Thesis in LaTeX.

The background.

When I was writing my diploma thesis in 2009, I wanted to do it somewhat ‘special’ and I tried to write it in LaTeX (more precisely in LyX). That time there were so many geeks around Internet that praised LaTeX to be superior to any other typesetting method that I was eventually persuaded to give it a try. That time I failed miserably. Eventually I ended up writing the diploma in OpenOffice (currently LibreOffice).

Four years later I again faced the choice: either to take an easy way and to compile the PhD thesis in MS Word, or to overcome the challenge of steep learning curve of LaTeX.

As you might guess, I took the hard way. What was the motivation? Definitely, part of it was my loyalty to the open software. The other trigger was the thesis from a ‘competitor’ group that was beautifully typesetted in LaTeX. It made me a little envious, so I had to prove to myself that I am no worse than ‘that guy’.

Positive side of LaTeX

I like the idea of open source software. One of my favorite features is the commitment that OSS requires from its users. This commitment makes the users more responsible. It teaches to understand the inner machinery of the program rather than pushing the right button. The same reasoning pushed me to learn R for data analysis (hopefully, more on that later).

I have to admit, LaTeX has quite a steep learning curve. Actually, during writing my thesis, several times I was about to quit because of the desperation and fear that I won’t finish my thesis in desired time. But amount of work that would be needed to transfer everything to another software saved me from the wrong decision. Thankfully, my PhD advisor was very supportive and progressive and agreed to make corrections to my thesis using Adobe reader. It’s one mouse-click more complicated than in MS Word, but for most of the professors it’s already too much of a hurdle.

During writing I could figure out several advantages of using (pdf)LaTeX:

• excellent integration with vector graphics. Since output .pdf file is a vector by itself, one can use other pdf-files as images. In the end I could not stop zooming up the figures to 6400x to check again and again that no loss of the image quality occurs.
  • Add here Inkscape and Gimp for image processing and R/ggplot2 for data analysis and you’ll get a pretty neat output

Tracing bitmaps from PyMOL in Inkscape can make pretty cool vector pictures, too

• numbering of chemical compounds. Once one of my labmates asked me if it’s OK to number compounds 59, 60, 62, 63 … His professor had just decided to exclude #61 from the draft, because it wasn’t tested in bio-assays. I don’t know how the story ended, but the moral is clear: renumbering all the compounds in the text and (more painfully) figures would take hours of useless work in MS Word (which is used by 99% of chemists).

Combination of scatterplot generated by R with LaTeX formatting (via Inkscape) and compound numbering

Disadvantages of LaTeX.

It’s not intuitive. It’s simply not WYSIWYG. That means what you’re writing is not what you will see on the screen when you open your precious pdf document.

Left: what you see; Right: what you get.

If you have never written at least a short peace of code in your life (shame on you!) you would probably be overwhelmed by this difference. If you’re experienced coder you probably won’t notice the difference between two images above. That brings us to the second disadvantage: as any type of coding, writing in LaTeX requires learning by practice. That means time and efforts. Normally people don’t like it, and it’s quite rational. Moreover, as any open source project, LaTeX is nothing but “finished product”. It’s changing all the time – some new packages are popping out from nowhere, some are dying out. That means you have to learn all the time. And google a lot.

Also, it’s not user-friendly, requires lots of trial-and-error and generates significant amount of frustration. Does it sound familiar? Yes, it’s pretty much like doing science.

The results

My final Thesis draft.

After crafting the thesis, making doctoral defense presentation in LaTeX was a joke.

Concluding remark

I fully admit that this unnecessary challenge was yet another face of procrastination. But since procrastination is virtually unavoidable, gaining a new skill is not the worst kind of it, right? I wouldn’t claim that knowledge of LaTeX is in any way essential or even useful per se. But simply keeping one’s mind in a state of learning something new is a nice gymnastics for neurons.