Archive for October, 2007

Please don’t SPIT in my ear – VoIP Spam

October 19, 2007

Spam, spam, spam… the number one enemy of your inbox. It has almost killed email, my 10 (or is it 11?) year old ICQ-number is almost unusable thanks to all the unwanted messages I get and anyone who has a blog with comments knows that sooner or later you’re going to have to deal with unwanted comment spam.

On one of my personal blogs I am told that “Akismet has caught 228,182 spam for you since you first installed it”… (Akismet is a spam blocker for the WordPress blog tool.) That’s almost a quarter of a million messages. As a comparison, I have about 900 legit comments, so the spam comments outnumber the non-spam comments by about 245 to 1.

Spam happens whenever there is a way of contacting someone at practically no cost. Sending one million email messages doesn’t cost more than sending a hundred, if you have the addresses. If just one out of a thousand recipients respond to the message, it’s worth it. And people do respond. Spam works, otherwise we wouldn’t receive them.

So what happens when telephony becomes a zero cost communication channel?

We won’t get away. We will have spam over internet telephony, sometimes labeled SPIT:

SPIT hasn’t really taken root worldwide yet—but remember that e-mail spam was just 17 percent of e-mail traffic in 2002, where by November 2004 that figure was estimated at 93 percent.

Daniel Putz, a student at Växjö University has written an interesting undergraduate thesis on the subject. He writes:

As SIP numbers work nearly the same way as email addresses, it is easy for detection software to locate a SIP telephone simply by randomly calling numbers. Since a call can be established for free by any computer which is connected to the internet, and since the bandwidth of today’s internet connections allows sending many calls simultaneously, a sender nowadays can send many calls from anywhere at any time – for free.
Additionally, using commercial messages instead of voice is made very simple by playing a record the moment a receiver answers the
phone. As I have already pointed out, even someone new to the field of sending commercials on the phone, is able to send spam calls to
unsuspecting receivers.

He then proceeds to go through a few solutions to the problem. One of them is a pay-per-call variant:

This technique is based on the idea that a receiver gets paid for every message that is not wanted. Every sender is charged for a spam message. Only senders which are on a non-charge list are not charged.

Clever, but difficult to implement on a large scale. It’s still interesting how one of the core features of IP-telephony (free calls) also is one of its largest downsides. Maybe we don’t want free calls but cheap-enough calls and easily managed calls. Maybe a small cost per call is actually a feature to keep the network clean.

Telemarketing is still a problem (or at least most people would consider it a problem) but not bigger than today. No matter what, I will never have 245 as many telemarketing calls as regular calls to my phone. They might be annoying, but I’m not drowning in them. And of course, a good reach management solution would help deal with the problem.

Putz then puts his theories to the test by performing an experiment involving volunteers that (unknowingly) receives spam phone calls from the brave researcher. He then measured their response and asked them how they felt when receiving the calls.

Let’s just say they weren’t very happy. (Although as far as I know, no undergraduate students were hurt when performing the study.)

They also, interestingly, expected the operator (or the government) to fix the problem.

Voice spam is far more intrusive than email spam. It’s so intrusive that it might make the vision of free phone calls on the net a dream only. At least not without blocking out anyone but your closest friend. In my world, that’s not very free.

A solution to SPIT will be the difference between free calls and cheap calls for VoIP. It remains to be seen where we end up.

By the way… some good news! I just received an email telling me that “this email address was randomly selected” and that I “have subsequently
emerged [as] a winner and therefore entitled to a substantial amount of 753,000.00 Pounds”. Wow! Maybe finally we can afford to finish our beta.

Erik Starck
Managing Director, Co-founder GlocalReach