The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair.
– Douglas Adams, Hitch Hikers Guide to the Galaxy
I was asked recently `How do you focus and manage to finish tasks?’ and I wanted to take some time to think about this. I know I do have methods which I follow, to some extent, I’m just not always thinking of them when I go about my day.
Much of my practices have come from a lot of trial and error to find how I work best, so they are certainly not general but they do work for me. I also enjoy talking with others about how they work. A nice draft of similar ideas is available on Terry Tao’s blog. I will probably re-state some of the exact same sentiments.
Context
To give some perspective, I am a graduate student studying in the field of Atomic physics and find quite often that my day to day activities require an overwhelming amount of skills / knowledge I don’t yet possess. So one is faced with the challenge of determining how much time to invest learning / acquiring before beginning the task at hand. For example, If I need to diagnose a problem with a laser there are many subsystems that one needs to consider
- Lasing material- diode, dye, pumped crystal or fiber
- Control electronics – temperature controllers, current controllers, feedback controller. noise, bandwidth, resolution, range.
- feedback mechanisms – diode current, optical power, temperature. bandwidth, advantage / disadvantage.
- noise pathways – electronic, mechanical, pressure, temperature. frequency response to each, magnitude.
So I need to know something about basic electronics, circuit analysis, integrated components, noise, standard building practices. Something about mechanical and temperature stability, material properties, basic statics analysis, standard design and building practices. Something about optics, optical properties of materials, diode frequency / power temperature/current dependence, standard practices. Of course much of this is reusable, I need to know these things for analyzing many problems that I’m likely to be concerned with. So it makes sense to spend a great deal of time and effort to understand them thoroughly. BUT, thoroughly is still very ambiguous since any one of these problems could merit a lifetime of study. So for many things I stop at a working knowledge, mainly being a number of rules of thumb I can use to get through these tasks, and pick topics that I have a particular interest in to specialize.
On time managment
Its of course wise to manage your time well and there are many ways to waste time and its important to realize what they are. For instance, most nights I’ll watch a few TV shows or a movie but I don’t consider this a waste of time even if I do have other work to do. You will always have `other work to do’ but its important to set aside time to relax and forget about the problem at hand for a moment. This will prevent you from getting stuck in a rut, repeating the same train of thought over and over as well as giving a refreshing break. So be sure you don’t ignore the utility of recreation.
Most of the time I considered wasted is due to an improper analysis of the difficulty of a task. Such that a lot of time is invested without anything to show for it. One can spend an endless amount of time iterating without direction, likely without success or gain. The majority of time on a project should be spent on planning and documenting, if these are done well the actual implementation should be minimal (aside from a few tedious and time consuming tasks that may occur).
Proper analysis of a project can take considerable effort and I find that I am rarely motivated enough to do so. In this way its useful to identify work that requires varying / different amount of motivation and effort. This takes some juggling and isn’t always possible due to deadlines, conflicts, dependencies and a mess of other `life’. I try to constantly have a handful of projects on-going, meaning I intentionally take on more than I intend to immediately work on. This way if I feel up for analysis I can get through a couple projects and this provides a lot of smaller low motivation / effort tasks that can fill in my `down time’. This also has the benefit of being able to delegate some of these tasks to others.
On choice
Before you even commit to thinking about how to complete a project you should think seriously about the `why’. This is critical to whether the project will ultimately be completed or not. You should try to keep a common theme or somehow make sure each individual project contributes to a larger whole. In my case, for most things I do I am under the impression that they contribute to being a better physicist, which in turn I believe is my best chance at contributing something useful and lasting to society.
It should also be the case that you find both the project and the process of implementing it interesting and rewarding. If you are only interested in the end result it will be much more difficult to complete the project successfully. Hopefully the project is not only rewarding in a self-fulfilling way, but provides some utility beyond the project for projects sake. For instance, don’t just make a noise triggered switch, make your own clapper.
Most importantly these projects should be within your current `skill level’. Meaning you posses all skills needed to properly analyze and plan the project. If not, then you should think much harder about stepping stone projects to put your self in a better position.
On planning
Planning is certainly the most difficult part of any project. Luckily, there are many that follow through with the documenting part of projects and a lot of great guides can be found online, e.g. How to build a hover craft.
The best place to start on planning any project is identifying how to break it up into smaller more or less self-contained projects. This has the added benefit of making progress noticeable. Each of these should be analyzed such that there is a clear plan and ordering of tasks needed to complete the project. I believe there is no unsolvable problems, only under researched problems. There should be no doubt about the possibility of completing the project and a definite time frame in mind for its completion. Have definite expectations and requirements. For instance, you should know you need between 5ft.lbs and 10ft.lbs of torque NOT about 7ft.lbs.
On investment
As part of planning, this should include investing in skills that will increase your chance of success or improve your efficiency. There is of course a fine line here to moving toward your goal and walking off on a tangent project. My method for avoiding this is to use your original task as an exercise to practice your new knowledge, keep your original task as the highest priority. If I wanted to program a gui for something, I might spend sometime installing glade and following a few tutorials but I would next try to build the gui I wanted WHILE reading details on specific functions as I need them. After I completed the gui, I may continue to read and master the potential of the program but this tangent would not now distract me from my original task.
These side/investment projects will to a large extent determine how successful your future projects will be and increase the range of projects available to you.