 Hi, hello everyone. Thanks for joining us for today's session. This is our first keynote session for the Unraveling the Cycling City course. And today we are happy to have George Leo with us to give us keynote talk. I will be assisted by Facundo along with me here today. I just wanted to give you a brief agenda for the today's session. First, I will be introducing George that will be followed by the keynote talk by George himself, which would be about 30 minutes. And after that, there will be a Q&A session for about 25 minutes. Of course, George doesn't need much introduction, but it's a little something that I wanted to let you know about George. George is a PhD researcher studying on how ideas from urban design can guide the creation of attractive environments that encourages cycling as a practical and desirable mode of transport. He also studies the emerging concepts of cycle highways, which is conceptualized by practitioners, academics and people cycling in the northern European countries. George is currently cross appointed at Einhofer University of Technology and University of Amsterdam as a part of the smart cycling futures project. I just wanted to give a little bit of a note for the Q&A session. During the George's talk, you can put your questions on the chat screen, which would be compiled by Facundo and he will be moderating and queuing the questions which we will ask on your behalf after the Q&A session is over. So, George, how are you? Hey, I'm doing excellent, yourself. Very good. And I hope everyone else is doing excellent too. We're all trying to do the best we can. Okay, thanks for taking the time to join us today. And without much ado, please do start and give us your keynote talk on beyond stated preferences. Excellent. First, I wanted to thank the MOOC team for, you know, you're all volunteers. So it's great to see the initiative to have these things happen and to have these talks being organized. I want to keep the MOOC fresh, but the course content itself isn't always easily updated without interrupting the progress of current students. So this is these live sessions are a great way to add on and to keep things current. I was going to talk about bicycle highways today, but then I realized there's a whole lecture in there about bicycle highways. So I wanted to talk about my research on cycling experience instead. And the two are not unrelated. They're actually quite closely related in terms of, I try to use bicycle highways as a way to apply the research on cycling experience. By the end of this talk, you can ask me questions either on the paper presented in the course or the presentation that I'm putting forward today. So today I'll be presenting some findings based on a paper that I wrote before the bicycle highway paper. I'm putting out a link to that paper in the chat box below. And it'll be about cycling experience. And right now I'm taking the idea of cycling experience and trying to figure out how it relates to the current body of research on stated preferences and revealed preferences, which is kind of like quite an academic way to talk about what people say versus what people do. So stated preferences generally is a way that we try to find out what people want or what people desire in a laboratory setting. So to do research using stated preferences, an example would be you're taking a few different infrastructure types, you know, protected infrastructure or or sharers or bike lanes or no infrastructure. So, and using that type of categories to ask people hey, it given this road type what would you prefer so it's very hypothetical. On the other hand, stated preferences, sorry reveal preferences is generally done using GPS study. So now, most of those reveal preference study is done by GPS, which means you track people on what they do what kind of roots they choose. For example, if people choose the shortest route, then you can ascertain from that data that people like the shortest route or people choose a route with separated bicycle infrastructure at the expense of perhaps longer detours that we can make the conjecture people would take prefer protected infrastructure or a separate bike path, even though they have to give up times for it. So in the era of big data and easy access to GPS right because all of us has GPS in our phones. It's the research had seems to be heading more that way. Now what my talk is about is how do we take these two categories and how do we move forward and learn more about people's motivation and meaning right because having GPS data from your phone tells us nothing or next to nothing about what our motivations are or why we choose what we choose. And, you know, unraveling the cycling city kind of gets into that, especially, you know, these articles on rule breaking for those of you week four. And I'd like to look more into the why you know why do people choose what they choose. So let's talk about experiments because they're, they're quite important, you know, as a, as a scientist, right what is an experiment and better yet what's the difference between an observation and an intervention. So, first of all, the experiment involves number one, specific instruments and specific people. Right, so usually when we think of an experiment, we think of someone in the lab coat, right administering doses of medicine, perhaps in a very controlled environment, which is number two, aims for a controlled inducement of changes in the laboratory. And that number three measurement of these changes