In January 2002 — at the tail-end of the last big economic shakeup — WebGroup Media LLC (WGM) had been a full-time venture for about nine months. Ben Halsted and I were the early founders, and we had enough contract gigs behind us to know we really wanted to work on something with a passionate, long-term mission. We were tired of working on projects where the specs called for technology decisions based on marketing bullet-points; leaving us to always learn the better tools in our spare time.

We were still paying the bills back then with a couple contract jobs and a few hundred customers on budget hosting plans. But we had our senses primed for finding the problem we’d enjoy throwing ourselves at for a living.

We had to spread ourselves pretty thin to bootstrap the company and keep the lights on, and accepting all that discounted work kept us in a mode of perpetual maintenance just answering questions and fixing problems. I remember feeling like we’d never find the time to build something new because the workload would reset itself each night.

We’d tried to organize the e-mail workload with several different apps, but primarily we had been using RT. The interface worked just fine, but even with our light volume it started to become unusable pretty quickly. It’s still around, and I’m sure it’s made a lot of progress since then, but at the time we were rapidly exhausting our supply of small talk for passing the several minutes it took to run searches. Beyond that, it’s written in Perl, and most web hosts were already strongly favoring the LAMP stack (Linux, Apache, MySQL, PHP). Pundits say the “P” stands for PHP, Python, or Perl. In the hosting market, it stands for PHP.

January 9th 2002 was the day Ben and I had an IM conversation where we brainstormed about how to break the monotonous cycle and start working on something more interesting. In good spirits, and youthful defiance, we kept joking about how “support is hell”. It was around the same time that we were establishing the mantra we’ve had to this day of “Most Important Thing First!” (or “MITF!”). It didn’t take long to realize other people were probably as frustrated with their shared e-mail as we were. Our biggest problem was the thing we had the most to say about. We had our project idea.

We quickly started thinking of names for the project. Some serious, most not. I’ve long since lost the original logs, but I believe it was Ben who (perhaps jokingly, but now immortally) asked what the name of the guard dog to hell in mythology was. I said “Cerberus”. He said something along the lines of “We should just call it that”. We laughed. It had been a long chat, and we decided to go with the code name of “Cerberus” until something better came along.

Two months later we sold the first copy. The logo still said “Cerberus Helpdesk”. There was no going back!

It’s been a really interesting seven years.

In a lot of ways, we underestimated what we were getting ourselves into (”I know! Let’s make a startup that stores years worth of e-mail and file attachments for thousands of companies worldwide; guaranteeing our office never closes and we never sleep!”). What we definitely didn’t underestimate was how many people shared our frustration with shared e-mail. We’ve had the good fortune of coming to the attention of well over twenty thousand teams in the past few years. All that feedback and positive energy has snowballed the project into something so much better than we had originally imagined. As cliché as that sounds, it’s still true.

I think the big reason this project has been energized for seven years is that it’s still as exciting and relevant today as it was in the beginning. We owe a lot of that to a project culture that’s still willing to take risks and make big changes while pursuing solutions that are as simple as they are powerful. That hasn’t been without difficulty. The transition from Cerb2 to Cerb3 was seen by many in the community as a disaster of Frankenstein proportions (I like to call it the “co-mutiny”); and it’s taken over 2 years for Cerb4, free of the baggage and clutter, to win people over based on its own merits as something entirely new.

Cerb4 is new, but it has seven years and thousands of brains behind it. Seeing that work pay off as people put it to work is hugely rewarding. Even this morning, somebody on the team sent me the link to a forum post where a community member said, “Cerb4 is naked compared to Cerb3!”, quickly followed up with another post by the same person saying, “I have to say, even naked, it’s very powerful and very easy to use. Thumbs up.”

That’s what gets me out of bed in the morning… 2,556 days later. ;)

With the mushy nostalgia out of the way, here’s an entertaining look at the project’s evolution.

The interface today still proudly shows its lineage, but a refined usability has replaced the gaudy attention-seeking of yesteryear:

The project website has gone through a bit of an identity crisis: from a blip (2002), to taking itself too seriously (2005-2007), to defying convention (2008), and settling into a matured self-confidence for 2009.

The project web site history was pulled from the Internet Archive’s “WayBack Machine”:
http://web.archive.org/web/*/http://www.cerberusweb.com

Happy Birthday, Cerberus! And thank *YOU* for being a part of the project!

-Jeff Standen, Chief of R&D, WebGroup Media LLC.
on behalf of the current Cerberus Helpdesk team: myself, Darren Sugita, Dan Hildebrandt, Joe Geck, and Mike Fogg; and retired team members: Ben Halsted, Jeremy Johnstone, Brenan Cavish, Jerry Kanoholani, and Trent Ramseyer; and many more community heroes.