How to write a technical report

There are many good sources on the basics of writing a technical report, see 1, 2 (pdf), 3 (pdf). What I want to emphasize in this blog post is the actual writing process.

I found that it is difficult to write a report after you have done the technical work, feel exhausted, and want to move on to greener pastures (i.e. new technical challenges). The most productive approach is to begin writing before you have started any analysis, i.e. when you don't have a clue what you're going to do. I start by taking notes on the blank report about why I am undertaking this effort, what mathematical tools I might need to develop, what scenarios I might try. Thereby, I make up my mind on how I will approach the problem. That allows me to get over one of the main procrastination reasons, namely not knowing where to start. This makes the report an integral part of the overall picture. The rest is following/updating your outline and filling the blanks (and sometimes realizing you are at a dead end and giving up).

Your report and analysis will be finished at the same time and both of them will be much better than if you had done them separately.

Update 31.10.2012: See In Defense of the Scientific Paper for similar ideas.

p.s. Having more than one computer monitor where you will use one for the report and one for the calculations, is a big productivity boost.

p.p.s For basic English writing rules, see The Elements of Style

music:
* Akrep Nalan - Sarhoş
* Adele - Set Fire to the Rain
* Placebo - Every You, Every Me (covered by Basement Alchemy)