E-mail spam slowly but exponentially grew for several decades to several billion messages a day. Spam has frustrated, confused, and annoyed e-mail users. Laws against spam have been sporadically implemented, with some being opt-out and others requiring opt in e-mail. The total volume of spam (over 100 billion emails per day as of April 2008) has leveled off slightly in recent years, and is no longer growing exponentially. The amount received by most e-mail users has decreased, mostly because of better filtering. About 80% of all spam is sent by fewer than 200 spammers. Botnets, networks of virus-infected computers, are used to send about 80% of spam. The cost of spam is borne mostly by the recipient, so it is a form of postage due advertising.
E-mail addresses are collected from chatrooms, websites, newsgroups, and viruses which harvest users' address books, and are sold to other spammers. Much of spam is sent to invalid e-mail addresses. ISPs have attempted to recover the cost of spam through lawsuits against spammers, although they have been mostly unsuccessful in collecting damages despite winning in court.
|E-Mail Spam by Category|
Mainsleaze is all but non-existent, as well-known companies do not wish to be associated with spam.
If a marketer has one database containing names, addresses, and telephone numbers of prospective customers, they can pay to have their database matched against a database containing email addresses. If the database contains erroneous information, companies will send out unsolicited commercial email. To protect against this, e-mail should only be sent to users known to have subscribed, and should be verified through a double opt in procedure. Purchasing a list of e-mail addresses that match a list of names is almost guaranteed to create unhappy recipients.
As the recipient directly bears the cost of delivery, storage, and processing, one could regard spam as the electronic equivalent of "postage-due" junk mail. Due to the low cost of sending unsolicited e-mail and the potential profit entailed, some believe that only strict legal enforcement can stop junk e-mail. The Coalition Against Unsolicited Commercial Email (CAUCE) argues "Today, much of the spam volume is sent by career criminals and malicious hackers who won't stop until they're all rounded up and put in jail.
Spam is legally permissible according to the CAN-SPAM Act of 2003 provided it follows certain criteria: a truthful subject line; no false information in the technical headers or sender address; "conspicuous" display of the postal address of the sender; and other minor requirements. If the spam fails to comply with any of these requirements, then it is illegal. Aggravated or accelerated penalties apply if the spammer harvested the email addresses using methods described earlier.
A review of the effectiveness of CAN-SPAM in 2005 showed that the amount of sexually explicit spam had significantly decreased since 2003 and the total volume had begun to level off. Senator Conrad Burns, a principle sponsor, noted that "Enforcement is key regarding the CAN-SPAM legislation." In 2004 less than 1% of spam complied with the CAN-SPAM Act of 2003.
Article 13 of the European Union Directive on Privacy and Electronic Communications (2002/58/EC) provides that the EU member states shall take appropriate measures to ensure that unsolicited communications for the purposes of direct marketing are not allowed either without the consent of the subscribers concerned or in respect of subscribers who do not wish to receive these communications, the choice between these options to be determined by national legislation.
In Australia, the relevant legislation is the Spam Act 2003 which covers some types of e-mail and phone spam, which took effect on 11 April 2004. The Spam Act provides that "Unsolicited commercial electronic messages must not be sent," which is an opt-in requirement. This contrasts with the U.S. CAN-SPAM act, which is opt-out (i.e., companies are free to send spam until the recipient directs the sender not to). Penalties are up to 10,000 penalty units, or 2,000 penalty units for a person other than a body corporate.
Legislative efforts to curb spam have been ineffective or counter-productive. For example, the CAN-SPAM Act of 2003 requires that each message include a means to "opt out" (i.e., decline future e-mail from the same source). It is widely believed that responding to opt-out requests is unwise, as this merely confirms to the spammer that they have reached an active e-mail account. To the extent this is true, the CAN-SPAM Act's opt-out provisions are counter-productive in two ways: first, recipients who are aware of the potential risks of opting out will decline to do so; second, attempts to opt-out will provide spammers with useful information on their targets. A 2002 study by the Center for Democracy and Technology found that about 16% of web sites tested with opt-out requests continued to spam.
Accessing privately owned computer resources without the owner's permission counts as illegal under computer crime statutes in most nations. Deliberate spreading of computer viruses is also illegal in the United States and elsewhere. Thus, some common behaviors of spammers are criminal regardless of the legality of spamming per se. Even before the advent of laws specifically banning or regulating spamming, spammers were successfully prosecuted under computer fraud and abuse laws for wrongfully using others' computers.
