Slashdot Mirror
Can the Malware Industry be Trusted?
Joe Barr writes "Is the entire anti-virus / malware industry as rotten as it appears? I started digging into it as a result of the recent lame, unsubstantiated assertions of viral threats to Linux by Kaspersky Lab, but the practice doesn't seem to start or end with them. Who knows, maybe it's pandemic in that entire segment of the IT industry."
Agreed, the industry is full of FUD, along with other substances.
Noticed a copy of AntiVirus for Mac OSX @ CompUSA last week. $59! Three questions:
1) Who buys this stuff?
2) Why so much?
3) Why?
To my knowledge there is only one virus in the wild for OSX and it never really made an impact. I understand that AV for Mac scans for the billions of Windows viruses, but considering that the Mac is extraordinarily unlikely to become infected, it's similarly unlikely a Mac will pass on a virus. I know it's part of being a good net citizen, but ultimately scanning email is your own responsibility. I don't scan for Linux or mainframe viruses, or iPaq scripts. Why should I scan for Windows viruses?
Or am I missing something?
Every year, US-Cert produces huge fireworks in the security trade press with their annual summary of misinformation about security flaws. The idiots in the press repeat the lie verbatim and the lie becomes real. What is the lie? That Unix/Linux is less secure than Windows. Granted, only the stupidest dolts in the universe -- and the trade press -- are going to buy that crap, but they put it out there anyway.
I got to that point in the article and remembered the red ink on a paper I wrote in grad school, wherein the professor said, "too pejorative to be taken as an objective analysis of the topic."
In all things academic or reporting, if you do not really have it, then at least fake objectivity....
"All great things are simple & expressed in a single word: freedom, justice, honor, duty, mercy, hope." --Churchill
Spam vendors and open source vendors make lots of whacko claims.
[...] extremely low false positive rate, with less than one in one million messages being a false positive.
A few years ago, Bayesian classification seemed a promising way to filter spam.
[...] best recorded levels of accuracy have included 99.991% by one avid user (2 errors in 22,786) and 99.987% by the author (1 error in 7000), which is ten times more accurate than a human being!
That translates to better than 99.984% accuracy, which is over ten times more accurate than human accuracy
In the game of cat and mouse between spammers and anti-spam vendors, spammers and hackers quickly developed new techniques to "fool" the Bayesian filtering software.
File these under UFO sightings.