Podcast "We are TUM" – Transcript, fourth episode

"We have to approach humans as humans and that's something that's very well implemented in video games, but also in advertising and in many other fields, for example in education. And we should be bringing that into computer applications. That's referred to as gamification."

[Moderator Matthias Kirsch:] The woman we just heard is Gudrun Klinker. And there's a good reason why she'll be speaking about video games and gamification in today's podcast: She's professor for Augmented Reality at TU Munich and is also behind the launch of the Games Engineering degree program. In this episode we hear from Gudrun Klinker about why universities will have to become more gamified and why it makes sense to develop video games at a university. Welcome to "We are TUM" – the podcast by and for the Technical University of Munich. My name is Matthias Kirsch and I'll be guiding you through today's podcast. As always the President of the university Thomas Hofmann will get things started by introducing the topics of today's episode to you.

[Thomas Hofmann:] Welcome, dear listener. Let me start off this episode of "We are TUM" with a question: What on earth are video games doing at a university? Not much, you might think, but you'd be wrong. In our section on Cutting-Edge Research we'll speak with Professor Gudrun Klinker about Augmented Reality, the necessity for video game degree programs and we'll address the question of why a leading university also has to be gamified. Then we'll turn to our staff members. Anja Bräunig is this episode's Hidden Champion. She and her team support the appointment of approximately 50 professors every year, helping us bring the best scientific talents from around the world to TUM. They also increasingly help protect us against outside attempts to hire professors away from TUM.

Then as always we'll present to you a start-up founded by TUM students. In this episode we visit Florian Grigoleit. He's the founder of "modelwise", a kind of spell checker for engineering models. He'll tell us about all the things his product can do. And we'll close with our section The Outside Perspective, this time featuring Peter Rösler of the German Speed-Reading Society (Deutsche Gesellschaft fürs Schnell-Lesen). He has five tips for us on how we can read faster and more efficiently. I hope you enjoy listening to this episode of "We are TUM"!

Cutting-edge research

[Kirsch:] Augmented Reality. Extended Reality. Virtual Reality: Terms we've heard with increasing frequency over the past few years, especially in the context of entertainment and video games. But Augmented Reality has many more uses; hardly anyone is as well acquainted with them as Gudrun Klinker. For the last 21 years she's been Professor for Augmented Reality at TU Munich and has set up the Games Engineering degree program. My colleague Marcel Laskus speaks with Gudrun Klinker.

[Marcel Laskus:] Welcome, Professor Klinker.

[Gudrun Klinker:] Hello, nice to join you.

[Laskus:] Let's start with a personal question: What was the last video game you played?

[Klinker:] [Laughs] My students ask me that all the time! Of course you see an older woman sitting here and that says a lot about what games I play. I'm more interested in quiz and board games and the like, but the online variants, sudoku puzzles; I'm not that into the games the younger generation plays. But that doesn't bother me a bit and doesn't keep me from conveying relevant findings to my students on how to generate good 3D games – my background means I'm well versed in Augmented Reality, computer graphics and computer vision.

[Laskus:] To those who aren't that familiar with the field, Games Engineering sounds a lot like playing in the real sense. Why do we need something like that at a university like TUM?

[Klinker:] We launched this degree program because on the one hand the computer game industry profits greatly from the latest modern computers and also helps design them. So there's a very close symbiosis; all young gamers, and also older gamers, know that they need very good graphic cards for their gaming. That means there's a connection, that we know the latest technologies, but we like to take the requirements of this very interesting community, of this industry, into account. That's one way of looking at the situation. On the other hand, the same thing is true for me personally and for other people who work in the field of 3D computer graphics, computer vision and their application, especially Virtual Reality and Augmented Reality: We know we have to work on a user-centric basis, we have to know how the people who use our application think, how they understand information.

[Laskus:] Do video games have an unfairly negative reputation in our society?

[Klinker:] Yes and no. There are certainly several aspects here which get out of control and which are generally not all that welcome, which at least some people don't welcome. On the other hand, it's simply a question of, let's say, psychology. Psychology can be used to manipulate people. When someone wants to manipulate people, it can have both good and bad consequences. If I motivate someone as a teacher, then as my student you could say: She manipulated me again, look here, I did three handstands, and I didn't really want to do that. But conversely, if I consciously hold back and only convey cold hard facts, well, that's not right either. We have to approach humans as humans and that's something that's very well implemented in video games, but also in advertising and in many other fields, for example in education. And we should be bringing that into computer applications. That's referred to as gamification, taking elements from games and placing them in other environments where they may not even be recognizable as games.

