February 16th, 2004

Stagnation or Growth

January 30, the day after I got back from Japan, Marshall asked me to commit to staying at MIT for another six months. I did eventually. But I was surprised at how much hesitation I faced trying to answer his request.

I'm not unhappy with my job. Marshall is one of the three good managers I've worked for. I enjoy working with all the members of the team and wouldn't mind working with them on future projects.

Only later that evening did I begin to understand why I felt like I was giving something up by agreeing to stay at MIT.

Exploring these feelings, I realized that I was more affected by the potential for being laid off than I had at first supposed. I managed to face the layoff possibility with reasonable equanimity.

As I was considering the layoffs, I realized that I'm stagnating somewhat at MIT. The design problems I'm solving here are not that complex; I was certainly doing more challenging work at Permabit and at FundsXpress. The core problem Mekinok was trying to solve is clearly harder than Kerberos--possibly so hard that it cannot be solved at all.

But more than that, I'm putting off some important explorations here at MIT. Long term I want to create a stable team of people that I can work with/for. This goal comes out of some discussions on fxiles, a mailing list for people who worked at FundsXpress. I should dig up those discussions and put them somewhere public to link to. But the short summary is that I roughly know what I want out of a work environment. I've had it for short times in the past, but I want to make it a stable feature of a team.

I can't work on that while at MIT. First of all, it's not my job to do that sort of organizational work. Secondly, I'm not sure that what I want is compatible with an organization dedicated to a single project like Kerberos, particularly when that project is not core to the surrounding business.

While considering what would happen if I got a layoff notice, I realized that at some point I'll need to focus on these goals. I'll need to find out if what I want is compatible with a successful business and if I have the skills (or can find the people) to make it work. What I'm doing with Netzah is an experiment in that direction, but I can't really say that I'm giving it that much of my attention. And I realized that if I were laid off, I would have something to do and to work towards immediately. So I began to look forward to my options if I were laid off. That certainly made facing the layoffs easier.

I want to make it clear that I wasn't hoping to be laid off. I was also looking forward to what I was doing at MIT. I had simply managed to turn a situation where I was facing the possibility of losing my job into one where I was facing two potentially interesting life paths. Better yet, the decision was completely out of my hands.

It seems fairly clear that staying at MIT is the right thing. It's doing a good job of building up my reputation as a protocol designer and Kerberos expert. I'm not really ready right now to go work on those other goals. And we do have some important work to do in order to make sure that MIT Kerberos has a future . I'm certainly happy with my decision, but I also miss the options that I don't yet get to pursue.