You are on page 1of 18

Tips for Avoiding Computer Crime

Copyright 1999-2000, 2002, 2004, 2012 by Ronald B.


Standler

Table of Contents
Introduction
1. Password
2. Anti-Virus Software
attachments in e-mail
3. Firewall
4. Avoiding Harassment and Phishing
5. Backups
6. Other techniques
Disabling Features in Microsoft Windows
7. Wireless Networks
Spyware
8. Links to other sites
Recommended books
Conclusion

Introduction
These suggestions are a companion to my separate essay on computer crime in
the USA. These suggestions are like installing high-security deadbolt locks on doors
of an office or home. There is no warrantythat following these suggestions will
prevent one from being a victim of computer crime, but at least one can make it a
little harder for a criminal, and maybe the criminal will find an easier target.

This essay contains only my personal opinions on the general problem of computer
security, and is not intended as advice on your specific problem. You should hire a
competent expert in computer security to review your situation and then advise
you.

1. Password
To access an online computer service or Internet service provider (ISP) one needs
both a user name and password. ISPs typically select a user name that is the same
as the last name of the subscriber. This means that user names are easy to guess,
therefore one must be especially careful with the password.

Select a good password:

Make the length of your password at least five characters. It is too easy for
automatic programs to sequentially try all combinations of characters in a
password of only 1, 2, 3, or 4 characters.

In short passwords, use at least one upper-case letter, at least one lower-case
letter, and at least one digit, for example, c5U3rN
A five-character password composed of only random lower-case letters has
about 8106 possible combinations, but a five-character password composed
of both upper- and lower-case letters and the ten digits, all chosen randomly,
has about 776106 possible combinations, i.e., about one hundred times
harder to guess.

To make a long password, use a concatenation of two unrelated words, each


with at least five characters, perhaps separated by at least one digit
(e.g., airplane5style). By having a longer password, it is no longer as
desirable to include a mix of upper- and lower-case letters.

Avoid obvious passwords e.g.,

o your name,

o anyone's first name (especially bad are your spouse's first name, your
child's first name, your dog's or cat's name)

o your nickname (e.g., "Flash" or "Buzz"),

o your home telephone number,

o your date of birth,

o your astrological sign,

o your mother's maiden name,

o your wife's maiden name,

o license plate number of your car,


o exact sequences of letters on a keyboard (e.g., QWERTY or
ASDFGH)

o sequences from the alphabet (e.g., ABCDEFG or ABCABC)

o or any other publicly available information.

Also avoid any of the above spelled backwards, and any of the above either
preceded or followed by a single digit.

There are about 8106 possible combinations of a string of five lower-case


letters that are chosen randomly. In comparison, there are only about
0.15106 entries in an English dictionary for college students' desks.
Therefore, it would make sense for a hacker to use a word list from a
spelling checker, instead of generating permutations of characters. In
response, you should make the hacker work harder by choosing a password
that is not in the dictionary of your local language.

For a given number of characters, the strongest password is a random


sequence of lower- and upper-case letters and digits. However, such a
password can be difficult to remember. My suggestion is to choose a
unusual foreign word that does not appear in the dictionary of your local
language. If you are tempted to use foreign characters (e.g., , ) in the
password on a computer in the USA, first ask the system operator if those
characters are allowed in a valid password on the system, as most operating
systems in the USA are limited to using only the first 128 characters in the
ASCII set (i.e., no foreign characters).

Of course, you don't need to limit yourself to official languages. <grin> You
can invent your own words, e.g., garkle6snerkle. Such words are much
easier to remember than unpronounceable clusters of characters,
e.g., c5U3rN.

Having chosen a good password, do not write it down, and do not tell anyone what
it is. (Get a separate account for your spouse, each of your children, each of your
co-workers, .... so that no one shares passwords.) This rule can create a problem if
you die or are incapacitated, so perhaps you should write it down once: on a sheet
of paper that you keep in a bank's safe deposit box (for your personal account) or
in a safe in the corporate office (for the company's computer).

