“…14-year-olds … would never imagine that a professional scientist would spend so much time drawing, sketching, and writing.”
I spend a lot of my time working on figures and illustrations. More than the person-on-the-street might imagine. During GCSEs and A-levels I didn’t see the point of the canonical “Draw a labelled diagram” but now it’s the sort of thing I often end up writing as comments on other peoples’ work. Making these sorts of illustrations can take a lot of work – especially to make something that is publication quality which conveys some important point. In my opinion a picture conveys a thousand words isn’t a strong enough statement when communicating science, as long as its a good picture. If you’ve ever tried to describe a picture in your head for someone to draw out you’ll know how difficult it is to convey visual ideas in words.
Publishers want graphics in a vector format – so they can be scaled without loss of quality – which usually means PDF or EPS (Encapsulated Postscript). They often have a long set of guidelines for the production of figures, for example journals published by AGU, Springer, and Elsevier. For line drawings I typically use OpenOffice Draw, although I haven’t found its EPS export to be that great, and Inkscape. But I know plenty of people who use Adobe Illustrator and Corel Draw. Often a hybrid solution is called for – so rather than trying to draw some complicated shape using Bezier curves I will make a plot and then import that into Inkscape and merge it in with the rest of the figure. This is the sort of thing that can consume time. For prettier 3D graphics I’ve taken to using Pov-ray – this is what I used to make the figure on the right – but since it isn’t in a vector format the image needs to be rendered at a high enough resolution for print.
Back as an undergrad we had a lecture on not using Excel to make our plots – the main reason being the lack of control and the effort required to produce a publication quality figure – dealing with similar journal issues as described above. I spend a lot of my time writing computer programmes to make plots for papers, using tools such as Matlab, IDL, and I know lots of colleagues who use PGPLOT. Back as an undergrad make my plots by sending plot commands to a vector drawing package on the Amiga using a scripting language (ARexx). The advantage of these approaches is that if you have some sort of script to make your figure then it’s a simple matter to change some aspect of the figure and then redraw it. Similar effort goes into making beautiful images that we enjoy from the Hubble Space Telescope. “Hubble images are made, not born.” You can see this at work in a movie made for the Hubble YouTube channel which shows someone producing an image of NGC 3982 over the course of three weeks (in time-lapse of course).
Throughout my scientific eduction I’ve been encouraged to draw. Throughout my software engineering training I had to draw flow charts, PERT charts, Gantt charts, JSP diagrams, class diagrams, use cases, and more. My PhD supervisor encouraged me to sketch in order to explore physical problems – especially those in three-dimensions. I can also still recall my work experience placement, seeing A3 sheets of paper stuck to desks that would slowly fill up with doodles, sketches, notes, phone numbers etc. over the course of a working week. I do something similar now. I also use drawing/sketching/doodling as a tool for thinking, laying out ideas, and considering different approaches to problems. In the course of writing this blog post I came across another blog post on drawing as a tool for thinking and largely agree with what was written there – although my sketching tends to be more freeform. Unstructured sketching anyone?? Sketching is something I encourage in my students although I worry about cognitive styles different to my own.
Of course these aspects contain different approaches, or rationales for drawing. The first is as a cognitive aid, the other is to communicate ideas. Sometimes my thinking-related sketches will somehow end up forming the core of an illustration for a paper, other times they might just be another page in my lab book.
Looking back on my education, a GCSE in Art or Graphic Design (both offered by the secondary school I attended) would have been useful in my future career. Just like more emphasis on writing. But 14-year-olds who are interested in computers and science would never imagine that a professional scientist would spend so much time drawing, sketching, and writing.