Brad Templeton Home
Origin of the term "spam" to mean net abuse
Much to the chagrin of Hormel Foods, maker of the canned "Shoulder Pork and hAM"/"SPiced hAM" luncheon meat, the term "spam" has today come to mean network abuse, particularly junk E-mail and massive junk postings to USENET.
(See also my reflections on the 25th anniversary of spam.)
How did the term get this meaning? I went on a mission of etymological research. In this article you'll learn how the term, born of canned ham, moved into BBSs and MUDS and then was applied to USENET postings and E-mail. I've put in a short history of the earliest big spams, including a special page about the first E-mail spam from 1978. (You'll be astounded to see which net celebrity defends the spam. But we were all younger then.)
(This is an interesting time for spammiversaries. March 31st, 2003 marks the 10th anniversary of the term Spam being applied to a USENET post, and May 3rd marks the 25th anniversary of the earliest documented E-mail spam.)
Plus at the bottom, a bit about the term surfing the net.
Most people have some vague awareness that it came from at first from the spam skit by Monty Python's Flying Circus. In the sketch, a restaurant serves all its food with lots of spam, and the waitress repeats the word several times in describing how much spam is in the items. When she does this, a group of Vikings (don't ask) in the corner start a song:
"Spam, spam, spam, spam, spam, spam, spam, spam, lovely spam! Wonderful spam!"
Until told to shut up.
Thus the meaning of the term at least: something that keeps repeating and repeating to great annoyance. How did the two get connected?
Canter and Siegel
In April of 1994, the term was not born, but it did jump a great deal in popularity when two lawyers from Phoenix named Canter and Siegel posted a message advertising their fairly useless services in an upcoming U.S. "green card" lottery. This wasn't the first such abusive posting, nor the first mass posting to be called a spam, but it was the first deliberate mass posting to commonly get that name. They had posted their message a few times before, but on April 12, they hired an mercenary programmer to write a simple script to post their ad to every single newsgroup (message board) on USENET, the world's largest online conferencing system. There were several thousand such newsgroups, and each one got the ad.
Quickly people identified it as "spam" and the word caught on. Future multiple postings soon got the appelation. Some people also applied it to individual unwanted ads that weren't posted again and again, though generally it was associated with the massive flood of the same message. It turns out, however, that the term had been in use for some time before the famous green card flood.
Later, some particularly nasty folks figured they could take mass e-mailing software (which had been around for decades to handle mailing lists) and use it to send junk e-mail to large audiences who hadn't asked for it. The term quickly came to be used to describe these unwanted junk e-mails, and indeed that is the most common use of the term today.
The "green card" spam didn't coin the term, and it wasn't even the first "spam" -- though it was the first really large commercial one.
My research shows the term goes back to the late 1980s and the "MUD" community.
A MUD is a multi-user-dungeon. That's a somewhat archaic term for a real time multi-person shared environment, which is to say a shared world where users can chat, move around and interact with locations and objects in the environment. MUDs were named that because the first reminded people of "adventure" or "Dungeons and Dragons" games that involved jointly exploring a cave or dungeon. Modern successors of the MUD include EverQuest, World of Warcraft and The Sims Online.
But most people used MUDs to chat, and to play around and impress one another with objects they created. They were at first a highly evolved successor for the chat room.
The term spamming got used to apply to a few different behaviours. One was to flood the computer with so much data as to crash it. Another was to "spam the database" by having a program create a huge number of objects, rather then creating them by hand. And the term was sometimes used to mean simply flooding a chat session with a bunch of text inserted by a program (commonly called a "bot" today) or just by inserting a file instead of your own real time typing output.
There are unconfirmed reports as well that the term migrated to MUDs from early "chat" systems. Rich Frueh believes the term originated on Bitnet's Relay, the early chat system that IRC (Internet Relay Chat) was named after. When the ability to input a whole file to the chat system was implemented, people would annoy others by dumping the words to the Monty Python Spam Song. Peter da Silva reports use in early 80s chat on TRS-80 based BBSs, but feels since they imported other Bitnet Relay customs, the term may have come from there. Another unconfirmed report from a BBS user claims to have seen it defined as a "Single Post to All Messagebases" though this origin seems highly unlikely in my judgement.
Another report describes indirectly a person simply typing "spam, spam..." in a MUD with a keyboard macro until being thrown off around 1985.
(Another report by a blogger that she believes she observed the birth of the term spam in AOL Star Trek chat rooms turns out to date from 94-95 and thus postdates the word entering common usage via other channels.)
My research has not found BBSers or Relay chatters using the term in USENET messages, so for now I conclude it was MUDders who brought the term to USENET and email.
Here we see a thread where MUDders discuss the meaning and origin of the word Spam in 1990.
After the MUDders used the term, from time to time it would be used to describe a net abuse, but this usage was fairly rare. People far more commonly talked of the luncheon meat. It wasn't a bad term at the time, and many people liked to use it in site names, userids, signatures etc.
1978: The first internet E-mail spam, sent by DEC
Einar Stefferud, a longtime net hand, reports that DEC announced a new DEC-20 machine in 1978 by sending an invite to all ARPANET addresses on the west coast, using the ARPANET directory, inviting people to receptions in California. They were chastised for breaking the ARPANET appropriate use policy, and a notice was sent out reminding others of the rule.
I have put up a page with the message and its reaction or you can see it directly in the msggroup archives (if they come back online.) It may amuse some to see a young Richard Stallman as one of the defenders of the spam!
Compaq now owns DEC, so perhaps Bill Hewlett and Dave Packard's heirs are simply among those who vowed never to do business with spammers? Of course in 1978 nobody called this a spam.
Even earlier, non-network spam
Tom Van Vleck, co-author of the CTSS MAIL command, reports an even earlier spam sent on MIT's Compatbile Time Sharing System (CTSS) as far back as 1971. A sysadmin named Peter Bos used CTSS MAIL to send everybody the anti-war message that: "THERE IS NO WAY TO PEACE. PEACE IS THE WAY." He reports the spammer defended it by saying, "but this is important." He was also an authorized admin, so this one is somewhat harder to classify. A great history of early mail systems provides more details.
And people were thinking about the problem even before it took place. In 1975, RFC706 by Jon Postel mused presciently on the issue.
email@example.com says: HELP ME!
Rob Noha, using the account JJ@cup.portal.com posted on May 24, 1988 to as many newsgroups as he could find a plea to HELP ME! and send money to his college fund, as he was, he claimed, running out.
This, the first USENET spam was, in a sense, a charity spam. Or perhaps charity 'scam' is a better word.
JJ posted to many groups, but each post was crossposted to 4 or 5 newsgroups, to reduce the total volume, but still annoy.
However, nobody ever used the term "spam" to refer to this posting until 1996. It did spark a lot of debate about the merits of having sites like Portal, which sold accounts to anybody with a few dollars. This somewhat elitist viewpoint got even more prevalent when AOL was poised to join the fray.
Make Money Fast
For many years, the classic unwanted post on USENET was the "Dave Rhodes" "MAKE MONEY FAST" post. This was a pretty standard chain letter, but a lot of people started sending it to mailing lists and posting it to newsgroups for no good reason. Most were roundly chastised. Some ran away and were never heard from again. Others learned to play well with others.
However, while there were many "MAKE MONEY FAST" postings in the early 90s and even the 80s, they were usually one-off postings, each one by a different person. Thus they weren't called spam (until after ARMM, below.)
ARMM -- first to be called spam.
In 1993, Richard Depew tried to make some changes in USENET. He advanced a somewhat controversial idea called retro-moderation, where newsgroups would be semi-moderated (that is to say, regulated so that not all postings would appear) through a moderator who canceled postings that broke the rules.
Moderated groups were common, but with those, each posting was pre-screened by the moderator. He suggested allowing a moderator to screen after the fact. Because this was new, and reminded people of censorship (since canceling the postings of others was normally a major faux pas), this was quite controversial. Depew got his own fan group, and both allies and detractors.
However, what really got people upset was March 31, 1993. He had been playing with some software to perform the retro-moderation task. His software, called ARMM, had a bug, and he ran it, and it ended up posting 200 messages in a row to news.admin.policy, the newsgroup where people discussed the running of the net. You can see an example here.
It really ticked people off, and some people, knowing the term from MUDs, called it a spam. The very day ARMM was run, Joel Furr, as far as I can tell, was the first to call a spam a spam.
Depew himself shortly apologized for having done a spam, using the term himself. However, many would say that this wasn't a true spam, since it was an accident.
First Giant Spam
The first major USENET spam came on January 18 of 1994. Every single newsgroup found in it a religious screed declaring: Global Alert for All: Jesus is Coming Soon.
This one caused a ton of debate and controversy. The Andrews University sysadmin (Clarence Thomas, no relation) who sent it generated a flurry of complaints against his institution and some press, though reportedly he never got more than a mild punishment at the time. He did however eventually leave the University, but was also known to have done some more minor religious spams at later dates.
Normally in USENET you can post a message to more than one newsgroup using the "crossposting" mechanism. The advantage with this is that the message only goes out once, and people who read both newsgroups only see it once. This feature is highly useful if not abused, yet most major conferencing systems never implemented it.
