Monday, April 9, 2012

On why I hate LaTeX/Beamer

Continuing with a last's year post, I make a case in point about how using LaTeX/Beamer is bad for scientific presentations.  

Last year I wrote about LaTeX/Beamer and how it was worse than bad PowerPoint. One or two weeks ago I found a presentation on Slideshare done in Beamer, by people I know. They are excellent scientists at Germany's best research center. It is partly not their fault, they are applied mathematicians, and in mathematics LaTeX and, to a lesser degree Beamer,  are industry standards.
The problem of using Beamer for scientific talks is that it employs the same narrative of a written essay: Title, introduction, first section, second subsection, conclusion. Another problem is that Beamer enforces the use of off-the-shelf templates, and the command-line nature of Beamer makes its customization very time consuming. The use of predefined templates one of the causes of  the death by PowerPoint. The creation of a math talk using Beamer is a soporific bomb.
To show what I mean, I extracted the layout of 4 slides of the presentation I found on Slideshare. I know that designing engaging visuals for the mathematical sciences is difficult. Not only is math hard to visualize, but the use of equations and mathematical symbols (which is a just) can easily obscure a talk's slides. 
First, the title slide:

Problem 1. As I mentioned on my last post, the visual follows the old paradigm of writing the name, place, and date of the conference. I get that presenters want to write this information for the people that will see their slides after their talk. If that's the intent, one could also write this information on the link that downloads the visuals, or in a description field or presenter's notes field, if provided. If none of the above is possible,  another option is to create a text file with this information and pack these two files in a zip.
Problem 2. Another problem here is that the name of the authors and their affiliation is written twice (what for?), one in the title part and one on the bottom left and bottom center-right as part of the "status bar".
Problem 3. No sense of design. The big dot and the little white square are the institution's logos, observe how they completely disconnected. The text on the top containing the name, place and date of the conference is floating. All in all, the objects are randomly thrown into the page, with no hierarchy, alignment or proximity.
Now let's go into the content:

Problem 1. This is their second slide and the "navigation bar" makes its first appearance.  This bar contains the sections of the talk and marks its progress. This is not the way of helping orientating people back to your talk, it is a way of torture them if a talk is bad and feel endless. This bar is useless table of content and is one of the things that make Beamer worse than PowerPoint. It uncovers that Beamer is spawned out of a linear, essay-like narrative.
Problem 2. Not only the top and bottom bars are distracting, there is a second top bar with the title of the slide, and to the very right, the dot is (again) their logo.  All of this courtesy of a template...
Problem 3. Beamer offers a "boxes" that are supposed to create focus by containing important information. Well if this is so, why let the important information compete with the item on the bottom?
Next, the monster slide:

This slide shows a math theorem...just no. Don't do it like this, please. In case you have  to show a theorem,  read chapter 10, "Writing a talk" of Handbook of Writing for the Mathematical Sciences, second edition, by Nicholas Higham.  Be aware that this slide is not Beamer fault, but the author's.
I'll finish with their last conclusions and thank-you-for-your-attention slide:

This bullet-point slide is, sadly, the bread and butter of today's scientific presentations. We are still a long way to create visuals that have the same quality standard of our research and writing communication. Unfortunately Beamer doesn't help us to improve our slides, quite the opposite, as  we have seen.

So that's why I hate Beamer, because it doesn't foster creativity,  forces to think inside-the-box, and ultimately I can only do visuals of the past. But, if you think there is no other way to make visuals for a math talk, click here, it might help you get some inspiration.


  1. We have to think something about the theorem issue. If it is a mathematical presentation, the most important contribution is indeed a theorem. Think about a math thesis presentation: most results will be theorems :(

    1. I agree that in a math presentations theorems have a big role. The question is how you present them. As University of Manchester Mathematics Professor Nicholas Higham in "Handbook of Writing for the Mathemetical sciences write" writes "When you write a slide, aim for economy of words. Chop sentences mercilessly to leave the bare minimum that is readily comprehensible." Think about it, people may take their theorems directly out of their latex articles and dump them into their beamer slides without further thought. That is that Beamer fosters, but ultimately it Beamer is just a tool. As I wrote on the post, the responsibility lies on the person using it.