EU Spam Symposium 2006

Last month the first EU Spam Symposium took place at the University of Maastricht in the Netherlands. Since this was the inaugural symposium, delegates had not set their expectations too high, but in the event the organizers did an admirable job. This forum invited a variety of technological, jurisdictional and commercial speakers from all over the world. It was also the first public event to have an ex-spammer on the programme.

Of course, like all conferences, it had good parts and not so good parts. But, overall, it was good to see that spam is no longer ignored worldwide. The only thing I was disappointed with was that the symposium did not address phishing at all – a subject that, in my opinion, should have been covered extensively.

The symposium was opened by the head of the anti-spam department of the Dutch OPTA, Danyel Molenaar, who presented 'Enforcing anti-spam law in .NL'. OPTA is a government body that regulates compliance with legislation and regulations in the areas of post and electronic communications. Danyel's message was pretty clear and I have to agree with him: 'It is effective to enforce! But cooperation is vital'.

Next on the agenda was John Graham-Cumming, who is well known for his papers and research in email filtering and (especially) spammers' tricks [as well as in the pages of VB - Ed]. His paper was entitled 'Three years of spam mutation' and was very entertaining. The presentation was an up-to-date overview of the latest spamming trends. All the techniques presented are also described in John's Spammers' Compendium, which is available on his website (http://www.jgc.org/).

Cristina Bueti from the ITU (International Telecommunication Union) spoke about how a United Nations (UN) specialized agency helps the world to communicate. Her presentation was called 'Countering spam in a digital world'. It was very encouraging to see that organizations like the UN are taking the spam problem seriously. However, I have some doubts that anything will really happen without someone employed to enforce it. Currently, the UN is providing advice to its members on how to fight spam and other digital threats. The other good thing about this presentation was that it was the only one where I saw mention of the word 'phishing' – which was listed among the various aspects of spam.

Next, Ann Elisabeth presented her paper on tracking spammers in Norway. Unfortunately, the presentation ran over the allotted time and had to be interrupted due to time constraints. She attempted to explain which email header fields are important in tracking a spammer and how this information is used. Tools to analyse IP addresses, MX information, DNS records and many others are part of her 'bag'. Regrettably, because the presentation was cut short, the audience was left with incomplete information.

The most controversial presentation was that of Spammer-X, a.k.a. Eddy. In his presentation 'Inside the world of spam: from the eyes of a spammer', Eddy explained to delegates that, for five years, he was the man in the shadow who sent 'many, many millions of spam emails'. He even wrote a book called Inside the Spam Cartel.

I have doubts that the guy was as good as he says he was. He wrote a book about the spamming industry, so why didn't he receive questions from his old 'friends' from the cartel, from police or other law enforcement organizations? Of course, he didn't reveal any real names in his book.

Eddy managed to break the circle and to get away from 'the spam cartel', and he also kept all the money that he made. The central point of his presentation was to explain why the spamming business works: it works because people buy the products advertised in spam. Unfortunately, they don't even get the desired effects from the blue pills they buy. Why? Because, according to Eddy, all the products advertised in spams are fakes.

Jose Maria Gomez Hidalgo, of the European University of Madrid, presented 'History, techniques and evaluation of Bayesian spam filters'. I enjoyed this presentation from an academic point of view, but it was rather long and very heavy on statistics. The presentation focused on Spam Assassin rule sets, on Bayesian poisoning and on known methods to improve statistical filtering.

The final speaker was Matthew Prince, from anti-spam software and services company Unspam, who explained 'Why anti-spam laws haven't worked, and what to do to fix them'. Those who attended the VB conference in Dublin last year might have had a déjà vu experience. Matthew enhanced that presentation with up-to-date information and described in a very entertaining and enthusiastic way what Project Honeypot does and how it helps with tracking spammers. I have to say that he also convinced me to get the honeypot they are working on and give it a try. However, I do have some doubts about who is using this honeypot – if I were a spammer, I would install the honeypot and develop some kind of obfuscation methods to bypass it.

Slides and webstreams for all the presentations are available at http://www.spamsymposium.eu/archivewebcast.htm.