Why we need to reinvent media - and how
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 275 | |
| Author | ||
| Contributors | ||
| License | CC Attribution 4.0 International: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/51965 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
| |
| Keywords |
00:00
Content (media)Data miningSocial softwareData modelBit rateContent (media)Computing platformMultiplication signExterior algebraMotion captureService (economics)Noise (electronics)Computing platformData modelAreaVideo gameMoment (mathematics)Similarity (geometry)BitGoodness of fitHypermediaWordSound effectBusiness modelData miningOrder (biology)Type theory1 (number)CASE <Informatik>Group actionRight angleInformationArithmetic meanAlgorithmOnline helpTwitterMereologyBit rateStreaming mediaGreatest elementComputer animation
06:50
Content (media)Data miningSocial softwareData modelBit rateInheritance (object-oriented programming)HypermediaPhysical systemComputing platformNoise (electronics)Content (media)FreewareLevel (video gaming)Motion capturePoint (geometry)Order (biology)Computing platformOpen setMultiplication signBitMereologyGoodness of fitPhysical systemInstance (computer science)Data modelSingle-precision floating-point formatSummierbarkeitInformationInternet service providerDecision theoryArithmetic meanTransformation (genetics)Business modelVector potentialSound effectSlide ruleGreatest element1 (number)Graph coloringState of matterPower (physics)Principal idealInternetworkingWeb 2.0Shared memoryDifferent (Kate Ryan album)ArmComputer animation
13:40
Content (media)Computing platformEquals signFlow separationTransformation (genetics)FeedbackLevel (video gaming)Direction (geometry)Transformation (genetics)Decision theoryFreewareShared memoryProcess (computing)Flow separationPower (physics)Different (Kate Ryan album)Content (media)Computing platformService (economics)Arithmetic meanMereologyPhysical systemProjective planeSound effectNoise (electronics)WebsiteMoment (mathematics)Equaliser (mathematics)Instance (computer science)Address spaceWordRandomizationVideoconferencingMessage passingInternet service providerWeb 2.0Video gameData modelArmState of matterKeyboard shortcutEmailSynchronizationCountingFile viewerPlanningRight angleFeedbackTouchscreenComputer animation
20:31
Flow separationEquals signTransformation (genetics)Content (media)RandomizationContent (media)Element (mathematics)Level (video gaming)Physical system2 (number)Computing platformSubsetData modelVotingMultiplication signInformationWeb 2.0BitType theoryCASE <Informatik>RobotCore dumpProjective planeRight angleArithmetic meanCircleMereologyGreatest elementGoodness of fitClassical physicsGame controllerTransformation (genetics)Noise (electronics)Moment (mathematics)Metropolitan area networkFilter <Stochastik>State of matterTouchscreenMeeting/Interview
28:39
Level (video gaming)Content (media)Physical systemRobotMultiplication signGoodness of fitVideoconferencingArithmetic meanSpeech synthesisCASE <Informatik>Order (biology)BitVulnerability (computing)Decision theoryComputing platformInstance (computer science)Functional (mathematics)Traffic reportingSet (mathematics)RandomizationRight angleSound effectCentralizer and normalizerComputer clusterMoment (mathematics)Video gameMeeting/Interview
32:09
Axiom of choiceCASE <Informatik>HypermediaComputing platformMultiplication signBitInformation securityLevel (video gaming)Form (programming)Vulnerability (computing)Meeting/Interview
33:10
Flow separationEquals signTransformation (genetics)Computing platformContent (media)Multiplication signTransformation (genetics)Level (video gaming)Order (biology)State of matterAxiom of choiceLine (geometry)Process (computing)Data modelPhysical systemComputing platformNetwork topologyWater vaporSimilarity (geometry)Decision tree learningDecision theoryBitFunctional (mathematics)Control flowContent (media)Link (knot theory)WeightTwitter1 (number)Group actionComputer animation
36:37
Computer animation
Transcript: English(auto-generated)
00:26
First of all, welcome to the live stream of the day. We're here in Zurich. We're here today to talk about the live stream of the day. We're here today to talk about the live stream of the day.
00:43
My first guest today is Vernon, and we're here today to talk about the live stream of the day. How things change in media. More precisely, why we need to reinvent media and how.