When you get a new computer account, it will come with an initial password,
which password was probably randomly chosen. Follow the instructions from the
system administrator for choosing your ownpassword, and change the password.
The initial password may have been seen by someone who gave or mailed it to
you.

Use a different password at each website, service provider, or computer account.

Changing your password every few weeks is standard advice from computer
security experts. However, changing your password every few weeks also makes it
easier for you to forget your password. You need to decide if it is worth the bother
of changing passwords every few weeks. If you do forget your password, you will
need to contact a system administrator, prove that you really are the official user,
and get a new initial password assigned.

Many users store their user name and password in a logon script on their hard disk
in various programs: e-mail (e.g., Eudora), webbrowser (e.g., Netscape), terminal
emulator (e.g., Procomm), and modem control programs (e.g., Trumpet Winsock).
This storage of user name and password is convenient, as it automates the logon
process. However, if you store your user name and passwords in logon script(s),
then:

You should definitely enable the password setting in the BIOS of your
computer, so that a password is required every time the machine is switched
on. You might also enable a password setting in Windows98 or other
operating system, to give an additional layer of protection against
unauthorized use of your computer.

If other people have access to your computer when your machine is running
and you are away from your desk, you should install screen saver software
that requires a password to return to the operating system or applications
software.

If your computer is stolen, it is possible for the thief to logon to all of your
accounts. Therefore, it is essential that you logon to each of your online
accounts and change the password for each accountimmediately after the
theft of the computer is discovered.

This remediation includes changing your passwords at online stores (e.g.,


amazon.com). The information stored on your computer in
the cookies.txt file that your webbrowser accesses identifies you to each
online store, and could make it possible for a thief to impersonate you and to
charge items to your credit card.

Nearly everyone has private data (e.g., medical and financial data on a home
computer; business secrets on a computer in the office) on their machine. The same
suggestions about a password in BIOS and a password in a screen saver apply if
you have confidential or proprietary information on your computer. However,
unlike changing online account passwords, there is no easy way to destroy the
value of confidential data in files on a stolen computer. Users with very sensitive
data (e.g., military secrets, major trade secrets) should encrypt all of their data
files.

2. Anti-Virus Software
In the 1980s, computer viruses were generally passed from one user to another user
via floppy disks. Hence, users in the 1980s did not need anti-virus software if they
both (1) only purchased software from reputable sources and (2) never copied
programs from floppy disks provided by their friends and colleagues.

Three developments in the 1990s made anti-virus software essential for all
computer users:

1. It became common to distribute software and updates via downloads from


the Internet,

2. hackers developed viruses that were delivered inside macros for


Microsoft Word, which malicious macros could be hidden inside a
document sent by e-mail, and

3. hackers developed malicious computer programs that were commonly


distributed as attachments to e-mail: clicking on the attachment executed the
malicious computer program and infected the victim's computer.

Since everyone uses e-mail and nearly everyone will download executable
software from the Internet, everyone should have a good anti-virus program
running on their machine.

Because an anti-virus program will likely object to the installation of any new
software, the user should disable anti-virus program before installing new
software. Of course, before temporarily disabling the anti-virus program, use the
anti-virus program to scan the distribution files (e.g., CD-ROM, floppy disk, or
*.exe file downloaded from the Internet) for viruses. Do not forget to enable the
anti-virus software after you install new software.

It is not adequate to purchase anti-virus software with a new computer, install the
anti-virus software, and forget about that software. The virus definition file for the
anti-virus software should be updated periodically, because new viruses are
discovered every day. How often you should update your virus definition file is a
complicated question: the answer depends upon your tolerance for risk, how you
use your computer (i.e., receiving e-mail or downloading software from bulletin
boards is risky), and which operating system you use.
1. If your computer runs a 32-bit Microsoft Windows operating system (e.g.,
Windows 95 or later), then I suggest that you update your virus definition
files at least once every day.

