Sections:

• Introduction
• Philosophy
• Gavin Stewart
• Timeline

Introduction:

OpenBottle began with Bluebottle Systems Pty. Ltd. at the start of 2002. I (Gavin Stewart) was in a discussion with the director of Bluebottle Systems over a potential mechanism to resolve the growing spam problem. Spam was a rather hot topic at the time. The mechanism discussed was that of getting an email sender to verify themselves to their intended recipient the first time an email is sent.

In March 2002 I wrote a whitepaper and a flowchart on the mechanism, at that point not having known about any similar efforts in the Open Source community. (Soon after I became aware that TMDA had a patch for sender authenticated whitelists.) As development progressed many other anti-spam businesses began to pop up, some also using authenticated whitelists.

From the very beginning we expected to make the source code available. We didn't know for sure what license, or when, or at what stage in development however. The decision to make the source open was a fairly quick one. It was plainly obvious that we would not find investors able to give us the backing to get us where we wanted to go. (We're still in the dot com gloom here.) Opening the source up would get people talking about it, testing it, criticising it and improving it. ("It" being either the system/software itself, or the logic behind it.)

Philosophy:

"World Peace" was looking tough, so I thought I should "start small" and would work towards eradicating spam first. (World Peace is looking easier now.)

As a consultant, I have quite a few customers ask the inevitable "can't you do something about all this spam?". (These are typically small to medium businesses BTW). After implementing one of the more well known mechanisms, such as RBL's (Realtime Blackhole List) or filtering, the customer eventually wants the anti-spam "solution" turned off. My experience is that business would rather put up with the spam, than the inconvenience of blacklisting a potential customer, or blocking a security paper on Internet misuse that may contain a few juicy terms.

We all know there is an inherent weakness in the current email systems we use. It is also what makes it as useful as it is. The ability for anybody to arbitrarily contact anybody else anywhere in the world at any time is an awfully powerful tool. (And awfully hard to put limits upon and yet keep that same essence). The late Jon Postel wrote an RFC identifying this weakness in 1975.

It is the aim of Bluebottle Systems (and hence OpenBottle) to resolve the weakness in the existing email system that permits spam, whilst preserving the spontaneous and even gratuitous nature of email as we know it today. (Note that the definition of spam is left up to the readers own interpretations as this appears to be the source of much contention, and boons little to re-hash that here.)

Gavin Stewart:

Just so you know who the "I" is in the above text. My name is Gavin Stewart. Much of my employment history has been with ISP's and more recently as a consultant supplying various Internet related skills (un*x systems, network security, whatever pays the bills). I have always tried to limit exposure of my own details on the Internet from my first introduction to it (1991). This is the most successful mechanism to avoid spam .... and flame wars. Obligatory homepage link.

Timeline: (reverse chronological order)

Feb 2003: Source code released.
Oct 2002: Bluebottle Systems launches end-user Bluebottle service.
Apr 2002: Start of development.