01:03
Now I'm looking forward to learn more about this. Thank you very much, Daniel. Hello everyone. It's a great pleasure and great honor to be able to speak in front of you, even if it's remotely and I don't see you directly. Who am I? Hi, my name is Vardon and I have a problem.
01:24
My problem looks a little bit like that. I consume a lot of content, media that feels like noise, content that doesn't have a lasting impact on me, content that maybe targets my emotion and feels good in the moment,
01:44
similar to junk food that also feels good in the moment, but it just makes you feel empty afterwards. So content that is really, we call it clickbait, content that doesn't have any substance, content that is really like junk food.
02:04
And I ask myself often what would the world look like if we had, the most prominent content was actually content that is transformational, content that makes you think, content that has this kind of substance.
02:23
I call it sometimes butterfly content that can have a butterfly effect on your life. I'm sure all of you have experienced that, that you have read something or watched something or listened to something that has really changed how you maybe how you look at something or has really given you a new insight into a topic.
02:41
So it has really been transformational to you. I ask myself what would the world look like if we had more of this type of content, if this type of content was the most prominent and not the junk food content. In order to answer that question, what would the world look like in such a case, we have to take a step back and ask ourselves, well, why is it not the case today?
03:05
Why is the world not like that today? Why does it seem like the content that screams the loudest is the one that is heard most? And to answer the question why is it not the case today, we have to look or to let's say follow the money
03:21
and look at the business models of the places where we consume media today. And there are two business models that are the most common ones. The most common is the advertisement business model, which you see on this illustration. And you see simplified how this model works.
03:41
So at the top you have an intermediary, which can be a platform or an online newspaper for example. And at the bottom left you have the advertisers and at the bottom right the users and creators. So what happens is that the advertisers pay the platform
04:01
and the platform provides free service with that money to the users. Why free in brackets? Because we don't pay anything financially but we pay the platform as users with our attention and with clicks and that's what the platform gives back to the advertisers.
04:22
So the advertisers pay the platform in order to get our attention and our clicks. And so what's the alternative today? The alternative today are paywalls and paywalls work like that. Again you have at the top an intermediary but this time the users and the creators are split into two groups
04:42
and this paywall model is mostly used with online newspapers for example where the creators are journalists for example. So the journalists they create content for the platform, the platform gives that content to us the users and we pay something to the platform. The platform takes that money and gives a little bit of that
05:04
to the creators back to the journalists. So it's a classic intermediary model. Now let's look at the good and bad sides of those two models. The good and bad sides about the advertisement model. Well first a good thing is that it's very inclusive
05:20
in the sense that you don't have to pay money in order to get access to the information. You don't have to be financially invested. You can be a part of the platform no matter who you are. And you can also be a user and a creator at the same time. So you can publish on the platform as well as consume content.
05:43
And it's openly accessible generally so if you think about Twitter the information is not behind the paywall. The information is generally openly accessible for everyone. That's the good sides but the bad sides obviously heavily outweighed and they are tied generally to the click rate business model.
06:03
Meaning that the business model has this incentive of generating as much clicks as possible, getting as much attention as possible. So that leads to low quality content and to attention and data mining. Why? Because it's simply not important
06:23
how high the quality of the content is as long as it makes us click, as long as it grabs our attention. So it's actually that's why we get clickbait content. Because there's a financial incentive for the platform to do that and to mine our attention and our data.
06:41
Meaning that the more data the platform has on us the better they can predict what we will click on next. And also obviously they work with the help of algorithms to find out how to best capture our attention. That's what makes those platforms so addictive. It's just the business model that provides the incentive for the platforms.
07:04
It's not that they are evil or something. It's really their financial interest to do that. And that obviously can lead to huge problems because it can be used also for political purposes. All kinds of people or organizations can use this model to pay
07:22
in order to push their content. And that's how we got to the point where we are on a societal level today. This race to the bottom is exactly because of that. So on top of that one of the downsides is that the creators don't get any payment or any financial reward. So they also don't have an incentive to provide high quality content.
07:44
And the platforms that use advertisement-based models are usually centralized. So you have a central instance that is being paid by the advertisers. Now the good and bad sides about paywalls. Well, first it's good that the quality of the content is higher because actually someone is willing to spend money on the content.