before jerry and after. And this is our traditional concept of what an experiment is. But the problem with bicycle research is that the world isn't very neat like that, and it's kind of hard to put cycling into a laboratory where we can control everything. But we mostly rely on observations for what we do. So we, we try and observe the way that people choose their roots, or observe the type of environments that people cycle through, and we try and make some conclusions. Another way is to ask people what they prefer. Right. And this is also not an experiment in the true sense because we, we don't have a before and after we're just trying to obtain data. So it's it's also an observation like type of way of getting data. Now where does meaning come in and where does the why matter. The why matters is because they're right now there's revealed preference and stated preference right. The why matters because neither of these categories kind of tell us exactly what the reasons for these preferences. Right. And I think even the word preference is is a bit constraining, because we want to know not just about preferences, but about the reasons behind people's choices. And the reasons may be motivation, right. Why, why, why does cycling matter. And that's also some of the things that we covered in the course. So the, the difference between stated and revealed preferences also seem to matter less. It seems that GPS has really supplanted the this research most revealed preference research can be done using by just tracking people to see where they go. So there's not many things are very interesting there, aside from more and more data. And on this stated preferences, if you're just asking people know what type of roots do you like. It seems a bit shadow, shallow, given our huge access to GPS data which tells us exactly what kind of roots people prefer. So what's the alternative to this type of research. I think the alternative is to look into research that looks more at at motivations specifically. Right. Some of the research I do is is right along research, which is doing interviews on the go and trying to ascertain notes, a meaning that's more than just transport, because if we think about cycling as just traffic determinism, right, as just traffic determinism, then we don't see a reason to obtain people's motivations for travel. But if we see transport as cycling as a meaningful way of being right so you know you can observe people standing outside of a cafe. You don't ask, you know, is that an efficient thing to be doing. They're not moving anywhere. If you think of many of our stationary activities we assume that there's meaning behind it. Right. If we're at the office we're working. It's a very clear why for at home we're doing home stuff. But for some reason when we're on the move. We immediately switch to this scientific mindset like where's this person trying to go. And that's the, that's not very helpful for getting motivations, and I think there's, there's a lot more that can be explored here. So, with that in mind, I, my research is on cycling experience and that's kind of the why for the aspect of the why that want to touch in today. And so I'm going to go briefly to a 2018 paper I wrote to this diagram, which, which outlines three different ways. Let me just go find that it was up here briefly. Here it is. So, which outlines three different types of cycling experiences of social sensory and spatial. Right and this kind of gives you an inside out view of what, what matters for people that that ride a bike. On the social side, you know, we tend to forget that interaction is important that, you know, the sense of normality that the social context is important for people who bike. For example, if you're in a context where no one cycles and it's kind of weird, then obviously that's going to affect, you know, the way that that you travel as well. You know, the way that we feel on a bike, it's very sensory way of movement. It's very different from automobile traffic. If you're in a box, right. Shielded from the environment, which insert interestingly, if, if you're in a car, it's actually much easier to do these laboratory experiments, because you simulate know the windshield you can simulate the visual effects of being in a car. It's a lot like a video game, but it's very much more difficult to simulate the feeling of being on a bike. Right. It's not a chair you kind of have to balance. You feel the pavement. So it's, it's very difficult to do laboratory experiments for cycling versus a car where you can simulate know what's coming up on the windshield. You can get people to play driving games, for example. And you can do experiments that way. So that's, that's kind of the reason also why it's so little is understood about cycling. And then spatial, right. How do people create a sense of the environment. How do people make sense or develop preferences for for certain environments around them, because cycling is more about than just environments. More than just infrastructure. What kind, why do prefer people prefer greenery, why do people prefer certain types of architecture. And maybe these are the kind of questions that we'll never know in the particular. But we do know in general, you know, what kind of environments people prefer riding a bike, and they seem to be approximately the same types of environments as people prefer walking. And that's why it's important, I think, to look at urban design is because we have a history of knowledge. Though