# Anti-Patterns in Presentations¶

I’ve seen a lot of presentations and given a couple ones myself. There are a lot of things I have to work on with my own speaking style. This article is intended to share the patterns that I have identified and work on myself.

Perhaps you first want to read the following lists which contain thoughts from a way more experienced speaker.

## Slides¶

### Legends too small¶

People in the last row still need to read the legends and axis labels on the plots. Therefore they must not be too small. The default size of generated plots is usually too large; their font is too small in relation. Therefore it is important to set the output size to something small, font sizes will increase.

When using pgfplots, one gets this for free, the font sizes are automatically sensible.

### Back & Forth¶

Sometimes there is the need to refer to something that has been said earlier. It is tempting to just go back in the slides. When I see a change in slides, I automatically assume that it is going forward. I am always irritated to see some slide again without the speaker announcing it.

I would try to use an overlay to remind the audience of the previously covered content. It is not as disruptive as flicking through the slides. Also it shows a lot of thought about the structure.

Another irritating thing is advancing to the next slide and going backward again after a second. It looks like the speaker does not know which slide is coming up next. Here a tool like pdfpc can make the talk easier: It displays the slide that will come up next on the laptop screen.

### Introducing axis labels¶

When showing a plot, I need around ten seconds to figure out what is displayed. I need time to read the axis labels and to match the legend to the data points. Ideally the speaker introduces these before starting to interpret any of the data. When the speaker switches to a slide and directly starts interpreting features seen in the data, I will get a bit lost. While trying to read the axis labels (if they are legible), I also need to listen to the interpretation.

### Slide numbers¶

Having no slide numbers make the slides really clean. The audience might want to take notes and come back to some of the slides. It is a nice help to give them slide numbers, then the audience can add the numbers to their notes.

### Orhaned text¶

As a listener, I expect that everything on a slide will be covered in the talk. The text on the slide is meant to help the understanding, but the talk should be comprehensible without reading the bullet points.

Whenever there are bullet points on the slide that are not referred to, I am a bit list. Was I supposed to read that content? If yes, when was the right time?

The other extreme is fading in each bullet point at a time. If this is done with precise timing, it can help to guide the audience. If it is done at random, I find it more distracting. Each time the slide changes, I will look at it. If bullet points are uncovered, my attention perhaps should have remained on the speaker.

### Contrast¶

One talk was in a building without blinds. The projector was not really powerful, so the only colors that one could distinguish were black and white. The slides were designed with a much better situation in mind. They had dark gray text on light gray background. It was really hard to read, though it probably looked very nice on the computer screen.

### Power management¶

A common setting for laptops is that the screen gets disabled after a few minutes of inactivity. This is very handy to extend the battery life. During a talk, this usually disables the projector again. The audience will just see “Searching for signal” on the screen, and it might take a minute to get the slides back.

Disable the power management features while giving a presentation. For the X server, this could be done like so:

xset s off
xset -dpms


KDE has a handy option in the battery tray. Most presets have allow much longer phases of inactivity when the power supply is connected, this might also be a viable option.

Similarly, automatic updates on Windows can be configured in such a way that it nags you to restart the system. If it actually does that during a presentation, one has to work without slides for at least two minutes or even until the remainder of the talk. Make sure that automatic updates are suspended. Windows 10 offers the definition of core working hours, it will not nag you then.

## Speaking¶

### Checking off bullet points¶

Having a talk that is just a reading of bullet points is bad enough. What I find also irritating is a speaker that starts off freely and talks about various points. Then that speaker looks at the slide and checks off the points with “yep, said that, yes, covered that”. It gives the impression that the slide has no connection to the talk and merely serves as notes for the speaker.

It would be better to just not say anything when one sees that all the points are covered. Just advancing to the next slide is what I would expect in the audience.

### Thinking noises¶

Memorizing everything in a talk is hard. It takes a lot of preparation and a few rehearsals to have everything formulated and ready to say. However, it makes a visible difference to the audience. Knowing what to say makes the whole talk more pleasant to listen to. It is irritating when the speaker starts to make obvious thinking noises to cover up.

Having small pauses in a talk can emphasize the key points. If there is a need to reformulate, just say nothing. The audience will be busy looking at the slides and perhaps not even realize that a pause is not intentional.

## Body Language¶

### Overuse of pointing device¶

Most presentations are held with a laser pointer or some pointing stick. This can be great to lead the audience’s attention to some detail on the slide. I find it mostly annoying when it is overused. There is no need to point to the title of a slide.

The problem with a laser pointer is usually the amplification of an unsteady hand. On the projection canvas, the laser dot will wiggle around. Also it is not really easy to aim from the hip without practice. Therefore I prefer to use it very sparingly. If the canvas is in reach, I often try to just use my hand for pointing. It is more natural and slower; this prevents me from overusing pointing.

### Not using a remote control¶

Try to get a remote control when delivering the talk. This will allow you to move in the room freely. Otherwise one has to return to the laptop for every single slide change.