The use of botnets can be perceived as theft. The spammer consumes a zombie owner's bandwidth and resources without any cost. In addition, spam is perceived as theft of services. The receiving SMTP servers consume significant amounts of system resources dealing with this unwanted traffic. As a result, service providers have to spend large amounts of money to make their systems capable of handling these amounts of email. Such costs are inevitably passed on to the service providers' customers.
Other laws, not only those related to spam, have been used to prosecute alleged spammers. For example, Alan Ralsky was indicted on stock fraud charges in January 2008, and Robert Soloway plead guilty to charges of mail fraud, fraud in connection with electronic mail, and failing to file a tax return in March 2008.
Senders may go to great lengths to conceal the origin of their messages. Large companies may hire another firm to send their messages so that complaints or blocking of email falls on a third party. Others engage in spoofing of e-mail addresses (much easier than IP address spoofing). The e-mail protocol (SMTP) has no authentication by default, so the spammer can pretend to originate a message apparently from any e-mail address. To prevent this, some ISPs and domains require the use of SMTP-AUTH, allowing positive identification of the specific account from which an e-mail originates.
Senders cannot completely spoof e-mail delivery chains (the 'Received' header), since the receiving mailserver records the actual connection from the last mailserver's IP address. To counter this, some spammers forge additional delivery headers to make it appear as if the e-mail had previously traversed many legitimate servers.
Spoofing can have serious consequences for legitimate e-mail users. Not only can their e-mail inboxes get clogged up with "undeliverable" e-mails in addition to volumes of spam, they can mistakenly be identified as a spammer. Not only may they receive irate e-mail from spam victims, but (if spam victims report the e-mail address owner to the ISP, for example) a naive ISP may terminate their service for spamming.
Increasingly, spammers use networks of malware-infected PCs (zombies) to send their spam. Zombie networks are also known as Botnets (such zombifying malware is known as a bot, short for robot). In June 2006, an estimated 80% of e-mail spam was sent by zombie PCs, an increase of 30% from the prior year. An estimated 55 billion e-mail spam were sent each day in June 2006, an increase of 25 billion per day from June 2005.
In terms of volume of spam: According to Sophos, the major sources of spam in the first quarter of 2008 (January to March) were:
When grouped by continents, spam comes mostly from:
In terms of number of IP addresses: The Spamhaus Project (which measures spam sources in terms of number of IP addresses used for spamming, rather than volume of spam sent) ranks the top three as the United States, China, and Russia, followed by Japan, Canada, and South Korea.
In terms of networks: As of 5 June 2007, the three networks hosting the most spammers are Verizon, AT&T, and VSNL International. Verizon inherited many of these spam sources from its acquisition of MCI, specifically through the UUNet subsidiary of MCI, which Verizon subsequently renamed Verizon Business.
The U.S. Department of Energy Computer Incident Advisory Capability (CIAC) has provided specific countermeasures against electronic mail spamming.
Some popular methods for filtering and refusing spam include e-mail filtering based on the content of the e-mail, DNS-based blackhole lists (DNSBL), greylisting, spamtraps, Enforcing technical requirements of e-mail (SMTP), checksumming systems to detect bulk email, and by putting some sort of cost on the sender via a Proof-of-work system or a micropayment. Each method has strengths and weaknesses and each is controversial due to its weaknesses.
One method employed involves using a white list of email addresses. For example, the owner of the email account can set the server to only allow emails from senders that are in the owner's addressbook. This is often used in combination with methods to give new senders an opportunity to request inclusion in the owner's addressbook. For example: 1. All email originating from senders not in the addressbook are sent an automatic response stating that their email has not reached the recipient (ie. the email account owner). 2. The sender is given the option of sending the recipient an addressbook inclusion request via an online form. 3. The online form includes a captcha to only allow requests from human (non-computer-automated) sources. 4. If the recipient (account owner) approves the request, current and future email from the sender reaches the recipient with no further filtering. Yahoo dropped a similar feature from their webmail service in 2005.
Anti-spam techniques should not be employed on abuse@ email addresses, as is commonly the case. The result of this is that when people attempt to report spam to a host, the spam message is caught in the spam filter and the host remains unaware that their network is being exploited by spammers.
