Unless SETI is as bad as Outlook et al, the concerns should be
dismissed.
The issue is not that writing a virus requires knowledge of a computer
system, the issue is that being infectable requires running code, not
just processing data. (buffer overruns occur because of missing
terminators -- the data is not being handled strictly as data, but is
in effect "interpreted" and hence trusted to be well-formed) [0]
Given that there's no way for the SETI processing to be interpreting a
given data format, much less executing arbitrary attachments, they'd
have to have written their clients in a very byzantine fashion to have
any hope of being a virus vector.
If it were April, I'd wager this physicist was playing a joke on the
Grauniad. [1]
-Dave
A possible explanation for the Fermi Paradox: the galaxy is chock full
of sentient beings, but they don't wish to have any contact with us for
fear of having 86% of their matryoshka clusters devoted to forwarding
interstellar 419 spam and offers for organic topological deformations.
:: :: ::
[0] this should be CS 101. On a meta level, think of it as Liberal
Arts 101 as well: any given policy recommendation may be good or bad,
but it is not dangerous as mere symbols sitting on a shelf --
information only becomes dangerous when followed blindly, without
thought or consideration of consequences.
[1] it is somewhat suspicious that
http://www.iaanet.org/publicat/aacont.html has no mention of any likely
titles for v58 n1 and instead shows more prosaic entries such as
"Electric control system for microgravity fluid experiment on SZ-4
spaceship". Is there a meta virus here?
