"98 Guy" wrote in message ...
FromTheRafters wrote:

without providing any shread of evidence that those threats
or exploits are operable on win-98 systems.

Considering exploits, it is not reasonable to assume that your
OS is more secure just because an exploit is not operable on
it.

I think you mean payload - not exploit.

Yes, I meant "threat" not "exploit".

If a given piece of exploit code is not operable on a given platform,
then how can that platform be vulnerable to the exploit or any
hypothetical payload / shell-code that might follow? How do you
define
vulnerable in that context?

The exploit in this case is against the application, what follows might
be OS platform specific. For instance If a demo exploit has a benign
payload (like executing notepad) it may work for all versions, but if a
real world exploit calls cmd.exe (which W98 doesn't have) then the
threat is still valid even though it isn't operable on your OS.

If the vulnerable software falls over, but the OS doesn't
recognise the shellcode, the system is *still* vulnerable
to the exploit.

But there's no consequence if either or both the exploit or the
shellcode does not function properly on a given system.

It could mean the difference between a worm instance being hosted and a
DoS against the vulnerable application.

If the exploit or the shellcode causes the application (or the OS) to
crash, well that's just a nuisance that's not likely going to be
repeated by the user.

I don't really consider DoS's to be a significant or credible threat
to
anonymous end-users (what's the point?).

Well then, a DoS exploit is not an exploit to you?

If it is an NT specific malware *payload* you might not be
vulnerable to the payload, but you still are vulnerable to
the exploit.

Until we see a functional example of an operable PDF exploit AND
payload
for the Win-98/Acrobat-6 combination then we can't be sure *if* there
is
a viable exploit in the first place.

I suppose you have your own unique definition of payload then?