Richard Feynman in the famous book “You are, of course, joking, Mr. Feynman” shared his way to quickly understand and evaluate new things:

I came up with a scheme that I use to this day, when someone explains something to me, and I try to understand it: I create examples. For example, mathematicians enter a room in an extremely excited state with a stunning theorem. As long as they tell me the conditions of this theorem, I build something in my mind that fits all its conditions. It's easy: you have a set (one ball), two non-intersecting sets (two balls). Then, as the number of conditions grows, my balls get their color, their hair grows or something else grows. Finally, mathematicians give out some kind of stupid ball theorem that doesn’t fit my hairy green ball at all. Then I say: "False!"

Nobody would accuse Feynman of being incapable of abstract thinking, but even he found it easier and clearer to look at particular cases first. When we speak, the same approach may be relevant for many of our listeners. Let us give them the opportunity to listen to us as they are easier and more understandable. In this article I want to highlight a few properties of good examples, knowledge of which will help to make your performances clear and memorable.

The need for examples for success

The examples in our presentation play several important roles at once. When working with IT speakers at trainings or when preparing reports at a conference, I discovered an amazing fact: many people don’t understand the importance of examples and don’t even think about them. At programmer conferences this is a noticeable problem that I would like to gradually solve. So, what are the purposes of the examples?

Explanation and entertainment

In the audience at the conference almost always there are several groups of people whose attention you can easily lose during the talk:

• Those to whom everything you are talking about is simple and clear. They quickly go to the phone, never to return from there.
• Those who, on the contrary, with great difficulty watch the course of your thought. Perhaps the topic for them is too new or exceeds their intellectual capabilities. These, too, quickly go to the phone.
• Finally, people who, in principle, are not very interested in the topic, and they wandered into the report by mistake or by despair. After all, when you attended conferences, you probably noticed that uninteresting reports often go in parallel? This is the case.

Examples help to make sure that not all of these people fall asleep in the course of the story. To those who find it difficult, our examples add clarity exactly in the form about which Feynman wrote. All other examples, at least, entertain. Without denying the fact that there is often a lot of value in abstract reasoning, I’ll note that well-chosen examples help the audience to assimilate this food correctly.

')

Proof

Induction (the transition from the particular to the general) is generally characterized by people much more than deduction (the transition from the general to the particular), which was the basis for the successful business of Sherlock Holmes. Even those who consider themselves thinking and objective are still inclined to make generalizations based on insufficient data. The more interesting the story, the easier it is for students to summarize its findings on all life situations.

Doesn't it just work in sales presentations?

It would be a mistake to think © that IT is something different from other areas and it is more difficult or unnecessary to give convincing examples here. Below I will try (by examples, how else) to show how this is done in a technical presentation, and at the same time to formulate signs of a good example. I hope they will be useful to you.

A good example is paradoxical.

A good example is casting an intellectual challenge to the audience. When something obvious is suddenly wrong, it is always remembered and often serves as food for subsequent analysis and self-search. Take the example from Alexey Shipilyov’s speech on the hashcode java.lang.String:





The fragment of interest to us lasts two minutes and begins at 19:53, here is a link to youtube with the exact moment of the beginning .



Two lines of the same length and composition of characters, being placed in the same benchmark that repeatedly computes their hashcode, produce results that are radically different in speed. WTF? - only the viewer has time to think, and then everything becomes clear: it turns out that the hashcode is cached, but if it is zero, the system considers that it has not been calculated yet. Thus, with a probability of 2–32, the line is not lucky, and its hashcode, equal to zero, will be recalculated every time it is called. It was just such an unlucky line that turned out to be the “over-neglected tool”.



I ask you to pay attention to the following: if you tell this feature of string implementation without examples, it will almost certainly be quickly forgotten. An example task is very good at remembering. People generally love to think, especially if nothing depends on the result of their thoughts. This can and should be used.



Often, an ordinary example can be turned into a paradox by changing the order of presentation: first tell the symptoms, and then explain where they came from. It is always more interesting!

A good example is large

A story from the life of a large company that cost (or, on the contrary, brought) a lot of money, or at least served as a cause of a career turn of a famous person, at least somehow resonant - is better than a story about an unknown Vasya Pupkin. Very well, if you manage to show these events from a new angle.



Look at the famous events version of 41 shades of blue in Jason Cohen’s presentation:



The fragment of interest to us lasts 4 minutes and begins at 18:42, here is the exact link to the beginning .







This is a story about Google, the majestic and beautiful Marissa Mayer is a fool, and the billions of Google users are experimental rabbits. It is both ambitious and paradoxical. And when a speaker calls on listeners not to repeat Google’s mistake and not to try to find a winning combination of parameters by going over the A / B tests, other evidence seems to be unnecessary to people. If you can illustrate your idea with a large-scale example, it is better to do so.

A good example intersects with reality.

Details from real life lend credibility to the story; they serve as anchors to help focus attention.



Recently, I watched and helped prepare several presentations on the topic of synchronization and parallel flow control primitives. The natural task that comes up with this is the task of readers and writers.





The task is fundamental, but the example of using synchronization primitives from it is bad. When you in your practical work, and not in the classroom, this task had to be solved for the last time? What are her manifestations in real life? Of course, the implementation of the file system is exactly they, the reader-writers. When was the last time you wrote your file system?



Another known case is ticket booking systems. Watch the available tickets can be all, and book - only someone one at a time, after which the booked option disappears from view for the rest. But nowadays all this is hidden in the depths of the database, again, we do not do it ourselves.



In other words, on variations of the task of readers-writers, the audience very quickly falls asleep, even realizing the fundamental nature of the task. By the way, I would be grateful for references to current manifestations of this task in real life.



In the same context inversion of priorities is often mentioned, and there’s a good example about it. In 1997, the Pathfinder rover computer suddenly began to go into reboot, while it could not complete the cycle of collecting and sending the next piece of data from research sensors.

Inverting priorities is a fairly simple concept, but it is still reasonable to follow it with an example. And you can attach beautiful photos to this particular example. Those who have been puzzled all this time, what kind of landscape is depicted on the starting picture and how it relates to the topic of presentations, now they will find out that this is one of the pictures taken by Pathfinder. Possible disadvantage: is this story too jaded? All of it took place in the second and third year, or did it turn out to be news to someone?



In general, a concrete example, with appearances, passwords and photos of participants is always better than the same, but abstract.

Conclusions and usefulness

Every significant concept you are telling is worth giving an example. It’s not a fact that it should have all three described features, but the more the better. This is not a waste of time, but a vital part of the mechanism by which your listeners will understand you. Good luck to everyone, speak clearly!

Bonus for attentive readers who got to the end

Have you noticed a mistake in the reasoning of Jason Cohen?

Links to topics mentioned in the text

What Happened on Mars: Notes from Pathfinder Launch Participants One , two (it seems to me alone that they contradict each other? ..)

Who is Jason Cohen .

Article Ron Kohavi comrades, which mentions successful experiments with the color of the links.