2. If your computer runs an Apple operating system or Linux, then I suggest


that you update your virus definition files at least once every month.

Of course, when there is an epidemic of a new virus reported in the news media
(particularly a virus spread by attachments in e-mail), it would be wise to update
your virus definition files as soon as the developer of your anti-virus software
revises their virus definition files to recognize the new virus, and daily thereafter
until variants (copycats) of the new virus stop appearing.

I have seen estimates that, as of the year 2001, there are only about 80 known
viruses or worms that are specific to the Macintosh operating system. (Compare
this to more than 61000 viruses and worms for Microsoft Windows known in
May 2002.) An exact number of viruses and worms is not possible, because of
uncertainty about how to count variants of a virus or worm. As a consequence of
this relatively small number of viruses for the Apple, it is probably adequate to
update the anti-virus definitions for anti-virus software on an Apple computer
every month. This suggestion is in contrast to advice for users of the Microsoft
Windows operating systems, where daily updates of anti-virus definitions are
prudent.

Owners of an Apple computer who also use Microsoft Word wordprocessing


software (versions Word6 and higher) are vulnerable to macro viruses that use
Visual Basic for Applications, which also affect computers running a Microsoft
Windows operating system.

attachments in e-mail

As a virus, worm, or other malicious program can be transmitted via an attachment


to e-mail, one should rigorously follow three rules:

1. Never open an executable attachment (e.g., an attachment with a file name


ending in .exe or .vbs, amongst many other types) in e-mail without first
knowing the contents and source of this file. There is no harm done in
waiting a few hours or a few days to contact the person who sent the e-mail
and learn the contents and source of the attachment. The Melissa and
ILOVEYOU incidents, on March 1999 and May 2000, emphasize that you
can receive malicious programs from a person who you know and trust,
since that person could be a victim of a malicious program that
automatically sent e-mail in his/her name.
2. Never open any attachment from an unknown source. Simply reply to the e-
mail and request that the sender send the attachment as plain ASCII text in
the body of the e-mail. Or, if the e-mail is obviously junk, delete both the e-
mail and the attachment.

3. Be cautious of any attachment that has a double file extension, especially


when the rightmost file extension is an executable file type. (A file extension
is a three-letter code at the end of a filename [e.g., .htm, .doc, .exe, .txt,
etc.], that indicates the type of file.) Examples of dangerous double file
extensions are:
filename.jpg.vbs
filename.doc.exe
filename.zip.com
filename.gif.bat
filename.txt.pif
filename.mp3.scr
filename.htm.lnk

where "filename" can be any sequence of letters and numbers. Files with
such dangerous double file extensions are executable programs (perhaps
malicious programs) that are pretending to be a picture, a document, text, or
a webpage. This list of dangerous double file extensions
is not complete, because there are many different permutations of a
nonexecutable file extension on the left with an executable file extension on
the right, and because there are more than sixty executable file extensions in
Microsoft Windows.

Note: the text of e-mail containing malicious programs often contains


ungrammatical text, punctuation errors, or misspelled words, because the author is
a non-native speaker of English. Such mistakes in English text in an e-mail
apparently from a native speaker of English should alert the reader to the
possibility of e-mail from a forged address, which may contain a malicious
program.

3. Firewall
It is good practice to erect a "firewall" between parts of a computer system that an
external user can access (e.g., via modem or Internet or voice mail) and parts that
are supposedly accessible only by a local user. Many hackers run programs that
randomly search the Internet and probe ports on computers that are connected to
the Internet. If the hacker finds a port that is "open", the hacker might be able to
access that computer and view/alter/delete files on that computer. Worse, hackers
may also hijack the victim's computer and use it to launch their illegal attacks on
other computers.