[Laskus:] The tone at universities is often highly ambitious, very serious. Could the university world use a touch of gamification?

[Klinker:] Yes, of course, just as in any other institution where knowledge is transferred, for instance in schools. And as I already said, good teachers know how to get people interested in a given thing. How they actually do that can vary widely from case to case. But the same is certainly true at the university and professors should know how to convey course content in a way that makes the students enjoy participating. That's also becoming more and more common today, people are working on a project-oriented basis, in teams, in order to include all these aspects and not just to grind in a bunch of formulas to be spit them back out on a test. Seen that way, we should all gamify.

[Laskus:] As we said, you are a professor for Augmented Reality at TU Munich. Isn't that kind of a game as well?

[Klinker:] No, I wouldn't say that. Originally, if you look back to the 1990s or '80s or even earlier to Ivan Sutherland, the vision of enhanced and virtual reality was really quite different. It was the possibility of giving people abstract knowledge, as compared to tangible things we could hold in our hands, these atoms, so to speak. You can visualize other knowledge, make it available to humans. And you can do that by painting pictures or screening films. But you can also try to blend it in with the reality of the human, the physical reality that surrounds us. Industry 4.0 for example requires a lot of Enhanced Reality in order to support humans on location in performing their work.

[Laskus:] If I, as an interested layman, take a look at everything that Virtual Reality, and Augmented Reality, has to offer, then as you said I find for example the IKEA assembly instructions where my smartphone can support me, there are also video games. But when will I really be able to step into a completely different world, to become part of it with all my senses, not only with my eyes, but with my whole body, my nose, everything at the same time?

[Klinker:] I find it difficult to foretell the future. I see us taking steps in this direction. And as you say, we have a number of senses, with vision the one we understand the best, at least from the point of view of IT. We can also already do an awful lot with acoustics. In terms of tactile sense, I think we're working towards robots that would be able to give us tactile feedback. But when it comes to the olfactory senses, to smell or to taste, we're still very far away from being able to artificially enhance the world in such a way that we don't end up with gigantic systems which are then very confusing and difficult. I'm not really sure. I have the feeling that right now we're making tremendous progress in optics and acoustics, so that Augmented Reality is now at the threshold to the situation where products can really enter the market and also be perceived as such by society. This may be particularly true in the area of computer games, but we should also keep our eyes on the industrial applications. They may not be getting such spectacular attention in the press, but they embody enormous progress in establishing this concept, especially Industry 4.0, how we can support people who aren't sitting at a desk.

[Laskus:] Professor Klinker, thank you very much for speaking with us.

[Klinker:] My pleasure, thank you too.

Hidden Champion

[Kirsch:] TU Munich has over 11,000 employees. Naturally there are occasionally positions which have to be re-staffed, for example professorships. Every year TUM appoints about 50 new professors. And of course other universities have to do the same, they need new people as well and occasionally also attempt to hire talented individuals away from TUM. Anja Bräunig is the woman who keeps just that from happening. She and her professorial appointment team make sure that the Technical University of Munich retains its best people and at the same time attracts renowned colleagues from other universities. She's our Hidden Champion in today's episode; let's listen as she speaks with Clarissa Ruge.

[Clarissa Ruge:] Ms. Bräunig, you're one of the people on the appointment team who is responsible for bringing the best minds to TUM. Who actually picks the individual candidates?

[Anja Bräunig:] The scientists themselves do that. There is always a commission composed of scientists and other members, like gender equality officers and student representatives, who then select the best candidates.

[Ruge:] And what do we have to invest in order to convince a given individual to come to Munich and TUM?

[Bräunig:] Well, it's not all about money, although money of course plays an important role. The personal aspects are important, like salary, but also that the family which usually comes along as well as the scientists' spouses or partners also get accustomed to Munich comfortably. That's becoming more and more important. And the second thing of course is that the facilities here are excellent, that there is enough scientific personnel, that there is good equipment available. And another important thing is that the candidates have a very good scientific environment here, very good, interesting fellow professors with whom they can conduct their research.

[Ruge:] And how do the candidates usually react when TUM comes knocking?