In order to send spam, spammers need to obtain the e-mail addresses of the intended recipients. To this end, both spammers themselves and list merchants gather huge lists of potential e-mail addresses. Since spam is, by definition, unsolicited, this address harvesting is done without the consent (and sometimes against the expressed will) of the address owners. As a consequence, spammers' address lists are inaccurate. A single spam run may target tens of millions of possible addresses — many of which are invalid, malformed, or undeliverable.
Sometimes, if the sent spam is "bounced" or sent back to the sender by various programs that eliminate spam, or if the recipient clicks on an unsubscribe link, that may cause that email address to be marked as "valid", which is interpreted by the spammer as "send me more".
In an effort to cut down on this abuse, many of these services have adopted a system called the captcha: users attempting to create a new account are presented with a graphic of a word, which uses a strange font, on a difficult to read background. Humans are able to read these graphics, and are required to enter the word to complete the application for a new account, while computers are unable to get accurate readings of the words using standard OCR techniques. Blind users of captchas typically get an audio sample.
Spammers have, however, found a means of circumventing this measure. Reportedly, they have set up sites offering free pornography: to get access to the site, a user displays a graphic from one of these webmail sites, and must enter the word. Once the bot has successfully created the account, the user gains access to the pornographic material. Furthermore, standard image processing techniques work well against many captchas.
The standard was written in an era before spamming when there were few hosts on the internet, and those on the internet abided by a certain level of conduct. While this cooperative, open approach was useful in ensuring that mail was delivered, it was vulnerable to abuse by spammers. Spammers could forward batches of spam through open relays, leaving the job of delivering the messages up to the relays.
In response, mail system administrators concerned about spam began to demand that other mail operators configure MTAs to cease being open relays. The first DNSBLs, such as MAPS RBL and the now-defunct ORBS, aimed chiefly at allowing mail sites to refuse mail from known open relays. By 2003 less than 1% of corporate mail servers were available as open relays, down from 91% in 1997.
An open proxy is one which will create connections for any client to any server, without authentication. Like open relays, open proxies were once relatively common, as many administrators did not see a need to restrict access to them.
A spammer can direct an open proxy to connect to a mail server, and send spam through it. The mail server logs a connection from the proxy -- not the spammer's own computer. This provides an even greater degree of concealment for the spammer than an open relay, since most relays log the client address in the headers of messages they pass. Open proxies have also been used to conceal the sources of attacks against other services besides mail, such as Web sites or IRC servers.
Besides relays and proxies, spammers have used other insecure services to send spam. One example is FormMail.pl, a CGI script to allow Web-site users to send e-mail feedback from an HTML form. Several versions of this program, and others like it, allowed the user to redirect e-mail to arbitrary addresses. Spam sent through open FormMail scripts is frequently marked by the program's characteristic opening line: "Below is the result of your feedback form."
As spam from proxies and other "spammable" resources grew, DNSBL operators started listing their IP addresses, as well as open relays.
Most of the major Windows e-mail viruses of 2003, including the Sobig and Mimail virus families, functioned as spammer viruses: viruses designed expressly to make infected computers available as spamming tools.
Besides sending spam, spammer viruses serve spammers in other ways. Beginning in July 2003, spammers started using some of these same viruses to perpetrate distributed denial-of-service (DDoS) attacks upon DNSBLs and other anti-spam resources. Although this was by no means the first time that illegal attacks have been used against anti-spam sites, it was perhaps the first wave of effective attacks.
In August of that year, engineering company Osirusoft ceased providing DNSBL mirrors of the SPEWS and other blocklists, after several days of unceasing attack from virus-infected hosts. The very next month, DNSBL operator Monkeys.com succumbed to the attacks as well. Other DNSBL operators, such as Spamhaus, have deployed global mirroring and other anti-DDoS methods to resist these attacks.
Zombie networks are particularly active in North America where about half of the Internet users are on a broadband connection and many leave their computers on all the time. In January, 2008, 8% of all e-mail spam was sent by the Storm botnet, created by the Storm Worm, first released in January, 2007. It is estimated that as many as 1 million or more computers have been infected and their owners are unwilling and unknowing participants.
The principle of this method is to leave the word readable to humans (who can easily recognize the intended word for such misspellings), but not likely to be recognized by a literal computer program. This is only somewhat effective, because modern filter patterns have been designed to recognize blacklisted terms in the various iterations of misspelling. Other filters target the actual obfuscation methods; such as the non-standard use of punctuation or numerals into unusual places, for example: within in a word.
(Note: Using most common variations, it is possible to spell "Viagra" in over 1.3 * 1021 ways.)