Some hackers randomly search the Internet, probing ports controlled by a


malicious program called SubSeven. When the hacker finds a computer that
contains SubSeven and is not protected by a firewall, the hacker can access the
victim's computer through the backdoor provided by SubSeven. The SubSeven
program was first detected in June 1999, and there are many similar programs in
existence, for example, the Back Orifice program that was first detected in
August 1998.

When I first installed firewall software on my computer that I use only for e-mail
and webbrowsing, just 14 minutes later, someone in Atlanta, Georgia tried to probe
the FTP port on my computer in Concord, NH. During 21 hours of my use of the
Internet at the end of February 2002, there was an average of approximately
1 attempts/hour to probe a port on my computer. That statistic convinced me that
firewall software is prudent on every computer connected to the Internet. In
March 2004, there was an average of 57 attempts/hour to probe a port on my
computer. This dramatic increase in the number of attempts per hour to access my
computer shows that the Internet is becoming a more dangerous place and that
firewall software is necessary for a secure computer.

separate machines

Now that computers are relatively inexpensive (e.g., less than US$ 1000), I believe
that it also makes sense to have totally separate and isolated machines for external
access. The cost of having a separate computer that is dedicated solely to receiving
incoming modem connections and requests from the Internet (i.e., e-mail and
webbrower software) is offset by the increase in security with minimum
inconvenience to authorized users inside the building. When a secure computer and
a computer for external access are in the same building, communications between
them should be via floppy disk or rewritable compact disk, not via wire or cable.

For example, this document that you are reading was obtained via the Internet from
a computer in Pittsburgh, which provides webhosting service, for which I pay a
monthly fee. I wrote this file, and the current archive version resides, on a
computer in my office, which is 750 km from Pittsburgh. Whenever I make a
revision to one of my essays, I upload a new copy from my office computer to the
webhosting computer in Pittsburgh via file transfer protocol (FTP) on the Internet.
Here are some hints about how to make a computer secure from incoming
commands:

Set terminal emulator software so that the modem either

a. never answers an incoming telephone call or

b. answers on the 99th ring and also connect a telephone answering


machine to the modem's line to pick up on the fourth ring, so there is
never a 99th ring.

Do not install any software that allows even an authorized user to access the
computer remotely, via a modem.

more important to have firewall on computers


with wideband Internet connection

During the 1980s and early 1990s, communications between computers in different
places were made by an analog modem that connected to an ordinary voice-grade
telephone line. The state-of-the-art analog modem in 1998 (i.e., using the V.90
standard) could download data at 56000 bits/second and upload data at
36000 bits/second. With compression (i.e., using the V.42bis standard), effective
download data rates of more than 120000 bits/second were possible on an ordinary
voice-grade telephone line. Each time the user wanted to connect to the Internet,
the user would need to have the computer's modem dial the local access number of
the Internet Service Provider (ISP) and establish a connection, a process that takes
about 30 seconds. The ISP then assigns the user a numeric IP address for that one
Internet session. This IP address is known as a "dynamic IP address", because it is
different for each session. If the user does not send/receive some data over the
Internet during some period of time (e.g., ten minutes), the ISP will automatically
disconnect from the telephone line and sever the user's connection to the Internet.

Beginning in the late 1990s, it became common to connect computers to the


Internet via cable television lines or on DSL telephone service, which have a much
higher download data rate, so-called "wideband" service. The cable or DSL service
is always connected to the Internet, unlike the modem on an ordinary voice-grade
telephone line. Specifically, the ISP assigns the user an IP address that is either
constant or rarely changed, a so-called "static IP address".
Cable or DSL makes a user more vulnerable to intrusions by hackers in two
different ways:

1. The static IP address allows a hacker to return to the victim's computer, once
the hacker has found that victim's computer has no firewall or an ineffective
firewall.

2. If a computer is on all the time, hackers have continuous access to that


computer, since cable or DSL is always connected to the Internet.

For these reasons, firewall software is more essential if one uses either cable
television or DSL for an Internet connection.