[Bräunig:] Luckily in the great majority of cases, actually in every case, they're very happy when we call. I think TUM has a very good reputation, nationally in any case, but internationally as well. And we usually have the president call first in person and let the individuals know that we would like to appoint them to a professorship. That's very well received by the candidates, since it isn't all that common, even on the international stage.

[Ruge:] How does a procedure like that, which actually takes a lot of time, fit in at a university like TUM, where things can never happen fast enough?

[Bräunig:] Oh, you could say, that doesn't fit at all. But in spite of everything else, we of course want to make a careful selection. And that takes a certain amount of time. Sometimes it's also a good thing to take a little more time, although we could also speed things up here and there. For example maybe in the case of tenure track professors, where there's also a tenure evaluation later where a second quality-based selection can be made, here you could probably shorten the process somewhat. But our possibilities are often very limited because of legal regulations.

[Ruge:] And has there ever been a, let's say, almost bizarre request for facilities?

[Bräunig:] Yes, there was someone who wanted to have a real slate blackboard installed in their office. We were more than ready to do it, but there were other mostly practical reasons why it didn't work out in the end. This individual was to be accommodated in rented office space, and it turned out that a real slate blackboard, which, as I found out, is relatively heavy, would have had to be mounted on a wall that would not have been able to support it.

[Ruge:] What do you personally enjoy most about your job?

[Bräunig:] Handling professorial appointments is a nice job, because it always involves a certain sense of achievement whenever a scientist agrees to join us. But even if someone declines, then you also know the results right away, so it's different from projects which take forever, there is a quick result and that always results in a lot of satisfaction. Especially when the good people agree to come to TUM, which happens in the great majority of cases.

[Ruge:] In your opinion, what would be the most effective way to work even better in your team, together with your team?

[Bräunig:] I think it would be important to grant universities the autonomy to essentially decide on how they would like to design the appointment procedure. Of course we have to follow certain basic principles, maybe legal requirements, that makes sense and is the right thing. But to be able to define the details of the procedure ourselves, that would certainly be an improvement in several aspects.

[Ruge:] When you joined the legal faculty at TUM in 2007 I don't think you expected to still be here in 2021. What's kept you here so long?

[Bräunig:] That's right, in the beginning I couldn't have imagined it, but in the meantime it's been 14 years. It's always been exciting at TUM, I've always been able to continue developing personally; and the topics I've worked on have also constantly developed. TUM is always up to something new, it keeps developing and there are great challenges which I can work on in responsible roles. I can make a difference at TU and there is a lot of freedom to realize your goals, at least in my field.

[Ruge:] Thank you for speaking with us.

The Young Perspective

[Kirsch:] Imagine yourself writing a text at your computer, maybe for a job application, maybe for a course at the university. Suddenly, shortly before you're done, you notice – hey! – the spell checker isn't working. These days that has become practically inconceivable. We're all used to being corrected by our computers when we make a typo or spelling mistake. A functionality like this was missing for a long time in the world of engineers, where the texts are models and planning. Now a former TU student has changed all that: Meet Florian Grigoleit. The idea behind his start-up "modelwise" is to fill exactly this gap. How does that work? My colleague Marcel Laskus speaks with Florian to find out.

[Laskus:] Hello Florian.

[Florian Grigoleit:] Hi Marcel.

[Laskus:] In a recent interview your explained that "modelwise" offers a kind of spell checker, just not for text, but rather for engineering models. What exactly is that?

[Grigoleit:] As you can imagine, in engineering it's of course necessary to check quality properties. Just as proper spelling is the issue with texts, for engineering products it's about quality, reliability and often also about safety. And the same applies for technical products like aircraft or cars: No human can be endangered. Today's engineers verify that by hand, in the future "modelwise" will do it automatically with the help of a computer.

[Laskus:] Could you give us a concrete example of how that works?

[Grigoleit:] Take for example a control unit in a car which controls, let's say, the airbag. There are a number of sensors which check how the car is accelerating. And when a couple of threshold values are exceeded, the control unit triggers the airbag. And if there's a technical defect, for example something as simple as a sensor which is sending an incorrect signal, perhaps due to a short-circuit, then the sensor fires the airbag at the wrong time, which would of course be a safety risk. And the engineer checks each component in the entire airbag system, asking what could go wrong and if something does go wrong, say the sensor sends a bad signal, what would happen then? Of course this entails an enormous number of small inspection tasks which our software can automatically take care of in a matter of seconds or minutes.

