26 August 2008

Tools of the trade

In my job as a professional science writer, I use no less than eight different software packages to get the job done. And that number is tending to go up, not down.

On the writing side, I use word processor, obviously. I'm using Microsoft Word for most of my work, but lately I've started using Google Docs when I have to collaborate with students.

A word processor alone doesn't help much with the task of references, though, so software package number 2 is a specialized reference manager; I'm currently using EndNote.

Now that I have the words under control, I need to deal with numbers. Microsoft Excel is great for data entry and simple manipulation, but again, I'm moving towards Google Docs for certain tasks. But Excel can't do the sort of high end analysis that I need as a scientist. So I have a statistics package (currently SPSS), and scientific graphing software (currently Origin), bringing the total to five.

I find I also need two kinds of graphics software: one for vector based images, and one for a pixel based images. I've been using Corel Draw since version 3, so I'm going to continue to use it for as long as I can for the vector images. Luckily, the package also includes Corel Photo-Paint for the pixel images.

I could, strictly speaking, stop there. But I consider posters a sort of manuscript, and I could use my graphics packages for making posters -- and have done so in the past. But I've found a simple desktop publishing program can actually be faster and easier. Microsoft Publisher is little known but very useful in this regard.

And that's what I use just for mastering my manuscripts, never mind presentations or websites or other odds and ends.


Anonymous said...

There's a (kind of) danger in the system of software packages you've chosen to use. Every one of them is proprietary, which may not seem like a big deal but not having scientific data in open formats really is a big deal. Consider how many times SPSS has essentially made your data at least temporarily unavailable to you while you jumped through the hoops of their mandatory upgrades. What's to say one day they decide to do something more imposing or even go away altogether for unrelated reasons (a bad economy?). I wouldn't say you should worry about going completely with open-source software, but you should be aware of your options. For writing and formatting and references (and even presentations now) not much beats LaTeX. There's a learning curve to it but you're a scientist!--you'll get it. For statistics and for much better graphics than SPSS or Excel you should consider R. It's command-line based and everybody who's never used a command-line thinks it's ugly and hard. But you really have never used a computer until you understand what kind of an open horizon the command-line really indicates. It is much, much more controllable. And consider this: R is a derivative of S, the statistics program that John Tukey himself had a hand in developing. And everything I've mentioned so far is free!--forever and always. No mandatory uploads, no licenses, no proprietary data that you own but can't get at. You can work with your data in text files!

I have a pretty streamlined system that I use for my scientific writing that I won't go into here. But I've been meaning to write about it on my own blog. I'll come back here and comment when I've posted it. But one quid pro quo is that I use a Mac so if you use a PC I don't necessarily know what some of the equivalences are. However, if you are on a Mac, start by taking a look at Scrivener--http://www.literatureandlatte.com/scrivener.html. You won't write the same way again.

Zen Faulkes said...

Thanks for the evangelism! For the moment, my system works for me, and I do not pretend that it is the best or only system.