4. Avoiding harassment
For casual on-line activities, you can establish a free e-mail account at Yahoo,
HotMail, or some other provider, and use an alias for that account. If someone
harasses or stalks you, then you simply close that account and chose another alias.
In other words, you adopt a disposable identity for your life in cyberspace.

Never give out your real name, address, city, telephone number, or other
identifying information to a stranger in a chat room, computer bulletin board, or
other public place.

Obviously, this simple solution has disadvantages. I use the Internet to post essays
and I hope attract clients who need my services as an attorney or scientific
consultant. I wouldn't get many clients if I used an alias like "CyberRocket" or
"Wolverine" at my website. <grin>

Avoiding Phishing

The only connection between phishing and computers is that modern phishing uses
e-mail and a bogus website to get a gullible person to disclose personal financial
information to criminals. That having been said, it is worthwhile to alert people to
the existence of phishing.

People first encounter phishing when they receive a fraudulent e-mail that typically
purports to be from a bank, credit card company, or other financial institution. The
e-mail might mention something about your account is suspended until you
"verify", "update", or "validate" some information. The e-mail invites you to click
on a link in the e-mail. The link typically takes you to a webserver located in a
foreign country and operated by criminals, who display webpages with the logo
and trademarks of a bank, credit card company, or government agency, which
makes the webpage appear legitimate. (It is trivial for people familiar with HTML and
webbrowsers to download some webpages from a legitimate financial institution, make some changes to the
text, then upload those bogus webpages to a criminal's webserver. The bogus webpages really do appear
legitimate because much of each bogus webpage is an exact copy of the real institution's webpage. ) The
bogus webpage asks you to supply your account numbers, passwords, and other
personal information (e.g., Social Security number, date of birth, mother's maiden
name) that can be used to fraudulently access your financial accounts and
perpetrate identity theft crimes.

Citibank has been a popular target for phishers. In response, Citibank posted
a webpage on some specific phishing e-mails. The U.S. Federal Trade
Commission (FTC) also has a consumer alert on the subject of phishing. There is
also an industry antiphishing working group.

5. Backups
If a computer virus or an invading hacker deletes your files, or either one corrupts
your files, the easiest way to restore your computer may be to reformat the hard
drive(s) and then copy files from a recent backup. Backups also offer protection
from more common (and less exotic) threats such as accidental deletion of a file by
an authorized user or failure of a hard disk drive.

Because making a backup is a chore that takes anywhere from a few minutes to
more than an hour (depending on the amount of files copied to the backup medium
and the speed of the backup device), and because backup copies are rarely needed,
most users do not make backups as frequently as they should. The interval between
backups should be determined by the amount of data files that you can afford to
lose.

One should carefully choose a backup medium that will be readable ten or twenty
years in the future. For example, in the mid-1980s, a common backup medium was
either 5 inch floppy disks or less commonly a tape cartridge.

Floppy disk drives for 5 inch floppy disks became obsolete in the early 1990s.
The Windows98 operating system, and subsequent Microsoft operating systems,
have no support for such floppy disks.

There are many different kinds of tape cartridges and different formats for each of
them. If one purchased a tape drive in 1991, and that computer and tape drive were
destroyed in 1998, there is no assurance that one can find a compatible tape drive
in 1998 or 1999 that will read one's "old" tapes.
Since the year 2000 I have used backups to CD-R, and more recently to DVD-R,
optical disks. For archival backups, I use recording speeds of 2X or 4X, and high-
quality disks (e.g. Imation "Business Select" or JVC/Taiyo Yuden disks with Hard
Coat).

There are a number of other methods for making backups that are mentioned in
instruction manuals for backup devices or backup software, or in comprehensive
computer user manuals sold at bookstores. For users with large numbers of files, or
with backups to media with a small capacity, it is useful to make three kinds of
backups:

1. A full backup of all files at least twice each year, and immediately after
completing a major project (e.g., writing a book).

