J. Gomez writes:

 > Yes, the user did it to himself, but what does he know?

Obviously too little to be trusted with an email account.  Fire the
corporate training department!

 > Please note this attack works successfully even if the user has no
 > administrative rights on his computer, and could potentially be
 > made to work equally well with Linux and MacOSX users too (if they
 > were a big enough demographic target to make it profitable vs cost
 > of development).

AFAICS the cost of porting is very low compared to original
development.  I would guess that the real issue is that they don't
have a good way to identify the executable format for the recipient
system, and so send a payload that will work on well over 90% of
recipients.

I also doubt it would work as well on Mac OS X, where the user would
be prompted for his password to confirm permission to execute an
application received from an untrusted source.  Surely some would type
the password, but I suspect enough would be deterred to lower the
click rate to unprofitable levels.

 > The lessons which I think we can learn from this are:

ISTM we already knew the lessons you list, and they inform every
discussion I've seen on this list.

My personal opinion is that, on the contrary, people are already way
too quick to discard proposals simply because they involve changes to
MUAs.  Of course, the reality that this is an IETF WG, and what we can
do that has effect with high probability is change wire protocols.
MUA presentation is outside of our bailiwick, and nobody really has a
good way to get ideas for MUAs broadly implemented the way we can
influence MTA implementations.

 > So I thought this could be of interest to keep in mind, when some
 > solution may be suggested to the DMARC indirect flows problems
 > which advocates some kind of MUA behavior regarding message
 > presentation.

"No Silver Bullet".  There are no "solutions" to these problems, only
improvements.  The fact that *some* users will dig a phish out of the
spam bucket and cut/paste a disabled URL into their browsers so that
they can be victimized despite the best efforts of their mail agents
doesn't mean that others who *would* click if it were presented as
valid mail would *not* go to such lengths for mail in their spam
folders, or perhaps would be deterred at the "cut/paste" stage.

_______________________________________________
dmarc mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dmarc

Reply via email to