08:05
So that's one good thing. And the other good thing is that a part of that money goes to the creators. So for example the journalists are then paid with the money that people provide. But the problem with the paywalls is that they are not inclusive. Meaning that if you don't have money you don't get the information.
08:23
You're left out. And that it also means that the users and creators are not the same. You cannot be as a user a creator at the same time. So you could not say for example oh well I'm going to publish tomorrow at the New York Times or something like that.
08:41
Meaning that the platform decides or the newspaper decides who is a creator and who is a user. And the paywalls fragment the web. Meaning that they go against the very basic principle of the internet. The open principle, the open accessibility of information. And that fragments the web.
09:01
It puts content behind, literally behind walls where only people who have money can access it. On top of that it's not transparent. You pay a paywall, I mean you pay a monthly sum to a medium and you hope that they will provide you with good content.
09:21
But you don't pay for like single pieces of content. You pay a monthly sum to the medium. And paywall based models are usually even more centralized. Because the decision lies with the newspaper or the platform. The decision who publishes and what gets published.
09:41
This decision is in the hands of the platform or the newspaper. In the business model of paywalls. Now I've asked the question what would a world look like if we had more of this transformational content. More content that really has a super high quality makes you think.
10:00
Content that provides you with new insight. So in order to have that new world what would a business model look like to provide that? Well it would have to combine the good things of those two models that we have talked about now. And with as little of the downsides as possible. Meaning that it would have to have the overarching goal of high quality transformational content as I said.
10:27
That would have to be the big goal, the big incentive for everyone, every stakeholder in that system. And then everyone who contributes in that system towards that goal of high quality transformational content.
10:41
Everyone who contributes to that should be also financially rewarded. So that they have a stake in that goal. Plus it should be openly accessible meaning that even if you don't have money you should be able to take part and also get the information. It should be inclusive in that sense.
11:01
And it should also enable everyone who wants to also publish to be able to do that. So everybody should be able to also be a creator or a user as they choose. And then it should be decentralized and democratic meaning that there should not be a central instance that decides what is high quality content, what is published and not.
11:21
Those should be decisions that should be democratic and crowd-sourced. So that's actually what we are working on. We think we have found a solution that could be very interesting. We call it butterfly without the L, so butterfly. And we call it a crowd filtering system.
11:45
And you see on this slide how it would work. So at the top left you have the creators. The creators create content. And this content then goes through a filter with four different stages. And that's actually where the users are. So the people are split into four stages.
12:03
And they filter the noise away and they filter through that noise and actually the content becomes better and better and better. So each person decides for themselves, well, is that piece of content something I want to send to the next stage? Is it transformational? Does it have this butterfly effect on my life?
12:22
Or does it have the potential for that? If yes, I'm going to send it to the next stage. If not, it's not going to go to the next stage. So it's a democratic filter in that sense. Now, you see that stages three and four have a blue color. And that is because they are the ones who spend money. So they pay money for what?
12:43
They pay money for the privilege of getting only this high quality, transformational, substantial content. They pay for the privilege of not having to filter through the noise themselves. Other people do that for them and they pay for that work.
13:01
Important is that they only pay for the pieces of content which they think have that transformational power, which they think can really have this impact on other people. So they pay only for the content that they send to the next stage. And also very important, the money that they pay is split.
13:23
So the money is split and it goes first to all the people who have filtered the content for them. So that means if you are in the stage one or two, you consume everything for free. Of course, there's a little bit more noise in there,
13:41
but you consume everything for free. Plus, you can even make money because you get a little share if you find high quality transformational content that the next stages are going to pay for. And apart from the people who filter, the money is also split towards the creators, obviously,
14:01
because they created high quality content that went through that filter and also a small share goes to the platform for providing the infrastructure and the service. So every stakeholder, everyone in that system has an incentive of finding and producing high quality content,
14:23
transformational content that has really an impact in other people's lives. Very important, you cannot come on the platform and say, oh, okay, I want to be at stage four directly. I'm going to pay, no problem, but I want to be at stage four. Not possible.
14:40
Everyone has to start at stage one. So it's really democratic in a sense that you have to start at the stage one and you have to prove that you provide the system with value. You have to prove that you can do that and that you are willing to do that and only then can you go to the next stage. And that's a democratic decision,
15:00
not something that I or the platform decides. It's just a democratic decision-making process. And what else makes the platform democratic is that there's a separation of powers, meaning there's four different stages and none of those stages have more power than the other.