HTML-based e-mail gives the spammer more tools to obfuscate text. Inserting HTML comments between letters can foil some filters, as can including text made invisible by setting the font color to white on a white background, or shrinking the font size to the smallest fine print.
Another common ploy involves presenting the text as an image, which is either sent along or loaded from a remote server. This can be foiled by not permitting an e-mail-program to load images.
As Bayesian filtering has become popular as a spam-filtering technique, spammers have started using methods to weaken it. To a rough approximation, Bayesian filters rely on word probabilities. If a message contains many words which are only used in spam, and few which are never used in spam, it is likely to be spam. To weaken Bayesian filters, some spammers, alongside the sales pitch, now include lines of irrelevant, random words, in a technique known as Bayesian poisoning. A variant on this tactic may be borrowed from the Usenet abuser known as "Hipcrime" -- to include passages from books taken from Project Gutenberg, or nonsense sentences generated with "dissociated press" algorithms. Randomly generated phrases can create spoetry (spam poetry) or spam art.
After these nonsense subject lines were recognized as spam, the next trend in spam subjects started: Biblical passages. A program much like Mark V Shaney is fed Bible passages and chops them up into segments. The reasoning is that this text, often very different from the writing style of today such as the King James Version, will confuse both humans and spam filters.
However, as many or most Bayesian filtering programs only use the most spam-like and least spam-like words for deciding whether an email is spam or not; injecting extra non-spam related words means that these extra words do not correlate well with spam, and so do not usually affect the result. However, they do decrease the effectiveness slightly, which, for spammers can make a significant percentage difference in the number of users actually seeing their spam.
Another method used to masquerade spam as legitimate messages is the use of autogenerated sender names in the From: field, ranging from realistic ones such as "Jackie F. Bird" to (either by mistake or intentionally) bizarre attention-grabbing names such as "Sloppiest U. Epiglottis" or "Attentively E. Behavioral". Return addresses are also routinely auto-generated, often using unsuspecting domain owners' legitimate domain names, leading some users to blame the innocent domain owners. Blocking lists use ip addresses rather than sender domain names, as these are more accurate. A mail purporting to be from example.com can be seen to be faked by looking for the originating ip address in the mails header, and Sender Policy Framework for example helps by stating that example.com will only send email from xx.xx.xx.xx ip.
Some Internet hosting firms advertise bulk-friendly or bulletproof hosting. This means that, unlike most ISPs, they will not terminate a customer for spamming. These hosting firms operate as clients of larger ISPs, and many have eventually been taken offline by these larger ISPs as a result of complaints regarding spam activity. Thus, while a firm may advertise bulletproof hosting, it is ultimately unable to deliver without the connivance of its upstream ISP. However, some spammers have managed to get what is called a pink contract (see below) — a contract with the ISP that allows them to spam without being disconnected.
A few companies produce spamware, or software designed for spammers. Spamware varies widely, but may include the ability to import thousands of addresses, to generate random addresses, to insert fraudulent headers into messages, to use dozens or hundreds of mail servers simultaneously, and to make use of open relays. The sale of spamware is illegal in eight U.S. states.
So-called millions CDs are commonly advertised in spam. These are CD-ROMs purportedly containing lists of e-mail addresses, for use in sending spam to these addresses. Such lists are also sold directly online, frequently with the false claim that the owners of the listed addresses have requested (or "opted in") to be included. Such lists often contain invalid addresses. In recent years, these have fallen almost entirely out of use due to the low quality e-mail addresses available on them, and because some e-mail lists exceed 20GB in size. The amount you can fit on a CD is no longer substantial.
A number of DNSBLs, including the MAPS RBL, Spamhaus SBL, SORBS and SPEWS, target the providers of spam-support services as well as spammers. DNSBLs blacklist IPs or ranges of IPs to persuade ISPs to terminate services with known customers who are spammers or resell to spammers.
On September 27, 2004, Nicholas Tombros plead guilty to charges and became the first spammer to be convicted under the CAN-SPAM Act of 2003. He was sentenced in July 2007 to three years probation, six months house arrest, and fined $10,000.
On November 4, 2004, Jeremy Jaynes, rated the 8th most prolific spammer in the world according to Spamhaus, was convicted of three felony charges of using servers in Virginia to send thousands of fraudulent e-mails. The court recommended a sentence of nine years' imprisonment, which was imposed in April 2005 although the start of the sentence was deferred pending appeals. Jaynes claimed to have an income of $750,000 a month from his spamming activities. On February 29, 2008 the Supreme Court of Virginia affirmed his conviction.