[Laskus:] And don't the engineers feel a little bit like their expertise is being called in to question, since you'd think they actually can do all that themselves?

[Grigoleit:] Yes, sometimes we get this kind of reaction, the anxiety that people will be replaced by automation or that their work is unimportant – and naturally there's a little bit of injured pride, too. But as we like to say, our software can take over the spell check, but it doesn't take over the writing. It's just a check, just support, and it takes over the boring testing tasks that most engineers don't want to be bothered with. The computer handles it for them.

[Laskus:] How did it come about that, after finishing your studies at TUM, you didn't just take on what would certainly be a high-paying office job, but instead you decided to start a company in addition to writing your doctoral thesis?

[Grigoleit:] That developed out of the doctoral program. Actually, I already thought I'd finish my studies, earn my doctorate and either I would go into academia or for example do industrial research. But during my doctoral studies my thesis adviser thankfully gave me an enormous amount of freedom. I was allowed to organize my work myself, choose my projects myself and to think about what direction I wanted to develop in. And after a relatively short period of time I started to really like all this autonomy and I didn't want to work inside a box anymore. This made independence a fairly obvious option. And because of the technologies we had in our research group, the idea not only of independence, say as a consultant, but of starting a technology company came up very quickly.

[Laskus:] You've had a long journey here at TU Munich and now with your own start-up "modelwise"; were there low points? Do you have any advice for other founders on how to get out of a slump like that?

[Grigoleit:] One low point is good – There were plenty of them. Of course less so during my studies, but a doctoral program also consists to a large degree of a lot of low points. Looking at "modelwise", 99 percent of a start-up consists of hard work and a very, very large number of setbacks. You get refused by funding sources, turned down by customers, by investors and when nothing moves forward, it doesn't take long before the fighting starts in the team. We had a whole range of discussions with the university about intellectual property rights, a lot went on there and I think there are two important things. One: Remind yourself that this will all pass. It could be the worst possible day in the start-up and you think you just want to give up. And I also had days where I did nothing else but look for jobs online. But that passes and the next day there's a small success, and that's what you really have to keep in mind. The second tip: Celebrate these little successes. Even if you've just been turned down by someone, at some point somebody will say yes. We had just gotten into a start-up program and we were thinking, should we do the program, there were plusses and minuses. And then we saw that there was an acceptance rate of one percent and then the question of "Should we do it?" didn't matter anymore, we were completely for it. Just a simple success, although it doesn't have to be the best solution.

[Laskus:] At what point or according to what criteria can you say, that's a good start-up?

[Grigoleit:] Well, an investor would say a good start-up is one that generates cash. I would say, especially at the beginning when the money is still a long way away, the important thing is the team. If the team doesn't fit well together and the team can't really work well and effectively together, then the best technology or the best product in the world won't be successful.

[Laskus:] And what's the trick to creating a good team in a start-up?

[Grigoleit:] I don't think there is a single formula, but the most important thing in my experience is to really try to take a step back and not only see your own interests, but rather to look for ways to move ahead together. In the initial phase in particular we often had the problem that each of us had very different ideas of where the journey was supposed to take us, who would take on which role and who was expecting what. Because of my internships at companies like Daimler my expectations were perhaps formed too much by the corporate mentality, while my co-founders had a working style which was much more inspired by start-ups, I'd say it was almost anarchic – and bringing those two approaches together was difficult.

[Laskus:] Florian, thank you for speaking with us, all the best for the future!

[Grigoleit:] Thank you, Marcel!

Five Tips

[Kirsch:] We'll conclude today's episode of "We are TUM" with another journey outside the TU universe in our feature "Five Tips". Our guest today is Peter Rösler of the German Speed Reading Society (Deutsche Gesellschaft fürs Schnell-Lesen). You heard right, there really is one. The stated objective of the society is to promote speed-reading, and also to investigate it. Speed-reading of course improves efficiency and Peter Rösler has five tips for us on how each of us can improve in this regard. My colleague Marcel Laskus meets with Peter Rösler.

[Laskus:] Hello Mr. Rösler!

[Rösler:] Hello Mr. Laskus!

[Laskus:] Ten years ago you were one of the founding members of the German Speed-Reading Society. Today the Society has about 100 members and the objective is, of course, to read more quickly and efficiently. Why did you want to read faster, Mr. Rösler?