not entirely scientific, we do have an history of knowledge from architects and as people who have made cities, you know, according to certain design principles that have been developed through the year. And it's, it'll be interesting to make that body of knowledge, much more scientific, do experiments on it, do more observation to obtain why and how what type of environments people prefer exactly. So let's, I want to just go through a few quotes here that some of the you in the course might find interesting this paper wasn't in the course. You know, here. So, So what were the things that you notice on a bike. Right. Not McCarthy 2011 finds that not only do cyclists list of hosts of risks attributed to this driver behavior and attitudes, but they have also formed through the process of sense making a common framework that explains the origins of risks posed by drivers. And that's a piece of research on cycling, on cycling experience that was done using interview and direct transcriptions, right, which is very different from the type of research that we typically do, which is based on data and surveys. This, this is an example of the sense making that we need to do as researchers that cycling is not just about going from a to b but it's also about kind of figure out what kind of questions are valuable to be asking. Especially when it comes to risk, especially when it comes to danger. You know, these fields of research have much more in common with researchers who study pedestrian environments than they do with people who are in the safety of an automobile. Let's go to an example of a sensory experience. Right. So, let's go to this one about shoes, which I think, again illustrates the, the similarities between, you know, cycling and walking shoes, I quote can be understood as part of a hybrid unit of analysis. Right. So, human socks shoes pavement, we rarely do we break it down that that far, where shoes intervene and disrupt the flow between the body and the pavement. You know, and, and that we can think of that in a similar way to cycling where the type of equipment you use disrupts or intervenes with the way that you actually experience the city. Let's say that's, that's also something that we could look forward to researching more is, you know, in some contexts, the traditional Dutch bicycle, right, is is is almost universal in other contexts such as North America it's much more mountain bikes. So, what does a machine have to do with human. How does that interaction happen, and what kind of implications are there for for cycling. Now, perhaps to close off this talk, I want to also talk a bit about e bikes. So, e bikes are kind of taken over the market, they're more than in the Netherlands it's more than 50% of all sales I believe so it's that they're more expensive so maybe not the number of bikes but definitely the sales the value of bikes, more than 50% of new bikes being sold and I suspect it's, it's been way past that mark in places like China already where these much cheaper e bikes are are on the market for quite some time. And I think e bikes are an interesting case where we focus on a lot of the transportation aspects of it, right, goes faster enables people to pet pedal further. But there's, there's something underlying that, and that is, I think e bikes change the way that cycling is experienced right it's becomes much more of a motorized vehicle and. The extent to which e bikes norm bikes share characteristics and defer, and that would be a very interesting avenue of research, and also know if, if you're able to throttle it completely or if you need to pedal assist. So there are a lot of questions about how people get around but there are also questions about the meaning and the experience of cycling that changes as a result of putting a motor on a bicycle. So let's go back to the original question. You know, which we started with was these research with large sample sizes may reliably capture preferences, but what about motivation. And I think that's a place that we can improve on understand more of the why, and only by understanding more of the why, can we then in the future, you know, understand exactly what is what research needs to be conducted. And also the blind spots that we have on cycling research. So that's my, that's my two cents in this talk, and that is also why I think, you know, we need to go beyond stated preferences, and why we should be exploring motivation and meaning more as we look at how cyclists interact with the world around them. So, hey, thanks for tuning in. Look forward to the questions. Thanks, George, because I need a very informative talk. We're waiting for the questions to stream in. And I think Facundo will help me out with this. Facundo. Yes. For the moment, we have two questions. So the first one would be, George, how does the sort of spatial data from apps like C dot sends right insights fit in. Okay, so I'm just looking at the, the webpage here briefly. So, I think that's, let's, let's think about the underlying problem here. And that is that the perhaps the responsibility of safety and and who bears it. And whether, whether, or and whether, you know, we should be requiring people to have technology in order to be safe. I think that's, that's a question here. So, but this question is also about data, like apps. So, I'm trying to get to the data page. Let's look at this together actually. Let me put this up. Perhaps we could unmute Kim and Kim could further inform us about what data