|
Today I concluded that my mail-box with 120
fresh messages constituted of about 110 messages where the sender address
is either falsified, or is coming through a hijacked computer.
In my opinion S/MIME is the primary culprit
for this unbearable situation.
That Windows have showed some weaknesses with
respect to virus attacks is undoubtedly true, but viruses would also have
had a much less impact if we have had a useful e-mail security
architecture. The same goes for phishing.
A do believe that the designers of S/MIME did what they could back in the
90'ties. However, now when we know better [*], shouldn't these guys who
indirectly contribute to an annual waste of hundreds of millions of
good working hours from the Internet community rather try to create a
system that to some extent compensates for the mistakes done in the
past?
DKIM is a step in the right direction but it does not address
confidentiality. That DKIM was designed to support people who want to
run their own mail-servers but cannot afford a domain-certificate is also a bit
off since these entities represent at most 0.1% of today's Internet
users.
Anders Rundgren
*]
- Client certificates are [still]
uncommon
- Encryption at the desktop by consumers
does not work
- Security administrators want central policy
handling
- Trusted third-parties is the norm (from your
employer to Google)
- You cannot send an encrypted e-mail to the IRS
and you probably never will
- e-mail encryption is incompatible with
many organizations' internal policies
- Security should be transparent, default, and
non-intrusive
|