[Rösler:] All my life I had always been a bit of a fanatic about efficiency. For example in high school I learned ten-finger typing so that I would be able to type faster. And at about the age of 40, when I saw a speed-reading course offered, it made perfect sense to me to take it.

[Laskus:] Today you want to give us five tips on how each of us, students, university staff or anyone else, might be able to read a little faster. I'm looking forward to you sharing your knowledge.

[Rösler:] Taken together, the first four of the five tips are a kind of exercise we use to train people to take their reading speed from the average of about 250 words a minute up to somewhere in the range of 450 words a minute.

#1

The first tip is simple: Read an easy text for ten minutes, as fast as possible. Push yourself! But you still have to understand everything you read.

[Laskus:] And will just any old text do? You say an easy text, what could that be, for example?

[Rösler:] Of course we don't want to take a difficult text in which the reader is tempted to reflect or think about background. That's why simple texts are good, or at least texts from the reader's own area of expertise where you already know all the terminology.

[Laskus:] So for example I could start training with a Grimm's fairy tale to eventually understand specialized literature on quantum physics faster?

[Rösler:] Let's put it this way, you can learn speed-reading with simple texts and then you can read more difficult texts faster after that, in the sense that the language areas of the brain can follow these texts more quickly. But if the specific text in question requires a lot of reflection, then unfortunately the limiting factor is the reflection and not the ability to speed-read.

#2

The second tip: Feel free to allow yourself a couple of seconds of rest after each page. That means you don't have to practice for ten minute straight, but go page by page.

[Laskus:] Why is taking a rest so important?

[Rösler:] The way we understand it now, the trick to this training format is that you make the brain's language areas pick up the tempo just a little bit. That means you have to be at your limit. And it may well be that you can't keep up this full-power level for ten minutes at a time, but rather that you need say a five or ten second break to mentally recover and then start off again.

#3

Tip number three: Repeat this exercise with other simple texts every two to three days. We don't think it makes sense to cram too many exercises into too short a period of time. The brain of course needs sleep at night to digest all the things it learns. So I wouldn't recommend doing more than one exercise a day.

[Laskus:] If I want to get better, faster at reading, then it would make good sense if I just read the texts I have to read anyway for my dissertation or for my studies. Why do you recommend reading entirely different texts, simple texts, and not the texts I wanted to read in the first place?

[Rösler:] Basically there's no reason not to use the texts that you would also otherwise be reading, but if they're difficult and make you want to reflect, then it will be a little more difficult for you to follow this approach to reading as fast as possible.

#4

The fourth tip tells us when it's time to stop doing this exercise. If there's no increase in speed after a couple of weeks, stop the training. It's that simple. We know from experience: The average final value is about 450 words a minute.

[Laskus:] And how can I measure how close I really am to the 450 words a minute mark?

[Rösler:] A typical book page has maybe about 300 words on it. Usually you have to count them. That means you take a novel, count the words on a typical page and then you know that if you take other pages of the novel, then about the same number of words will be on the other pages.

#5

And now we come to the last tip, what's referred to as reading management. And the tip is simple. Use diagonal reading and other types of reading management. Reading management simply means you don't try to increase your reading speed in terms of words per minute, instead you make a targeted choice of what you actually read. And many books are already structured to help you do this. You can see the table of contents, maybe at the beginning of a chapter there's a summary which already tells you that you can skip this chapter, or that there's less risk involved in skipping this chapter. Reading management is always a matter of balancing; sure there's always a small risk that you've skipped over something important. But using focused reading management is still the best way to minimize that risk when you have a lot to read within only a short period of time. Simply starting to read and then stopping when your time is up is a much worse strategy.

[Laskus:] Thank you very much for the interesting discussion!

[Rösler:] My pleasure.

[Kirsch:] And that's it for this episode of "We are TUM". In the next episode we'll once again be showcasing TUM's cutting-edge research, student life and we'll be speaking with all those who make TU Munich the unique place that it is. This has been "We are TUM". This episode was produced by Marcel Laskus, Clarissa Ruge, ProLehre Media and Didactics and by me, Matthias Kirsch. Sound design and post-production by Marco Meister at Edition Meister in Berlin. That's all until the next episode. Make sure to join us and discover the big and little secrets of the Technical University of Munich!

Back to podcast episode