On December 31, 2004, British authorities arrested Christopher Pierson in Lincolnshire, UK and charged him with malicious communication and causing a public nuisance. On January 3, 2005, he pleaded guilty to sending hoax e-mails to relatives of people missing following the Asian tsunami disaster.
On November 1, 2005, David Levi, 29, of Lytham, England was sentenced to four years for conspiracy to defraud by sending e-mails pretending to be from eBay, his brother Guy Levi, 22, was sentenced to 21 months after pleading guilty to conspiracy to defraud, and four others were each sentenced to six months for money laundering.
On June 28, 2006, IronPort released a study which found 80% of spam emails originating from zombie computers. The report also found 55 billion daily spam emails in June 2006, a large increase from 35 billion daily spam emails in June 2005. The study used SenderData which represents 25% of global email traffic and data from over 100,000 ISP's, universities, and corporations.
On August 8, 2006, AOL announced the intention of digging up the garden of the parents of spammer Davis Wolfgang Hawke in search of buried gold and platinum. AOL had been awarded a US$ 12.8 million judgment in May 2005 against Hawke, who had gone into hiding. The permission for the search was granted by a judge after AOL proved that the spammer had bought large amounts of gold and platinum. In July, 2007 AOL decided not to proceed.
On October 12, 2006, Brian Michael McMullen, 22, of East Pittsburgh, Pennsylvania was sentenced to three years supervised release, five months home detention and ordered to pay restitution in the amount of $11,848.55 for violating the CAN-SPAM Act of 2003.
On October 27, 2006, the Federal Court of Australia fined Clarity1 A$4.5 million (US$3.4 million; euro2.7 million) and its director Wayne Mansfield A$1 million (US$760,000; euro600,000) for sending unsolicited e-mails in the first conviction under Australia's Spam Act of 2003.
In November, 2006 Christopher William Smith (aka Chris "Rizler" Smith) was convicted on 9 counts for offenses related to Smith's spamming.
On May 30, 2007, notorious spammer Robert Soloway was arrested after having being indicted by a federal grand jury on 35 charges including mail fraud, wire fraud, e-mail fraud, identity theft, and money laundering. If convicted, he could face decades behind bars. Bail was initially denied although he was released to a half way house in September. On March 14, 2008, Robert Soloway reached an agreement with federal prosecutors, two weeks before his scheduled trial on 40 charges. Soloway pleaded guilty to three charges — felony mail fraud, fraud in connection with e-mail, and failing to file a 2005 tax return. In exchange, federal prosecutors dropped all other charges. Soloway faced up to 26 years in prison on the most serious charge, and up to $625,000 total in fines. On 22 July 2008 Robert Soloway was sentenced four years in federal prison, a ruling that sends a message to other spammers and online criminals.
On June 25, 2007 two men were each convicted on eight counts including conspiracy, fraud, money laundering, and transportation of obscene materials in U.S. District Court in Phoenix, Arizona. The prosecution is the first of its kind under the CAN-SPAM Act of 2003, according to a release from the Department of Justice. One count for each under the act was for falsifying headers, the other was for using domain names registered with false information. The two had been sending millions of hard-core pornography spam e-mails. The two men were sentenced to five years in prison and ordered to forfeit US$ 1.3 million.
August 19: A survey on Marshall Limited's website showed that 29.1% of the 622 respondents had bought something from a spam email. Other studies, one by Forrester Research in 2004, which surveyed 6,000 active Web users, reported 20 percent had bought something from spam, while a 2005 study by Mirapoint and the Radicati Group showed 11%, and 57% indicated that clicking on a link in spam caused them to receive more spam than before. A 2007 study by Endai Worldwide (an e-mail marketing company) showed 16% had bought something from spam. In response to the Marshall study, the Download Squad started their own study. With 289 respondents, only 2.1% indicated they had ever bought something from a spam e-mail.
Often, image spam contains nonsensical, computer-generated text which simply annoys the reader. However, new technology in some programs try to read the images by attempting to find text in these images. They are not very accurate, and sometimes filter out innocent images of products like a box that has words on it.
A newer technique, however, is to use an animated GIF image that does not contain clear text in its initial frame, or to contort the shapes of letters in the image (as in CAPTCHA) to avoid detection by OCR tools.
Blank spam may be originated in different ways, either intentional or unintentionally: