...because you didn't get a soundbite from Graham Cluley, Senior Technology Consultant from Sophos, the clearly omnipotent overlord of all things malware. Or is that just the BBC?

Symantec also turned 25 this year. Coincidence?

http://www.symantec.com/enterprise/security_response/weblog/security_response_blog/evolution_of_security/

The (harmless) "IBM Gremlin" was old news in my undergraduate days. And I left Cambridge in 1983.

...but John Walker's AMINAL variant was infecting UNIVAC machines in 1975.

http://www.fourmilab.ch/documents/univac/animal.html

I think Nick Kew must be misremembering his youth through the rosy-tinted glasses of middle age. According to

http://www.ciac.org/ciac/documents/CIAC-2301_Virus_Information_Update_5-98.pdf

"gremlin" was released in 1997. It's pretty hard to see how it could be "old news" in 1983 given that the very first models of IBM PC were only released in 1981.

Got any references to back up your memories?

It's true, Elk Cloner was the first personal computer virus, ever. See

http://www.skrenta.com/cloner/clone-src.txt

Symantec has nothing to do with this. At that time, it didn't have anything to do with viruses or anti-virus programs. That happened much later, when it aquired Peter Norton Computing - the company that used to make Norton Anti-Virus (among other things). But even Peter Norton himself as late as 1988 (when several viruses for the IBM PC, like Brain, Vienna, Cascade and Jerusalem, were already widespread) was saying that computer viruses were just an urban legend, like the aligators in New York City's sewers.

Naturally it depends on how you define a virus. I "understand" that there was a small program that ran on the DEC10 at York Uni before 1979 that simulated the action of a virus on a cell by making multiple copies of itself and running them, though it didn't infect other machines. However, Metronet Bunnies allegedy ran round the London Uni network (Metronet) from mainframe to mainframe in about 1979-80. So it seems that viruses didn't start with PCs: PCs just made it easier.

To Anonymous Coward. I don't like turning this into a discussion, but the IBM Gremlin in question was of course a *mainframe* thing.

I've no idea what the 1997 "gremlin" was, but it wouldn't surprise me at all to hear the word had been re-used, perhaps many times.

To Vesselin Bontchev: I was kidding about Symantec, attempting to imply that Symantec's entire business is based on creating computer viruses, which obviously isn't true.

I remember viruses at MIT in 1972 or 73...

Historically I wrote my first virus - a self-replicating worm - on a (data general eclipse) time shared mainframe in an Essex College, in 1978. The week after that I tried phishing and harvested many *many* user credentials by emulating the Login Screen. Having proved the concept I concentrated on work and beer and stuff and left the virii to others, making my old college BOFH a much happier person!

I suspect the first viruses/malware just came about when powerful computer hardware was exposed to thoughtful spotty youths on a large scale, my team had contemporaneously built and programmed a Nascom-1 in the evenings - viruses not possible due to h/w & s/w considerations and we were too busy trying to make it work , but during the day we delighted in crashing the Eclipse! Somebody else's Problem!

Now , of course, I'm professionally working in Cybersecurity!

Viruses... they were a plague on Amiga and Atari ST systems. Both tend to be forgotten by the "professionals" who hid and pretended that both of these systems were not considerably, cheaper, more flexible, more powerful and more capable than any PC available at the time. Yes, games could be played on them and it was quite a large part of the marketing, however the serious applications of both were very extensive (think Amiga for video titling and raytracing and Atari ST for MIDI). It took a very long time for anything PC based to displace just these two markets, and it's still not entirely uncommon to find Atari STs in music studios.

It wasn't long after these machines came out that viruses started to spread on them, mostly floppy diskette boot sector viruses but with the introduction of hard drive models the viruses became much more sophisticated. By 1987 and the release of the Amiga 500 and Atari ST, viruses were very common (the release of these more popular, accessible and, primarily, cheaper systems gave a much broader base of virus attack). By the time the IBM PC limped into view with 16 colour graphics, and other wonderful "innovations" and slowly became dominant, many estabilished virus writers moved to the newly emerging technology as it was the place to be.

At this time viruses typically just spread themselves, popped up screens "demonstrating" just how cool the virus writers were and a few had nasty payloads capable of trashing systems. Many virus writers were proud of just how small they could make their viruses. Since the advent of the Internet they've come a long way and viruses are now more for profit (through spam scams, phishing and DDOS blackmail) than for vanity or the wilful destruction of systems.

one of the earliest known prank viruses, the first known version of this application was first deployed on Multics systems in the early 70's.

networking was rather primitive, so the virus spread between systems mainly through "sneakernet", leading to some debate as to whether whimsical humans with admin privileges can be considered a valid infection vector (i vote yes, as any effective infection vector is, by empirical demonstration, a successful propagation method).

the program would be triggered by a timer, much like a cron job, but without the easy log tracking. when it woke up, the cookie monster would stop whatever the user was doing, and ask for a cookie. once the user gave up fighting it and typed "cookie", the virus would go back to sleep until the next timer event, and the user was returned to the original process.

ah, the early days...

The IBM PC "limped into view" in 1981, 4 years before the Amiga A1000 and Atari ST first shipped in 1985.

And b shubin: "cookie" was a prank program but I don't think it can be considered a virus. It didn't have any means of spreading itself to other systems, except maybe manual copying by disgruntled operators...

Yes, indeed. Although I wasn't a victim, it was explained to me by a regular user (Ph.D student at Melbourne Uni in Victoria, Australia) of the MelbUni mainframe systems (character based terminals, remember).

Cookie would wake up, select a terminal to bother, and would type

"Please may I have a cookie?"

If the user typed "Cookie" the program (more a worm than a virus) would fill the entire 80x25 character screen with "Yum! Yum!" and go away for a while.

If not, the requests became more stern and insistent over time, until the screen was locked with the banner headline text

"I DEMAND A COOKIE!!!"

and the user could not continue without typing in "cookie" when it was released with the "Yum! Yum!" response of satisfaction.

The program was adaptive as it eventually favoured the users who responded more quickly. They got more "Yum! Yum!"s on their screen until it became too much of an interference with their work. [I wonder if the programmer had a liking for The Mikado?]

The www.viruslist.com site says that in 1983 "he defined the phrase 'computer virus'" however the concept of machines being infected or diseased was part of the 1974 film Westworld where the pleasure driods finally get pissed off by always being the victim of their human customers.

There it is rather prophetically pointed out that the scientists and engineers that support the project that they do not understand how the machines actually work, as they were designed by other machines, and that the pattern of failure is not concentrated in the obvious actuators, but is more like a disease in humans.

...and some interesting comments.

Ultimately though, your're all old farts.

Which is why C and Verilog are safer than VHDL and VB.

A C programmer can normally tell you what sort of assembler will come out,

A verilog designer can tell you what gate's he's expecting,

A VHDL designer actually believes in bits as signed numbers.

And a VB "programmer" believes that his language is compiled, and it's all done by magic little elves that run around inside processor, which is made faster by having "more ram"

What.... you mean it's not little elves making my VB work? And more ram does make it faster - they like a good roast mutton, the little devils.

Microsoft Windows.

The first virus affecting computational mechanics was when Ada Byron caught a cold. Babbage could be a right old tramp with his hygiene sometimes.......

I was one of those spotty yoots when I had my Atari 800 back in 1979. In 1980 I took a small piece of malware someone else wrote and turned it into a virus which would remain memory-resident and self-replicate. After formatting any diskette the victim inserted into the drive, it wrote a hidden file to infect any machine the disk was then used on. This was a payback for the people who were getting pirated software free and then turning around to sell it. I'm pretty sure I still have the source code for it somewhere.

This got me thinking and I went up to the attic and found this punched card virus I had kicking about. I would email it to you but for some reason my CD Drive wouldn't read the 1950's Remington-Rand UNIVAC format.

Mac : I'm a Mac,

PC : And I'm a PC,

PC : I've got a nasty virus Mac.

Mac : Well PC I'm a mac and I dont get viruses.

PC : Dont Talk bollocks Mac you guys go the first virus ever you got pwned by a 9th grader and the world has been suffering since.

Mac : Well I dont get them anymore.

PC : Great so you cause a problem and then just pass the buck.

I see that Kaspersky , in their virus history, claim the term was coined in 1983 - I don't see how, when a lecturer told me & a load of other students at U.C.London about computer viruses (& calling then viruses) in 1982 - mainly to tell us not to release any(please)