she's talking about. Here, here's a webpage. I don't know, Kim, if you have a particular information about the app. I was going to ask about, it gives a look, there's more to it than just what's there, you can collect data on how people are reacting to the road. So how they turn sharply or if they experience a lot of vibration and things like that. It can always be brought into people's experience and you can see the hotspots and maybe survey the hotspots or something like that. It's, it's all rather new. So I don't know how you'd use that. Okay, so that, good. I, I note here that there's, there's a qualitative service that delivered via the app, which I find that quite interesting. You, you're, we're now able to, you know, time quite precisely what the quantitative measurements for for cycling are, and put that together with the qualitative data. So, for example, if you experience a near miss, and you're able to give a reaction to that immediately, that would be combining an outside in perspective which is probably accelerometers on the device and GPS data, and also matching with that some reaction from the cyclist. So, I, without understanding this more, I think I can talk to that concept. And that is quite useful. It's, it's, because what has been difficult is trying to measure people's reactions while on the move. Usually, it's been take people on a bike ride, and then you kind of perform a survey afterwards about, you know, when you experienced a near miss, how did you feel. But that's, by the time that that happens by the time the researcher gets to the person they're trying to research, the instant may have happened, you know, a few days ago, a week ago, a year ago. And the memory clouds, what the sense was in the moment. So by being able to sync up exactly what the instant was, in terms of the accelerometer so however the near miss or the incident is calculated with a feedback by the person who was involved in that instance, and be able to have those two data sets together. I think that's, that could be really useful to understanding how people interpret, you know, these events in the environment. So thanks for that question. I look forward to following up on this. Is this, is this a, yeah, I wonder if this is this would be good for for data collection for the research community. I think on the consumer side it's much more targeted at these lights and things. So the question, I'll look into this product a bit more and see what I see what's there. Thanks. Okay, so we have another question from Anthony, and they're asking, do you have knowledge about preferences depending on the type of bike, for example, an e-bike or a cargo bike, and the type of use, for example going to work going to the grocery traveling with your kids, etc. I just published a paper about cargo bike preferences. A few, a few months ago. See if I can look that up. And, and the results, it's, it's, it's hard to actually categorize these different types of bicycles. Even that's one comment I'll make while doing research. Even cargo bikes have their subcategorizations, right? You have three-wheeled cargo bikes, which actually feel quite a bit different than two-wheeled cargo bikes. And then you have these cargo bikes with the load on the back, you have cargo bikes with load in the front. Electric cargo bikes feel very different from non-electric cargo bikes because of the weight, especially if you're in a hilly environment. So just trying to narrow down the selection and the choices to make of what type of bikes we're studying and what types of bikes we're not studying was, was difficult. And as to the motivation, you know, so why people ride, where they're going as a leisure ride, as a day utilitarian ride. And that was, it was also difficult to determine in conjunction with the, the quantity of data we're collecting. So, you know, it's, it's, it's hard to determine on the one hand, if you're doing a traffic count or GPS study, exactly what type of trip it is. Right? Like, there's some figuring out and puzzle pieces to be done, figuring out, okay, if this person is going from here to here, you know, they're clearly going from home to work. But it's, it's not, it's not cut and dry in terms of being able to figure that out for, you know, thousands of people. And if you're doing a traffic count, it's even more difficult. Well, well, you can get the type of bicycle from a traffic count. Right. So you're standing there and trying to figure out who's riding what, and you can figure out the gender. It's, it's much more difficult to then get the purpose because you're not going to stop everyone to ask them. Whereas with GPS studies, generally, if you're doing research, you're getting a big data set from someone else because they're expensive to conduct now with cell phones, but privacy issues are important. So you can't then track down the people who are in the data set to ask them, you know, exactly what type of bikes you were riding. And to confirm the data, are you going to home or work. So there's quite a bit of methodological issues with connecting all these different elements that we want to know about. And then perhaps more difficult and the same question is what kind of socioeconomic group are they in, etc, etc, and having a survey that determines that with the GPS data and trying to get to know what type of bike exactly that they're riding. And maybe one day we'll be able to get all that information together and make a useful determination. But for now, I think that