However, it was not practical to crosspost to every single newsgroup, nor desired. Still, this event provided a button at the USENIX Unix conference saying "Jesus is coming and he doesn't know how to crosspost."
One of the most annoying things about this message was that not only did you see it again, and again, for every newsgroup you read, but it also showed up as the only message of the day in newsgroups that had low traffic levels. Most people like their low traffic newsgrous with low traffic, and this posting and others like it would soon spoil that serenity.
Around four months after the Jesus spam, in April of 1994, Canter and Siegel posted the famous Green Card Lottery - Final One? spam. It also was posted to every group.
Unlike prior net abusers, Canter and Siegel didn't turn tail and run. They were proud of it, even though net residents attacked in return, flooding their phone lines, fax machine and mail boxes. Every account they had associated with this activity was pulled. Their ISP was overloaded by the heat of the reaction, actually hurting a lot of people.
They were unrecalcitrant, and this is what really made people angry. This spam made the newspapers; it made them famous. They announced plans to form a consulting company to post such ads for other people. They wrote and got published a book about their exploits with the long title of: How to Make a Fortune on the Information Superhighway : Everyone's Guerrilla Guide to Marketing on the Internet and Other On-Line Services.
The book was dreadful, and didn't sell well, and in a remarkable show of restraint, people on the net ignored calls to stage protests at their book signings, and instead ignored them. It worked, and they vanished into deserved obscurity.
I wrote this to be a history of the term spam, though that required a bit of the history of the act itself. Keith Lynch has a Spam Timeline which details some of the later history of spam. In fact, back in 1989 I wrote what I think was the first timeline history of USENET which can now be filled out with the use of google.
Spam did a lot to ruin USENET, unfortunately. Newsgroup spam was brought into check, oddly enough, by people applying Depew's ARMM principles with better software engineering skills. Today the only reason USENET manages at all is an army of anti-spam software that send out cancels and NoCem messages, with filters that block it on the way into servers.
But E-mail spam caused even more harm, in a way. Records show people quickly started talking about email spams not long after the term became popular for USENET spam. As E-mail spam grew, it became apparent that many mailings were being generated to anybody who had posted to a USENET newsgroup. Programs were "harvesting" the addresses from messages to make mailing lists and send junk mail to them.
This in turn made people afraid to participate in the net. Post to the net (or go into a chat room) with your real E-mail address, and you would soon find your mailbox flooded with spam. Not very appealing. A number of people left the net or stopped posting altogether. Others took to no longer putting their real e-mail address in their postings, either putting in a useless address, or putting in an address that a live human could figure how to change into the real one.
One great damage of that was it broke the E-mail reply function in USENET. One thing that had made USENET quite different in the early days from other online conferences like BBSs and Compuserve was that not only could you reply to a posting with E-mail, you were often expected to. The recipient was expected to summarize replies if they were interesting. That way it kept down (but certainly didn't eliminate) private messages and me-too posts presented to everybody. You were supposed to take your flame wars to E-mail; now it's very difficult to do so.
I did this research because I've been doing a lot of thinking about the spam problem and solutions to it, particularly in E-mail. I have a page of essays on spam if you are curious.
When Google expanded their USENET archive to go back 20 years, it became possible for the first time to really research these questions again.
By the way, Hormel eventually gave up complaining about the new definition for their trademark and decided to embrace the fact it boosted sales of their merchandise.
More remembrances of USENET
At about the same time as I was researching this, I wrote an article for the O'Reilly Network called I Remember USENET with some stories of the early days of the net. You may enjoy it.
Other Etymology (Net-Surfing)
The new USENET archive at google is a gold-mine for etymology. For example, a bit of research seems to show that the term "net-surfing" originated with Brendan Kehoe, also known as the author of "Zen and the art of the Internet," an early internet book.
In this thread from 1991 he uses the term to refer to somebody browsing telnet sites (there was no web at the time). Two messages later, Ron Newman talks about how he likes the term and wants to spread it!
However, others claim independent coinage, including possibly Mark McCahil the Gopher developer (they used the metaphor a lot) and others back to the 80s who talked about Information Surfing. Paul Saffo used the term "information surfing" in a 1988 magazine column and reports it was commonly used and "definitely already in the zeitgeist" before he wrote it. Even Marshall McLuhan used the metaphor of surfing data like ur-surfer Duke Kahanomoku.
In addition, the term "channel surfing" shows first use in January 91, and seems to have originated at the same time. Several of the early users of forms of the term claim they did so due to a love of real water surfing, so this appears to be a metaphor of many parents.