Product Philosophy:
... Or got policed - so that today there is virtually
no unsolicited commercial email sent. Valid email marketing
lists are all opt-in, and every message has an opt-out
option.
No, the problems today with open Internet email are
the con games and the criminals. The billions
of messages from these "people" are not actually commercial.
They are not marketing anything real. They are trying
to either steal from you or your company directly, or
infiltrate your computers with malicious code (e.g.
viruses and trojans) which allows the criminals to add
the systems to their illegal networks.
The "offers" for "effortless diplomas," weight loss
programs, cheap pharmaceuticals, replica watches, celebrity
gossip, discount software licenses, sales lead sources
and help with your libido are all bogus. They
just want you to click the link and become a statistic.
And if you try to figure out where these messages come
from, you realize that the senders are also bogus (typically
"bots" on already compromised PCs using fake or spoofed
addresses).
What is the REAL Problem to be Solved ?
In the simplest terms, BLOCK ALL MESSAGES FROM BOGUS
SENDERS from ever reaching your email server and your
users' inboxes. It doesn't matter what they send you,
since none of it is legitimate.
Secondarily, we also want to make sure that legitimate
messages from real senders don't inadvertently carry
malicious components such as viruses or trojans. These
two basic requirements are the definition of email
system protection. If we could accomplish that,
email systems management would be just work and not
an on-going nightmare.
Well, (surprise surprise) this is EXACTLY what Sendio
can do. Check out the rest of the Philosophy topics
and the Facts topics for how we do it.
Solving the Problem:
It comes down to a simple distinction between "good
guys" and "bad guys". Good guys are the customers, partners,
suppliers, other business associates, friends, service
providers, teachers and everyone else you already communicate
with or who you may want to communicate with in the
future. Bad guys are whoever sends (or gets some bots
to send) any bogus messages.
If you already knew all of the good guys you
ever wanted to communicate with, you could simply create
a "white list" with their addresses. That would solve
the first part of the problem. Unfortunately, there
are always new good guys coming online (like new prospective
customers). You do not want to create bad impressions
and administrative headaches by simply rejecting messages
from unknown people and then having to manually add
them to your white list.
Community
Open Internet email was essentially "invented" in
August 1982 with the publication of RFC 821, which described
the Simple Mail Transport Protocol (SMTP). At that time,
everyone using the Internet was a good guy, so no mechanisms
were included for blocking bad guys. More recently,
communications tools like Instant Messaging (AIM, GTalk,
Skype et al) and social networks like Facebook have
been released, and their designers learned from watching
how bad guys abused email.
All of these more modern communications tools assume
that there are good guys and bad guys (i.e. "friends"
and "everybody else"), and they provide a means for
creating each user's "community" of friends (or "buddies"
or "contacts" or "followers" etc.). You interact with
your community, and are protected from everybody else.
New people can ask to join your community, and you control
the process.
Community for Email
This concept of community is EXACTLY what email needs
in order to solve the abuse problems. But, it must be
simple and intuitive, and not add any administrative
overhead. As an email administrator (presumably), you
have two basic choices: request that your email system
vendor add this functionality to their product, or externally
augment your system. (We, of course, would recommend
adding our products as the ideal course of action.)
At a high level, Sendio's definition of communities
for email can be summarized as:
- Take the effectiveness of white listing
- Add a straightforward automated mechanism for
new "good" people to join that "bad guys" won't
use
- Provide an easy way for dumping a "semi" good
guy that you just don't have the need or interest
to communicate with
- Add extra features for dealing with mailing
lists, newsletters and other automated messages
that come from "software" and not "people"
- Finally, perform a virus check on messages just
to make sure nothing is trying to sneak in via a
message from a "good guy"
Building an Initial Community
So, how do you go from lovely concept to hard reality
? First, you recognize that there can be both company-wide
communities and personal communities. A company-wide
community is the list of people that everyone may want
to communicate with, whereas a personal community would
include people that are specific to an individual.
The company-wide community would import all of the
system contacts in your email server and, typically,
all of the contacts in the your CRM system (SAP, Siebel
et al). Personal communities would import all of an
individual user's contacts (from Outlook for example).
In this way, everybody you already do business with
or interact with is automatically in your community.
Maintaining a Community
Once your community is defined and deployed, the
communications process is straightforward:
- Every email sent to your server first has a
series of SMTP validity and header checks performed
to eliminate obviously bogus, forged or spoofed
messages
- Next, the message is scanned for viruses and
other malware, and it is rejected if anything is
found
- Then, the message is analyzed to determine if
it violates any corporate policies (e.g. message
size, attachment types, standards compliance [DKIM,
SPF, etc.])