having having that's a challenge and also determining what type of bike to include or exclude. What is a cargo bike, for example, is also a challenge and perhaps something that we won't really ever figure out, given the wide variety of bikes out there. So the next question is from, I believe it's pronounced Rohit, I apologize in advance if the names are different. They say Arafa Long reads articles stated that using GPS for navigation might potentially make our memory weaker. In comparison cyclists should have a better memory and visual cue and hints on local markets. So this is a question in comparison do cyclists have a better memory and visual cues and also will that then help local economies. Wow. Intuitively, I have no doubt. That's the case. In terms of research, I think that can be backed up by quite a few studies for drivers. The whole talk about how cyclists and drivers are different, but I suspect, you know, if people driving cars are interpreting, have better memories of their roots, which is a field that research has been done on, then perhaps it's even applied more so to people cycling who have a better sense of their environment. So I suspect that's true the link to economics. And, sorry, can you repeat the last question I can't find out the chat here. The just the last part. Yes, they were asking if this difference would help the local economies. I don't know how to answer that one. That's, that's, that's a bit of a stretch to go from memories to local economies. But I think there might be an argument to be made that cyclists, well, we do know that bike lanes promote businesses from a few recent studies. So maybe part of that effect is a more intimate interaction between people on bike and their local environment. So, especially if it's main street businesses, which was a particular study, then if you're driving at the speed of a car, you know, those types of environments aren't designed for being passed at 60 kilometers an hour. But if you're on a bike, then the, the, the number of buildings is you're going slow enough that you can actually, you know, find what you're looking for. And it's possible, et cetera. Whereas, if you're driving really quickly, you're looking more for like a highway rest stop where there was giant signs that lured you in, you know why you're going really fast so I think the speed of mobility, not so much the memory of the environment would determine whether you're able to locate what you're finding in that environment. And also being to pull over and park your bike. Maybe that's a factor, but this is no all personal conjecture on that point. Then we have Jain, who says, on choosing routes, avoiding traffic lights is a factor. Has there been good studies to quantify how much energy or time is lost in stopped start journeys? How much energy is lost? I think this is interesting because it's more of a mechanical question, right? This more speaks to the cycling as a human machine, something that we can figure out using physics, really, about power loss and how much energy it takes to get started and stopped versus the wind resistance we're going at full speed. But I don't think that's the really relevant factor here because we do know that it takes quite a bit of energy to get something that stopped into motion. I think this question though, what's more relevant is how people interpret this energy being put in, right? How do people experience it? And whether they're adverse to it or whether they appreciate this input of energy. Perhaps there's a distinction there to be made between people who are out for a recreation ride and people who are out for a ride to work. Maybe more importantly, maybe in the future, this distinction will become less relevant as more people ride e-bikes. Because what I do know with e-bike riders is that when I was doing my personal research that they mentioned quite frequently the eradication of wind. So it takes the unpredictability out of the route that if you have a lot of headwind, for example, it affects how long it takes to get to your destination. So when speaking to some Dutch people, as they were growing up, getting to school the largest variable was not the traffic light, but actually the amount of headwind they get. If you're going 10 kilometers on the countryside, flat environment, and one of them mentioned that with an e-bike you would completely eradicate that variable, making your journeys more predictable and therefore making cycling more of a better transportation alternative and you're not exhausted by the time you arrive at work. So perhaps the question to the traffic light, the answer to the traffic light question is to what extent does it make you exhausted and tired by the time you get to work? Maybe that's the biggest variable. And maybe in the future, this variable will become less important. If more people are on e-bikes and the waiting time at traffic lights, maybe that becomes a much more important variable. I'm also gonna put, yeah, sorry, I'm gonna post the cargo bike paper in the chat here. Maybe we'll add that to the course at some point, but that's for you to read. Oh, sorry, to a public group. Yeah, go ahead, next question. Yeah, the next question's from Sean and the question is, do you have any suggestions on how cities with lower bike commute rates can normalize bicycling and also reinforce that the biking is generally safe? And also, what about cities that face potentially dangerous weather? For example, here that's over 115 degrees Fahrenheit