linking unknown work to known

The job I need to do right now is to implement Radia kick maps in AT. Its something that's been done before in other codes, and probably even in AT before. I have a supervisor who knows enough about it to help me with it, and is interested that I do it. So why do I go so slowly?
One of the problems is that I see this as an obscure topic that only a few people understand. Further its already been done before, so I am not breaking new ground. So its hard to gather motivation, since there will be few people who can benefit from this and some people who know that its not even new.

What would be valuable about it, is if it allows the code to be easier to use in the process and connects the topic both conceptually and software-wise to a larger community. So, to get myself to do it, I sit in a place which is more well-known and approach the problem with pedagogic bent. Initially, I came to the problem from the beam dynamics side. From here, one is naturally led to the questions of dynamic aperture and lifetime. But the kick maps come from insertion devices which are there for the purpose of creating the x-rays. So now I learn the Radia code, and try to sit on that side. Somehow in the piecing together of these two perspectives, I hope to make it a little more useful or at least understandable, and hopefully not take the rest of my life to do it. (really its not so hard, and perhaps one could say I'm just lazy and procrastinating writing such posts as this)