-=> Mocking Michael to Kurt <=-
  (Mock, mOck, moCk, mocK)
 KW> last time i read a virus bulletin test (i believe you posted it
 KW> actually) they got close to using about 400 different viruses... hardly
 KW> sufficient when there are over 11,000 known...
 MM> Why it should not be not OK to test the scanners just with the viruses
 MM> that are most common seen out there in the wild. If they test this way
 MM> even new upgrowing AV products (like Dr.Web) will have a chance to
 MM> show up there qualities.
well, it's like this... there are over 11,000 known viruses out there...
if you only use say 400 viruses for the test then it's entirely possible
that the product doesn't detect the other 10,600 viruses and you won't
be very well protected in that circumstance...
the viruses in the wild are only the viruses that have been reported by
trusted individuals to be in the wild at this point in time... it says
nothing about which viruses will find their way into the wild in the
future or which viruses are in the wild that the trusted reporters (not
news reporters) aren't around to report about...
scanner tests are designed (and assumed) to represent the
number/percentage of viruses a scanner can detect and thus how much
protection it can offer... if you don't use most of the viruses known
then you can't come to a reasonable figure for the detection rate..
viruses don't behave like people, statistical analysis on small samples
doesn't usually work because a small sample usually isn't representative
of the whole set of known viruses and it's next to impossible to get a
representative sample without first having all known viruses and testing
the scanners against them...
as for displaying scanner qualities... a scanner is a security product,
it's security depends on detection of viruses... therefore it's most
important quality is it's ability to detect a large number (preferably
more than any other scanner) of viruses... as i've already said, you
can't test that without actually having a large number of viruses...
... ever notice how ours is the only planet named after dirt?...
                   
--- Maximus 2.02
---------------