15:23
People are split in those stages, so no one can be at all the stages at the same time. It's different stages, different people. And that means that, for example, also the stages three and four who pay, they don't have more power than the others because they get only to choose from the things that have been provided to them by the stages one and two.
15:43
So it's really democratic in the sense of there is a separation of power. The users have to go through that transformation process themselves. That's also democratic. Nobody can come and say, I want to be at stage four directly. And it's decentralized, meaning that the crowd decides and the people decide,
16:02
and it's not something where a central instance decides what is good or bad, transformational or not. And then there's the equality of opportunity, which is also very important. Everyone can be a creator. Everyone can publish to the platform. And if they provide a story,
16:20
if they have something to share that is really of value for other people, then it doesn't matter where they are or who they are if they live in a so-called third world country because I truly believe that today the most important stories have not been heard yet because in the paywall-based model,
16:43
well, many people don't have the opportunity to publish at such a newspaper, for example, at such a publication. And in the advertisement-based model, the important stories are really being drowned out in the noise of everyone else. There's really no quality filter in there.
17:02
So I truly believe that this could be a system that leads to also those voices being heard. So it's a democratic system in every sense. That also means that, well, let's go to the first stage here. That's what we're working on right now.
17:21
So that means that you can go in this moment to butterfly.me. Again, butterfly without the L. butterfly.me is the website where you, if you open it, you see a screenshot here, you see the stage one, actually. So you get the question, what are the most life-changing things you've ever read on the web?
17:42
And then, like, please think hard and try to provide other people with value. If you have something that has had this kind of butterfly effect on your life, share that with the platform. And if people democratically also decide that they see value in that, you will be promoted to the next stage.
18:02
And very important, this is a project that is non-profit. As you have seen, the system is self-sustaining. It doesn't need any third-party investment or something. So that means it's a community project. It should be as democratically as possible. So that's why we're really looking for people,
18:23
maybe people like you, who can help us with that, who want to be a part of that, because the more people are a part of it, the more democratic it actually becomes. So with that, I think I will give back the word. And if you have any questions or also have any feedback,
18:42
please also feel free to reach us either through our website, butterfly.me, or at support at butterfly.me, which is our email address. Please feel free to reach out to us. And now I'm available for, yeah, if you have any questions.
19:02
Thank you for your interesting talk and this kind of invention you just presented to us. I'm pleased to hear. Maybe for all of you, again, the hints that you can ask questions directly now to Veron by using our chat channel.
19:24
I will try to ask them for you to. But let me start with my question first. So you're saying that there are many stages
19:43
and then my question would be, how would then the content pass on? Would it be a majority vote, or can one user decide if an article or a news byte, if you have broader sense of the video or whatever it is, then it can pass to the next stage?
20:03
Well, it's a decision obviously of everyone, but then we plan to introduce actually randomness to prevent bias. That means that it's not like if you have 100 people, it's not like the case that if 51% of those people
20:20
don't like the content, it doesn't go through. We add a little randomness, meaning that maybe the content is not shown to all 100 people directly, but maybe only to 40. And then if the majority of those 40 randomly chosen people think that the content should go to the next stage, then it's actually being sent to the next stage.
20:41
So we want to prevent with that, let's say that the majority always can rule, there's a certain element of randomness in it. But this would then need to pass two times the vote?
21:01
Yes. Because it's two stages, right? Well, no, actually every person takes a vote, but maybe out of those 40 people that have randomly been chosen to be shown that content, if the majority of those decide that it should go to the next stage,
21:23
then the content will go to the next stage. So it's actually just one vote that the people take at that stage. Now a question from the web we get. So first, the person is kind of criticizing a bit
21:44
that you're still kind of fragmenting the web in stages three and four. Well, the information is not fragmenting because at the stages one and two, where everything is freely accessible for everyone, you have the information accessible for everyone.