2. An incremental backup of only those files that were changed since the
previous incremental backup. After I make a full backup, I reformat the disk
that I use for incremental backups.

3. An archival backup: a full backup of all files to newly formatted virgin


media. Activate the write-protect feature on this disk or tape cartridge and
never write to it again. Use a permanent marker to label this archival
medium with the date of the backup and name of the computer.

However, most users can simply do a full backup of all of their data files (e.g.,
wordprocessing documents, e-mail, downloaded documents, photographs, etc.)
once or twice a month.

If the only threat was a computer virus or attack by a hacker, it would be adequate
to store all of the backup media in the same room as the computer. Because there
are also threats of fire, burglar, tornado, etc., one should also make a backup for
offsite storage, such as in a safe deposit box at a bank. Archival backups are
particularly well suited to offsite storage, since they are rarely needed.

Most backup strategies involve reusing the media that are used for full and
incremental backups in some kind of rotating scheme. Suppose some problem
(e.g., corruption of file by a computer virus or malfunction of hardware or
software) occurs with old, infrequently used data or document files. One might not
detect the damage to these files until after copies of recently corrupted files had
replaced all of the earlier copies of good files on the rotating backup media. That is
why archival backups are necessary: one can then restore files from an archival
backup medium that was recorded several years ago, prior to any damage.

The cost of backup media is so low (e.g., typically less than US$ 1 for one
recordable CD or DVD) compared to the value of the data, that it is reckless
behavior not to make an archival backup at least several times each year.

6. Other techniques
Because Microsoft bundles an e-mail program, Microsoft Outlook Express, with
their operating system, this e-mail program is commonly used. Hackers write
malicious programs (e.g., the Melissa virus that struck on 26 March 1999) to use
the victim's e-mail address book in Microsoft Outlook, knowing that such a
malicious program will cause havoc on most personal computers, because of the
popularity of Outlook. Similarly, hackers have written macro viruses that affect the
Microsoft Word wordprocessing program. One can avoid being a victim of many
malicious programs simply by refusing to use any Microsoft applications software,
but instead installing software from other companies (e.g., Eudora for e-mail,
WordPerfect for wordprocessing). Avoiding Microsoft applications software makes
sense, because of both features of Microsoft products and the immense popularity
of Microsoft's applications software, which makes Microsoft's application software
a tempting target for hackers who design malicious programs.
Prof. Dorothy Denning, an expert on computer security at Georgetown University,
agrees that avoiding using Microsoft applications software provides some
immunity from malicious programs and she personally follows that practice.
Despite the known vulnerability of Microsoft Outlook to attacks like the Melissa
virus that struck on 26 March 1999 and the ILOVEYOU worm that struck on
4 May 2000, Microsoft waited until 16 May 2000 before beginning to release a
patch to their Outlook program that fixed some defects in security.

While it is possible to write malicious programs to attack users of the Apple


Macintosh operating system, only about 5% of desktop computers run the Apple
Macintosh operating system, which small market share discourages hackers.

In the Windows98 operating system, the default location for most applications
programs is a subdirectory inside C:/Program Files. A hacker designing a
malicious program to use a victim's e-mail address book or document files could
simply assume that the victim used the default location for these files. One might
frustrate such hackers by installing applications programs in a nonstandard
location, for example, put WordPerfect in D:/wordproc and put Eudora
in D:/mailprgm. (An additional benefit of partitioning a disk drive into C: D: E: before installing the
Windows operating system is that one can put the operating system on the C: drive, all applications software on
the D: drive, and all data files (including the disk cache for the Netscape browser) on the E: drive. One can make
daily backups of the E: drive, because files change or are added to the E: drive every day that the computer is
used, and less frequent backups of the C: and D: drives.)

Declare your computer and software on your homeowner's or business property


insurance policy.

