Simple tools

Scientific visualization has a reputation for complex 3D graphics and animations. In real life most of the graphs made by scientists are relatively simple 2D plots that try to clearly illustrate a point. What do you want out of such a workhorse tool?

1. Multi-platform.
2. Reproducible results between data sets. Scriptable.
3. Must produce at least high quality Postcript or PDF.
4. Non proprietary formats. Separation of data and plotting commands.
5. Possibility of creation of complex labelling.
6. Extensible, open platform.
7. Free :)
8. Easy to learn.

Number 7 is not strictly necessary, but it is nice, particularly for funds starved research groups.

One of the best tools that I have found that fulfills most of these requirements is gnuplot. I started using gnuplot quite a while ago in pure desperation after a particularly unproductive session with a commercial product that shall remain unnamed (origin). It took me hours to obtain a combination of graphics with consistent font sizes and when I got there and tried to generate an eps, it was 15MB! I looked for an alternative and found gnuplot.

Visualizing Science

Some thoughts and tools on clear, beautiful and efficient visualization tools for Science. This is an excuse to learn about best practices and what is out there in this field.

I will also port the contents of my previous gnuplot tips page over here given that it is completely messed it up since the phasing out googlepages.