22:01
So it's not fragmenting in the sense that all pieces of content are there at the stages one and two, freely available for everyone. But then they get filtered in the sense that people decide what they find valuable or not. And then at the stages three and four, you only have a subset of all this available information,
22:21
and people actually pay for the privilege of only seeing that because the crowd has decided that this is the most transformational content and the people at stages three and four maybe don't want to get through all the noise, they don't want to see all the information, they only want to see what is important to them
22:40
because maybe they don't have a lot of time or they choose to spend money in that sense. But it's not a fragmentation because everything is freely available for everyone. Well, the work that you do is something that you get paid for.
23:03
So it's not a lot of work. After each piece of content, you just have to decide is it something that has the potential to be transformational or not. That's the only work you have to do after each piece of content. And that's really literally a matter of seconds. But if you choose things that really prove to be transformational
23:24
and of value for other people as well, you get financially rewarded. And the other downside that you have when you're on the free stages is that you have a little bit more noise. But that's the case today as well in the advertisement-based models. Even worse, I would say, because on this platform
23:44
there's pretty much only people who have that aligned interest in finding transformational content, finding things that are not polarizing, not this kind of clickbait content. So, yeah, I think you definitely have a lot of upsides even in the free stages where you don't pay anything.
24:07
All right. I'm not sure if I get the question 100%, but maybe you do. So the question is the information is already there now, but people's feelings are the problem. So you are drawing attention to the wrong posts.
24:23
Is it possible in our system that the most emotional information prevails? Oh, okay. So that's like the classic question. What comes first? Do the people want sensational content or is it that we're so used to bad content that we don't know anything else?
24:40
And I'm really deeply convinced that none of us actually enjoys seeing this kind of junk food and trashy content. Some content can be entertaining, and that's really no problem. In that system, nobody says what is good or bad. You can also reward entertaining content
25:02
or content that doesn't always have to be very rational or something, but it's about providing value. And I'm sure you've all experienced that, that you've read articles where you just thought, like, I just wasted my time. So that's what we would try to get rid of in that system.
25:22
And maybe I'm believing too much in the good side of people, but I really don't think that we actually enjoy the mess we're in right now. Everyone I talk to has a similar problem in the sense that they feel overwhelmed
25:41
and it's really this race to the bottom that we're in, especially with the advertising-based model. I think it's really time for something else and time for something that has at its core the interest of transforming people, but also then society at large.
26:02
So that would be my answer to that question. Now we have three questions from two people, but they are at least to me very strong links, so I formalize all three together. So the question is, how are you reaching out to people to initialize the system first?
26:21
And then, related to this, how can you make sure that the initial stage is not biased because you reach out, for example, only to people like the one from Neutron now listening to RSV3, or similar, so that you have gear roller sense. And while that's probably the same problem
26:43
as the next question, which is a bit more trivial, as stated, so how do you prevent the system of being controlled by bots? So if you have in one layer all biased people or even biased bots. All right, so the first question, well, obviously we have to start somewhere
27:01
and I really think that I'm in the right place to tell all of you that idea because I think it's important who is on the platform first because the one thing that is important in the beginning is that you are open-minded in the sense that you look at the quality of the content
27:23
and everything in the system is designed to enhance that. So, for example, the user transformation, or maybe let's come to that later, but there are many features that should prevent this type of bias, and the most important one is
27:42
certainly the user transformation, meaning that everyone has to start at those first stages. And then the more people come on the platform, the less the bias of each individual matters, right? So the more people come on the platform, the less it will become important
28:00
who is on the platform right now. But in the sense that how we reach out, we start reaching out to people like you in the beginning, so you heard now about this project and we really invite you to be part of this project right now and we're not advertising this in other circles at the moment, but the more people come on it, the more democratic it gets
28:23
and the less it is important who is on the platform right now. And then the question with the bots, well, the system has this kind of score that the system looks at. For example, if you're a creator, the content that you provide,
28:41
or if you are a filterer, the content that you filter, it looks at how far it goes, like how much of the content that you send to the next stage is then sent to the next stage. And if you're a bot and the system detects that, for example, this score is really low,
29:00
like all the people in the next stage determine your content to be really, really bad and don't let it to the next stage, then the system would warn you and maybe block you for a certain time so that you cannot introduce new content into the platform in order to prevent spamming. And then also there is a report function