Use the latest version of e-mail and Internet browser software, operating system,
and anti-virus software. Install the patches that are released between versions, to
avoid security holes and other problems.

Before disposing of an old computer, by donation or sale:

1. delete all data and document files.

2. delete all application programs (to avoid software piracy).

3. run WIPEINFO, in Norton Utility for DOS version 8 (or the corresponding
version for the Apple Macintosh), to overwrite all of the free space on the
hard disk, thus making it difficult to recover your data and document files.

WIPEINFO is necessary, because the delete command only changes the first
character of the file name in the file allocation table (FAT) on the disk; the delete
command does not remove the information in the file from the disk.

If you are tossing an old computer or hard disk drive in the trash, first disable the
hard disk drive. You can use a hammer and chisel to remove some of the integrated
circuits from the disk controller that is attached to the hard drive, or you can
remove the hard drive from the computer and smash it with a hammer.

The design of a logon screen should include a notice that unauthorized use is
prohibited by law. One might refer to the state (or federal) statute and mention the
maximum penalties, in an attempt to deter people. On the other hand, making the
notice too strong (i.e., reference to secret, proprietary, or private information inside
the computer) may be a double-edged sword, in that it may entice a hacker by
increasing the thrill.

Disabling Features in Microsoft Windows

Some optional parts of the Microsoft Windows 95, and later, operating systems
make a computer vulnerable to harm by malicious programs. For detailed
instructions on how to disable some of these parts, see:

The Computer Emergency Response Team (CERT) at Carnegie-Mellon


University has links to their webpages on showing normally hidden file
extensions, disabling JavaScript in web browsers, and disabling ActiveX.
F-Secure, an anti-virus software company in Finland, has instructions for
how to uninstall Windows Scripting Host in Windows 95, 98, 2000 and NT.
This uninstall provides immunity from any malicious program that uses
Microsoft Visual Basic Script, such as the ILOVEYOU worm. Symantec,
who distributes Norton Anti-Virus software, has similar instructions: go to
their search webpage, and enter the search query "uninstall Windows
Scripting Host".

There are two separate issues:

1. There have been an alarming number of security flaws in both Microsoft


Windows operating systems and Microsoft applications software (e.g.,
Internet Explorer web browser, Outlook e-mail program), as well as
macro viruses for Microsoft Word.

2. Prof. Fred Schneider in the Computer Science Department at Cornell


University speaks of a "monoculture". When nearly everyone uses the same
operating system and same applications, then nearly everyone is vulnerable
to the same viruses and worms, which makes these viruses and worms
propagate more rapidly. A monoculture exacerbates security vulnerabilities
in software, as well as stifles competition (i.e., the usual economic
arguments against monopolies).

I have never used any Microsoft applications software. And, in May 2002, with the
above two points in mind, I abandoned the Microsoft Windows operating system
too. Instead, I now use an Apple computer for my websurfing and e-mail.

7. Wireless Networks
A local area network is popular in businesses, because it allows computers to share
files without using a modem and because it allows multiple computers to use a
single printer. Conventional local area networks in the 1980s required coaxial cable
to be strung between computers and printers on the network. Recently, it has
become popular to use a so-called "wireless network", in which computers and
peripherals communicate via low-power radio transmissions.

However, wireless networks have an obvious security problem. Radio


transmissions do not magically stop at the exterior walls of the building. Anyone
with an antenna and amplifier can intercept communications on a
wireless network, which raises the possibility of both privacy violations and
industrial espionage. It is easy to imagine someone sitting in a van in a parking lot
with a high-gain Yagi antenna and a laptop computer, intercepting communications
from a wireless network inside a nearby building.
If you want to use a wireless network, at least:

enable encryption on all transmissions and

locate the transmitters as far as possible from exterior walls, and especially
far from windows.

Personally, I refuse to use wireless networks, as I don't want to risk my clients'


confidential data on such networks.