- Finally, the message sender is compared to the
community lists, with two potential outcomes:
- If the sender is already a member of the
community, the message is forwarded on to the
email server for routing to the appropriate
inbox(es)
- If the sender is NOT a member of the community,
the message is held in a "pending queue" while
the sender is invited to join the community
The "Invitation" Process
Here is where the bad guys get shut down while new
good guys get connected. When a message from a non-community
member comes in, it gets held in a "pending queue" while
an invitation-to-join-my-community email is sent back
to the non-member. All that person needs to do to join
your community is reply to the invitation (click Reply
and Send). This proves that a real person sent the original
message.
Once they reply, they are now a part of your community
and their original message gets delivered. Then, from
now on, all messages from the new community member get
delivered to you immediately, unless you decide to block
then at some point. You "own" your community and completely
control who has access to your inbox.
The "trick" here is that the bad guys won't (and
in fact can't) reply to the invitation-to-join-my-community
message. Bad guys go to great lengths to be "anonymous"
so that they cannot get tracked down. A message sent
to them goes to either a fake address or a compromised
bot system, neither of which will respond. So, the invitation
gets ignored and the original message gets dropped from
the pending queue. You never see it.
This process is EXACTLY the same as when you joined
Skype or Facebook. You had to confirm you were "real"
and then you were in (until you break the rules at which
point you get dropped). Everybody is familiar with this.
Newsletters and Subscriptions
But wait, you ask, what about legitimate messages
that come from automated systems, not "people", like
newsletters and subscription notices and other online
services ? No problems. When you first implement email
system protection using communities, you "teach" the
system about each user's personal preferences. One user
may want notices from Adobe, while another user NEVER
wants those notices, and each can be accommodated.
No "Lost" Email
So, adding community management protects your email
system from attacks and abuse by:
- blocking all messages from bogus senders
- blocking all messages that include malicious
components (e.g. viruses and trojans)
In addition, it eliminates the "lost email" problem.
So called "email security" products create this problem
when they attempt to block bad messages using techniques
like "probabilistic risk assessment" and "reputation
scoring," which are fancy technical marketing terms
for GUESSING. And we have all seen how they frequently
guess wrong.
With communities, it is simply about message senders.
If they are a part of your community, you get ALL of
their messages. If they are outside, you get NONE. You
have total control.
Easy. Elegant.
Content Filtering :
In the IT world, if someone says the word "anti-sp*m",
everyone adds the word "filter" in their minds. It seems
that "of course they just go together". But why ? We
just cannot figure it out.
It must be something like "of course the world is
flat" or "of course cigarettes are safe" or "of course
General Motors is a solid investment". Those things
that "everybody knows" right up until they are shocked
by the reality.
Well, content filtering of network traffic is cool
technology. And, when it can help make simple, clear
yes/no, good/bad, black/white decisions, it can be very
useful. For instance, in data leak prevention (DLP)
products, filtering can work well. If the filter "sees"
a credit card number or a social security number where
there should not be one, then actions can be taken.
A totally deterministic process.
Even anti-virus scanning is a form of content filtering.
If a virus signature is matched, block the message.
Again, a totally deterministic process.
The Failure: Using Content Filtering to Determine
Meaning
But using content filtering for the "semantic content"
(i.e. the "meaning") of an email message seems crazy.
Sure, some content would be "obviously" bad, but most
of it is open to interpretation. If you have ten people
read a message, and four find it offensive, three don't
care and three really like it, what should software
do ? This is NOT a deterministic process. This is why,
with email, filters have failed.
Again, let's be blunt. If email filters really worked,
we would not still be having these problems (which are
actually getting worse). But since filters are really
just "guessing" at the meaning of a message, the bad
guys keep adapting their approaches and the garbage
just keeps flowing in.
What it comes down to: Who, or what, do you Trust?
It comes down to who, or what, you trust, and whether
information is flowing in or out. If you truly trust
your employees, then there is nothing to worry about
with regard to the wrong information leaving the company.
In the real world, even good people can make mistakes,
so many companies are deploying DLP products because
it really is all about the data.
On the incoming side, it is NOT about the data. It's
about whether you trust the sender of the data. If you
do not trust them, why do you want ANYTHING they send
? And the answer is, you don't.
Sendio
Philosophy