
Third, much of the information in a computer
today never gets printed to paper. This is a six-year-old study. At that time,
twenty to thirty percent of all the information put in a computer never got
printed to paper. I think it's much, much higher today. Recent studies say
seventy percent. I'd be very comfortable with fifty percent. How much of your
e-mail do you actually print? Not much for most people, right? Big databases
rarely ever get printed. Groupware, these groupware systems that are set up, are
not meant to be printed. Eighty percent of all CAD files don't get printed.
Tremendous volumes of data are being created and stored in systems, which never
go out to the fiscal media. So traditional discovery takes place and you're
automatically writing [*442] off half the data set and probably the
most interesting half, because it has the e- mail and the embedded edits.
Deleted data is still there.
Finally, and probably most
importantly right now, the way this is being used in the real world, simply the
threat of someone having proper electronic discovery conducted
against them is a powerful new negotiating tool. In mediation or negotiation
between parties I've seen electronic data used very bluntly. "We know what it's
going to cost to read those tapes, and we'll take ninety percent of that and go
home," using the difficulty, using the data sets and what it's going to take to
deal with them as a negotiating tool. Good trial lawyers know this. They know
that an unprepared organization, especially corporations that have disparate
data sets and lots of data, have to go to extreme expense to do this.
Here are some ideas for rationalizing electronic
discovery a bit. If our rule of law means applying law to facts, then
electronic discovery has to be part of it, because that is
where everything is being created. The facts of how an organization runs today,
how a business runs, the facts of any given matter are, for the most part, in
the computer. That's the reality of it. Computers are often the best, and
increasingly, are going to be the only source for information. If not there
already, they will be in the future. Increasingly, electronic data sets will be
the only source of discovery information, which means electronic
discovery is going to be increasingly important.
Unfortunately, legal institutions and the computer processing world have never
learned to live on the same planet. They have different backgrounds and even
different languages. It has been a very difficult process getting the legal and
computer worlds together. To move forward and to properly integrate electronic
data issues into discovery, the challenge is going to be in developing rules,
authorizing things, allowing things that make sense, given a rational evaluation
of the data sets and the ways in which the data was used within a given
organization.
I believe that a qualified privilege for
self-evaluation should be given. It's very important and can save time and money
and make things more fair if an organization can evaluate its electronic data
processing systems and data sets, current, legacy and future, and take some
steps to make electronic discovery more accurate and more
rational. I can tell you that the reason a lot of companies don't try to
integrate electronic discovery into their standard discovery
routines is that they don't want to create documents about their systems and
data and make it easy for the plaintiff to come in and find out all that
information about their system so easily. Organizations should not be penalized
for good faith efforts to evaluate and improve their electronic
discovery capabilities. By recognizing a self-evaluation privilege,
like we already have in the environmental arena and in some of the civil rights
laws, electronic discovery can be integrated into standard
discovery much more quickly, more fairly and certainly much more efficiently.
[*443]
Don't encourage stonewalling. Make it clear
to both parties, that electronic data is discoverable and it's going to be
allowed and they're going to be expected to understand it, to make proper
representations, and to do a fair level of discovery.
Don't use
cost as the primary criterion. I just talked about the fact that computers are
often the only source of the data. It is not necessarily cheap to deal with your
electronic data, but the reason this is true is because it's not been thought
about and mechanisms have not been put in place to do it. So the owner of a data
set cannot be allowed to sidestep discovery because, either by purposeful
design, by neglect or being unaware of the issue, they make the computer system
hard to deal with. That cannot be allowed if indeed the factual material is in
the computer. Organizations that have a lot of data must deal with it. The
challenge to them is going to be putting in a system in such a manner that it
makes it economic and timely. That's their burden to do it.
Some
innovative discovery techniques that we recommend are requiring parties up front
to come up with descriptions of their system in the meet-and-confer process. If
they have a problem and they're not cooperating and not sharing information and
they have to meet and confer, put some teeth in it. Set some standards for what
has to be exchanged at the meet-and-confer process. I've seen judges say that
they have to each bring two of their computer people and have an informal
discussion. They have to bring a certain set of information about the computers.
They have to bring a description of the data. I mean, you can put some teeth
into what happens. Or you can have a special master preside over that process
and walk them through and then have them jointly come up with an agreeable plan.