SpyWare
In the year 2002, a new kind of malicious computer program appeared, which is
automatically installed when one visits certain websites (e.g., file-sharing services),
click on some pop-up adverts, or click on some attachments in e-mail.

The more benign spyware programs track the websites that you visit and send you
adverts that are considered appropriate for your interests, which is an invasion of
your privacy.

The worst of the spyware programs record your keystrokes, particularly looking for
your passwords, user names, credit card numbers, and other confidential
information. Your information is then available to whoever controls the spyware.

There are many computer programs that are designed to search your computer and
remove spyware. Unfortunately, reviews in computer magazines show that none of
these anti-spyware programs are successful in removing all known spyware.

A better way to avoid spyware is simply to chose an operating system that


is not 32-bit Windows (i.e., avoid Microsoft Windows 95/98/NT/Me/2000/XP/...).

8. Links to Other Sites


I have moved the links that were formerly here to a separate webpage.

I no longer provide links to sources of anti-virus software, firewall software, and


spyblocker software for the Microsoft Windows operating systems. Every
computer that is connected (even temporarily) to the Internet should
have both current anti-virus software and firewall software installed.
Nearly all computer viruses, worms, and spyware target operating systems in 32-
bit Windows family (e.g., Microsoft Windows 95/98/NT/Me/2000/XP/...).
Choosing an operating system that is not 32-bit Windows gives immunity from
most of these viruses and worms.

I have moved the links for anti-virus software and firewall software for the Apple
computer to a webpage at my personal website: www.rbs0.com/apple.htm.

Who Is?

After you install a firewall, you will probably be curious about who is probing the
ports on your computer. You can type either the numeric IP address or the domain
name into one of the WhoIs databases.

recommended books
Dorothy E. Denning, Information Warfare and Security, Addison-Wesley
Publishing Co., 1998.

Dorothy E. Denning and Peter J. Denning, Internet Besieged, Addison-Wesley


Publishing Co., 1997.

Peter J. Denning (editor), Computers Under Attack, Addison-Wesley Publishing


Co., 1990.
A collection of reprinted articles from computer software journals, mostly from the
1980s.

Peter G. Neumann, Computer-Related Risks, Addison-Wesley Publishing Co.,


1995.
A collection of terse anecdotal reports in book format.
Dr. Neumann moderates the online Forum on Risks, the current issue of which is
posted at SRI's website. An archive of issues back to 1 Aug 1985 is available
at Lindsay Marshall's website at the University of Newcastle upon Tyne.

True stories of detecting hackers:

1. Cliff Stoll, Cuckoo's Egg, 1990. (Pursuing a German hacker, who conducted
espionage for the KGB.)
2. Tsutomu Shimomura, Takedown, 1996. (Pursuing Kevin Mitnick, a fugitive
career computer criminal who made the mistake of hacking into
Shimomura's personal computer.)

Conclusion
When a criminal perpetrates a crime, his attorney is likely to say that the criminal
did everyone a favor by calling attention to lapses in security of computers. It is a
criminal defense attorney's job to put the best possible spin on the client's horrible
activities. However, recognize that "blaming the victim" for the crime is a cheap
shot. Even if the victim behaved in an imprudent way, a victim never invites a
crime.

I want to make clear that there are two completely separate issues: (1) prosecution
of perpetrators of computer crimes and (2) steps that a computer user can take to
avoid being a victim.

Most of the really effective steps that a computer user can take to avoid being a
victim of crime make the user's computer less convenient to use. Each user must
balance for himself/herself how much security is enough, especially when faced
with daily inconvenience of high-level security measures vs. the rare occurrence of
attacks. Further, the user must be aware that a determined and creative criminal can
defeat nearly any security measure, so complete security is not possible.

this document is at http://www.rbs2.com/cvict.htm


created May 1999, last revision 27 Aug 2004, links updated 25 Nov 2007, backup
section revised 3 Aug 2013.

My essay on computer crime.

return to my homepage

You might also like