for a specific period of time, but are otherwise photographically ideal for biking? Hmm, no. The weather one's tough. The weather one, maybe there are just places that are too extreme for cycling. For example, if we take an example from the streetscape from pedestrians, there's cities being built on deserts and in these places, these climates, most of the architecture is built in a way that everyone is in an air condition environment, for example. So either above ground walkways or underground walkways, and apart from that, mostly just automotive travel. So I'm thinking of these places like Dubai, Houston, et cetera. So that was, I think that's the architectural solution on the pedestrian side. Is there a similar way to protect people from the elements on the cycling side? There doesn't seem to be a good way to do that, because if you get too much enclosure, it's a lot more weight at what point is it no longer a bicycle? And then if you're trying to get people inside, then spatial efficiency becomes much more of an issue. So then you get into walkways, escalators, et cetera, in which case, does everyone really need a bike in that environment? So that's a tough one. Maybe there are environments where cycling won't be as popular due to the, you know, the temperature variations or weather. And interesting about hills is I go back to the e-bike, is that maybe that's a solution that can be solved by technology, right? Hilly environments may become much less of an issue compared to places that are too hot or too cold. And that perhaps in the long term will be a much more difficult issue. Sorry, can you say again the first part of the question? Yes. They asked, do you have any suggestions on how cities with lower bike communities can normalize bicycling and also reinforce that biking is generally safe? Yeah, that's a tough one, I think. How do you make a normal, it's a vicious cycle. I don't think we really have figured it out yet. You know, it's more people makes the activity more normal. There are perhaps one area of very, lots of contention. There might be even a lot of contention in this room is helmet lost, right? Not wearing a helmet does make the activity more convenient. So maybe there's something to be said. If you look at Denmark, you know, Copenhagen where there's more people wearing helmets compared to the Netherlands, where almost no one wears a helmet. Maybe there's a study that can be done to kind of figure out the impact of helmet use on the normality of cycling. But then again, there's also cultures where the helmet is the normal thing to do, still with some degree of cycling. So maybe that's a key issue. I don't have a good answer. I don't think really anyone has a good answer for that. It's a tough question that we'll keep working at. You talked a bit ago about e-bikes and the next question is about e-bikes. So Cyril asks, what are your thoughts on where e-bikes belong within urban mobility spaces? How governance should impose or suggest policies? And what impact may be in terms of equity? For example, decreasing the size of some groups to cycle where e-bikes allow others. Perhaps more privileged groups to cycle for again, some different purposes, leisure versus utility. E-bikes are interesting. There's a few issues there. There's the equity issue. Like what does it mean for everyone to spend $1,000 to be included in this activity? Maybe the economic issue will be solved in the future. I think what's more interesting for me in that question is the safety issue. I think I was reading somewhere while I was doing automotive safety research that the safest car is a car with a spike point directly at the driver. The safest car is the car that is most dangerous for the occupant. So if we think about that logic, that perhaps the safest e-bike is the e-bike that's most dangerous for the rider. So what does that mean? That means that as e-bikes become heavier, there certainly looks like there's e-bikes on that very heavy side of the spectrum, e-bikes that look more like mopeds, they become safer for the rider. The more mass there is in the vehicle, the more force that vehicle is gonna take on impact versus a vehicle or a person that's much wider. All right, that makes sense, so basic physics. So the heavier we make e-bikes, the more dangerous, I believe, people are going to be inclined to act. I think it's not common sense, but that's just the way that risk would be distributed. If you're completely rational and you're cared for at your own safety, the heavier the bike, the more aggressively perhaps you would be riding it. So perhaps one solution that's not overly onerous or complicated on the legal side is to make sure that e-bikes are below a certain weight, that they are not excessively heavy. In terms of the top speed, there's a lot of talks about make sure people don't go too fast. Perhaps it's informative to look at racing cyclists. People can go really fast under their own power. If you have a professional cyclist out for a sport ride, they're going 40, 50 kilometers an hour on a flat ground. So we know that people going this fast are using the infrastructure. So it's not so easy to say if we limit the top speed of e-bikes, then we limit all fast cyclists. So I'd say that the weight more than the top speed would be more important. And obviously there needs to be some limit to the top speed, but going too fast isn't necessarily the danger. The danger is people making the choice to go too fast. But making it artificially feel too