Tell them right up front, "I want you to meet and come up with a reasonable plan
on how you're going to identify data, how you're going to search it, how you're
going to produce it," so you can get around this in a reasonable form. Make them
define that early on so that later on it's not a squabble.
Have
preservation requests put in place early on. One case in particular, ATI v.
Sprint where Sprint was sued by Applied Telematics for a patent infringement,
ATI did not ask for any electronic data until late in the case. n9 Then they
came back and made a very explicit request for data. They said, "We want to see
the relevant data that resided on your backup tapes that were in existence when
the case was filed," clearly when there was an obligation to preserve relevant
data. They didn't ask for stuff earlier. Sprint came back and said, "Sorry, we
don't have those tapes. They've been reused, they've been rotated, we wrote over
them."
Sprint argued normal course of business. They said,
"We've always rotated our backup tapes. This is how we've always done it." That
was true, and they used industry standard. They said, "Everybody rotates backup
tapes," and that's true, too, everybody does reuse backup tapes. The court in
this case said that Sprint, by not taking into account the possibility that
potentially relevant information may have existed on those backup tapes and by
not preserving them, purposefully destroyed the evidence. Sprint settled two
days later. They're on record as saying that decision was wisely settled. ATI's
lawyers are on record as saying they are very, very pleased with the settlement.
This hit the front page of Corporate Legal Times, a three-page story about two
years ago now. n10 It got a lot of attention in the press. This is used all the
time today. I can tell you that this is one of the other tactics being used
today by aggressive firms, this sandbagging of requests, [*444]
holding out as long as possible to ask specific electronic data questions and
then hoping that the data is gone, hoping that there was no preservation system.
Have the parties deal with preservation right up front so this doesn't become an
issue. Put the data in escrow if necessary. It doesn't mean they have to look at
it. Preserve it, put it in escrow, then squabble about if and/or how it is going
to actually be used.
Have a data mediator deal with
this. That's the next slide. We're seeing more and more special masters, brought
in to mediate between the parties and come to agreement. Who to use? Don't use
an academic. An academic might be good for a software dispute, software code,
operating systems, but in this kind of world, you want somebody who understands
how users use data, who understands that backup tapes go bad sometimes and
that's just how it is and won't spend three months on every theoretical
methodology of recovering data. You want somebody who understands how computers
work, how people use them, and the realities of data processing. This could very
well be the future of electronic discovery. This idea of making
the parties meet and confer and, with the guidance of somebody who is going to
walk them through it, put a plan in place right from the start.
So, in summary, we're seeing electronic
discovery being conducted far more professionally than ever before, not
attorneys making up computer questions but actually having the background
themselves or using resources that understand electronic
discovery. It's aggressive, and it's targeted. Electronic
discovery will become increasingly popular because they'll read about
it. How many times do you have to read Bill Gates's e-mail before you want it in
your case? Electronic data will be, and must be, increasingly targeted because
that's where the data is.
Accordingly, new tools, new knowledge
sets, new protocols, new ways of dealing with electronic
discovery have got to be developed. Otherwise, these wide discrepancies
in what's allowed and the burdens being placed on owners of large data sets will
literally end cases long before any issues of truth get brought up. That is not
justice, and it isn't fair. Fairness will come from developing these tools and
forcing parties to agree on ways to identify, locate, retrieve, review, preserve
and produce their electronic data sets.
FOOTNOTES:
n1 See Armstrong v.
Executive Office of the President, 90 F.3d 553 (D.C. Cir. 1996).
n2 See
Fed. R. Civ. P. 30.
n3 See Daewoo Electronics Co. Ltd. v. U.S., 650 F.
Supp. 1003 (Ct. Int'l Trade 1986).
n4 See Crown Life Insurance Co. v.
Craig, 995 F.2d 1376 (7th Cir. 1993).
n5 See Fed. R. Civ. P. 26(a).
n6 See S. 1034, 104th Cong. (1995).
n7 See Electronic
Communications Privacy Act of 1986, Pub. L. No. 99-508, 100 Stat. 1848 (codified
as amended in scattered sections of 18 U.S.C.).
n8 See Fed. R. Civ. P.
30(b)(6).
n9 See Applied Telematics, Inc. v. Sprint Comm. Co., 1996 WL
539595 (E.D. PA 1996).
n10 See Bruce Rubenstein, Somebody Destroyed the
Evidence, Corp. Legal Times, Sept. 1997, at 1. 