29:25
where if you feel like there is content which is spammy or you feel like it's something that a bot has brought to the platform, you can also report that, but it's not reported to a central instance. It chooses randomly a certain amount of people
29:43
who are shown the content and then it's a democratic decision of this random set of people if the content stays on the platform or not. So it's kind of a democratic decentralized report function as well. So that's one aspect. And then also we have the aspect of time. I haven't talked about that right now because it's actually important,
30:03
but it's like a technical detail that we have a time lag in the system. So each decision you take is recorded on the system but not immediately done in the system, implemented in the system, meaning that Butterfly will ask you after a few days,
30:23
will show you again the content and ask you, hey, do you remember you have wanted to send that content to the next stage? Do you even still remember that article or video or whatever it was? If not, it's probably not been transformational. If yes, do you really still want to send it to the next stage? So this time lag also prevents bots from immediately doing something
30:44
or it enables or it increases the chances that people, that we all really take a step back and ask ourselves, is it really something that has had now over those days, over this time, an impact on my life? Or is it just something that has made me in that moment feel good
31:04
or I wanted to reward it in that moment, but today I don't think it's the case anymore? And that time lag would also increase, in my opinion, highly increase the chances that the content that goes to the next stage is really, people have thought about it
31:20
and really carefully chosen to send it to the next stage. So speaking of time, there is another question related to it, which sees this probably a bit of a weakness. So the question is, sometimes speed is crucial. Let us consider that security vulnerability publication, of course.
31:41
So passing all four stages takes time and you might miss the news if you only subscribe to stage four. How do you deal with that? Yes, that's definitely, I mean, that's a disadvantage of every democracy, right? That's why decision making takes time in every democracy.
32:01
And it's something that you, I don't think you can prevent it with that system. It just takes time and I think it's actually a feature that it takes time, because it's more, I think, probably it would not be the right platform to publish, or at least not in that form that I presented now. It wouldn't be the right platform to publish a security vulnerability straight away,
32:25
when maybe you could alert the public more easily through today's social media or something like that. But I also believe that if the security vulnerability that you would share with the world, that people would actually see that,
32:41
hey, there's a value in that, we have to make that heard and then it would also go through the stages. It still would take a little bit of time, but I'm really convinced that it would go through the stages. And of course, sometimes time is critical and then maybe that would not be the best use case.
33:00
But I think that this aspect that it takes time is actually really a feature, because it makes you take a step back and actually really reflect on your choice. All right, we're already over time, but there is one more question which is stated twice, one's on Twitter and one's in the channel here.
33:29
So a really cool idea, yeah. Then I do not understand why do you want everybody to start on stage one.
33:41
So that way you are cutting your target group and get less user or customer. And the person on Twitter has similar concern. It says, so if people who are willing to pay basically have a lot of money, they might be well with their time higher
34:06
and are not willing to pay more time to go through stage one, two, but they would be willing to pay for already moderate content in three and four. All right, so I think it's actually very important
34:20
that everyone has to go through the transformation model, because it aligns your interest. Only the people who take that time and go through all those stages prove that they are interested in that high overarching goal that the system has. Only if you prove that you are interested in that, you're allowed to proceed in the next stages.
34:42
That's really essential to that democratic function of the system. And I compare it a bit to Wikipedia, where also you cannot just immediately edit something and it will be edited. You have to take the time. It has to go through different stages. And I think it's actually very important,
35:01
because it also doesn't tie money to status or something. Money does not influence that decision-making process. And I think it just aligns the interests of everyone and it assures the platform or the system to,
35:23
it helps it to know that the people who go through the stages are really invested in the idea. And I think even if you have money now and would like to use such a system, but don't want to go through the stages, I think you would have to ask yourselves, why not?
35:40
Just do it, provide value, and you will get to the next stages. And it allows the system to stay democratic and not to give people who have money an advantage over everyone else. So a link to this, maybe you can answer it yes or no. It's a question, is it also possible to choose to stay on stage one or two?
36:01
Yes, definitely. That's something I haven't mentioned, but definitely it's always your own choice if you want to spend money or not. So you can definitely also stay at the stages one and two. You will be alerted that, hey, you could go to stage three if you want to, but if you don't want to, obviously you don't have to.
36:21
All right. Thank you very much for this talk and also all the answers to today's question. So we take a little break here and we are back in about 25 minutes with a talk about net politics in Switzerland.