safe for people to be going too fast on especially like pedestrian environments with dense urban environments where that kind of speed is completely uncalled for. And I think at some point we have to rely on people's common sense. If an environment is clearly designed with a lot of cross traffic and a lot of people sharing the space, that's where social norms kick in more than we can regulate top speed. And you know what top speed sometimes that's one of the advantages of e-bikes, right? If you're out in the countryside and there's wide open path, maybe going relatively faster than you would in an urban space is a completely rational thing to do. So that's my two cents. Weight limits over speed limits, if I had a choice. Okay. Sarah from the UK says that cycling infrastructure is really patchy and means you're cycling with cars, et cetera, which feels unsafe and a bit hostile. Whilst GPS data shows where people cycle at the moment, it doesn't show where this unmet need. For example, protected cycle routes. Is there a potential to capture or survey the unmet potential? Yeah, this is where the models come in, right? When we try and predict future bicycle traffic and actually the reason behind stated and revealed preferences, kind of like why they exist, in a sense is to calibrate these models, right? We take these revealed preferences so we track people via GPS and we have a pretty good sense of what kind of routes after they're built, what kind of routes they prefer. And if we're lucky, we have some natural experiments where a new route does get built and we see how many people shift over to the new route, right? And using that shift, we can determine the propensity of people to use that particular route. And then we can take a look at that new route to see if it's high quality, low quality, et cetera. And we kind of make a model out of it using all that data. And the input for that model could also be stated preferences. So we also have research on which people prefer separated bike paths, when is it appropriate to merge traffic, take all that data into a model and traffic engineers is using these models, urban planners to try and predict where a new segment of bicycle route could belong. Right? That's kind of the conventional way of thinking about it. But the problem is the models that we have currently and especially when I was doing research in bicycle highways, it turns out that they're completely useless. So these models are being used to justify funding for one or a rationale for building new bike infrastructure, but they're really not any, they're no good at the moment. And the numbers that they predict are quite far away from the numbers that actually are realized. Right? So that means there's, despite all the GPS data that we have, despite all the calibrations and the input that we're modeling these bike paths with, it's saying that there's something that we can't account for quantitatively. There's something in there that's, that even though it works for most car traffic models, this typical gravity model or agent based modeling, even though these techniques work for automotive traffic, it seems that there's something about cycling in which these models don't work. And that's kind of the residual leftover from the correct predictions versus, you know, how much is unaccounted for. And maybe, and if, if my hunch is correct, and there's so many micro elements and context dependent elements to cycling and things are unmeasurable, such as, you know, the type of building environment that the vault, the micro volumes of traffic day by day, maybe these models will never be 100% correct. And they might continue to be quite a far off just because it's difficult to predict how people respond to their immediate micro environment versus how they respond on a network traffic scale. Okay, so next we have a question from Chris and it says, how do you see this information being used in a practical sense? It's fine for most of us consumers, for example, bike riders to consider this, but how can we use this information to get our city engineers and administrators to provide what we need to feel safer and better about riding? So that on the one hand, and then they go on and ask, more relevant, perhaps, how can this be used to make cycling more accepted generally as a transport mode? So, let's talk about the city administration. It asks about how do we get this accepted. The local system and the system that runs transportation and traffic, right? The people that we must influence to change the environment. And the reason I talked extensively, just before about modeling, is that the people that control roads and public space in general around the world, it's actually quite surprising that that's the case because almost everywhere are increasingly traffic engineers. And there's a certain or an urban planners who are just finding their decisions using quantitative modes. So actually the research that I'm doing into these experiences don't translate well into models, right? And the challenge there is to figure out whether there is an independent or competing logic that our administrators would accept to justify a decision, right? Just because the models are wrong doesn't mean that the particular infrastructure is not worth being built. There are people who, practitioners who build bicycle infrastructures who know that the model is wrong. And given the model is wrong, they still have to make these decisions, right? And increasingly I think these decisions should be, maybe the engineering way of looking at these decisions can be replaced by a completely different paradigm, right? That if we don't think of traffic flow and carrying people as the primary objective of building bicycle infrastructure, it's a neighborhood beautification project. Some people have argued that bicycle infrastructure is a traffic calming project. It could be parks, it could be turned into a linear park, for example, enhancement of public space. These are all projects and logics are not related to carrying traffic. So perhaps one way to bring this research into some kind of practical use is to say that we're thinking about the problem the wrong way, that the primary purpose of bicycle infrastructure is indeed not as infrastructure at all. The fact that people use this environment or the bike path is a side effect, but the primary purpose perhaps of some bike paths is to create a better and more appealing public space. And that's kind of as far as I've gotten on the other side of the logic, because the alternative would be try to incorporate this into traffic modeling. And so far, efforts have been really unsuccessful. And perhaps, as I said, the reason it's unsuccessful is because it's not capable of being incorporated into a traffic model. Okay, so we're at the one hour mark. Do you have time for one more question? Yeah, sure. Let's do one more. Okay. So the question is, has there been any work done on accidents on e-bikes compared to normal bikes? Okay. Yes, plenty actually. Plenty of research have been done on e-bikes, accident rates versus normal bikes. I don't know the studies personally, but I do know that that's a huge field of study. And perhaps the related question there is, are e-bikes more dangerous? And maybe that's something that research, the numbers can tell us. Perhaps what's interesting for me on top of the, are e-bikes more dangerous question is, is how they enable, how the behavior of people in e-bikes can be interpreted by others. So how do we solve that problem of danger rather than just heavy regulation? For example, one of the ways that I think from personal experience, one of the ways, the visual cues of a cyclist stopping or going really fast is how fast their feet are moving. And if you're looking at a cyclist, you can tell that they're riding an e-bike because their legs aren't quite moving at the same rate as a speed that they're going. Their body motions don't quite make sense, given the way, how fast they're traveling. And perhaps one of the areas of research that we can do better is to determine what kind of visual cues both drivers and cyclists use to kind of estimate speed and then improving the design of e-bikes to make sure that our visual cues of estimating speed match up with what people expect. And further on that, maybe as more e-bikes are out on the road, perhaps we get better at interacting with e-bikes and maybe they'll lead to a safety gain. Or maybe not. Maybe e-bikes like motorcycles are just inherently dangerous. We do know that motorcycles are about 100 times more dangerous than driving, right? And so far, no one has been able to do anything about it. It's just the inherent design of the vehicle. Maybe it's true that e-bikes are just more inherently and inherently more dangerous, in which case it becomes a very tough question for policymakers. And people who are thinking about inclusivity and these cities have hills, what kind of trade-off are we willing to make, right? Between e-bikes and accessibility, if there is one. Okay, thank you very much. That is it from me. I'll hand the mic back to Sune. Hey, thank you very much for joining us today. And George, thank you very much. That was indeed a very informative session. I can get a last word in. I've given some perhaps controversial answers here. So it's good. I think we should be on the edge of controversy for anything to be, you know, to be useful, I guess. I'm trying to take us to the edge of that controversy. So, you know, let me know your thoughts for you. If I said anything factually incorrect, or if you have any perspectives on things. Yes, sorry, go ahead. No, no, sure. I actually wanted to ask you that if people want to follow up and ask you questions, what's the best way they can get in touch with you? Twitter or should they send us your questions and we can forward them to you? I think the best way to get in touch would be I've been trying to stay off social media. But if you, if you message me on LinkedIn, I'll leave the LinkedIn at Urban Cycling Institute or LinkedIn, if you just find me at George Liu. That's the place I'm at most. And if you want to get in touch with Mark, the promised route is also in the course, he's much more on the Twitter side as well. Thanks, George. Again, thanks everyone for joining in. And I hope this was an informative session for you. And of course, we look forward to having you for the next session, two weeks from now with Trey. And please let us know your feedback on how you found the session and how we could make them more informative for you. Before we sign off, I just want to also introduce the rest of the team who's been working behind the scenes to make these sessions possible. So two new mentors have joined us for this cohort. Yasmeen and Anna. And besides that, it's me, Sunay, Artem and Facundo, who have really been putting a lot of their personal time to make these sessions possible. And we will look forward to organizing more sessions for you guys. Thank you very much.