 Hello, everyone. Good morning. So glad you could join us today. I know it's early. I know it's Saturday. But thank you so much for being here. Let's kickstart World IE Day 2021 New Delhi edition. So, all right. So why do we need this? Why should we bother showing up early on a Saturday here? So it's our effort to make a world where information is understood. We want to be at a place where people can easily find and understand what they need. And World IE Day provides just a platform. So World IE Day is a nonprofit that provides foundation for learning about information architecture, our discipline, devoted to understanding of how information is structured and understood. So you could be a patron and donate to the cause. You can donate to the link that is here and we will share all the links with you in an email just after this session. So I'm glad to share that this year. In fact, today we'll be celebrating World IE Day across 50 locations, 24 countries and across five continents. Isn't that just great? And this definitely would not have been possible without the support of our sponsors. So thanks to our global sponsors and partners. Optimal Workshop who has been a long-standing partner for World IE Day at JAR and Veto, which is an even platform for who is also supporting as a platform partner at some locations. You ex-testing, Abby Covert, Enjoy HQ, Rosenfeld and our proud supporters across the world, Bentley University, Practical of Information, ProtoIO and others. And we have a surprise for you all. So for all the attendees who are here with us today, we have some attendees swag for you and we will share these links in an email just after this event. And also, there will be a small quiz at the end of the session. We'll send you an email. You can take part in that quiz and you can get a chance to enter into the global raffle for lotteries which has amazing prizes like Abby Covert's book, a signed copy. And you can also stand a chance to win noise cancelling set of headphones from Optimal Workshop. Also, you couple of you may stand a chance to win Azure RP team license. And that depends, but that depends on how you do in the quiz. All right. And again, thank you for our Platinum Sponsor Optimal Workshop. And yeah, we'll move on. Now coming to our event. So we have amazing set of talks lined up for today. But yeah, we have amazing set of talks lined up for today. We'll have Professor Ganesh Bagler who will be talking about computational gastronomy. Then we'll move on to Asad Ali Junaid who will definitely enlighten us with choice architecture and interaction design. And my personal favorite, a refreshing talk that I'm looking forward to is using IA to Garden Better by Anjul Tiwari. And of course visual narratives being the information designer that I am. Visual narratives to feed curiosity. This looks very interesting. And I look forward to this talk as well. And thanks to our local volunteer team. And this is an entirely volunteer run event. And this would not have been possible without Sovik. Thank you Sovik for leading it. And Abhishek will be tweeting all the live information for you guys. He'll be giving you a minute by minute updates on tweets and Manishani, that's myself and Saurabh. Thank you so much guys for putting this event together. And thank you to our New Delhi supporters as well, Miranj and Hasgeek. Hasgeek has been kind enough to be our platform support. And it just feels great to have this partnership with Hasgeek. Thank you Hasgeek. So quick thanks. If you are tweeting about this event, do include these hashtags. We would love to get in our conversation with you on Twitter. I know Abhishek is very active on Twitter. He will be very prompt in responding to you. If for any queries that you have, you can also ask us questions there. Within our talks, we'll try our best to include them and, you know, put this across to our speakers. All right. So like I was mentioning about earlier that after this event, we'll have a quiz shared with you. The winners of that quiz stand a chance to win all these giveaways. And yeah, we'll share the link with you. We look forward. Thank you and enjoy the event. We'll start at 1045 sharp. See you then. Thank you. Hello everyone. Thanks for joining us on World IE Day 2021, the New Delhi edition. As you can see on the screen, you should be able to see the schedule for the, for all the talks that we have lined up today. The first talk is going to start in a couple of minutes. Before we start the first talk, I would like to remind all our audience that we have a strict code of conduct for all World IE Day events around the world. You can read the entire code of conduct in by visiting the URL World IE Day dot org slash code of conduct. In case you face any harassment, firstly, we don't tolerate any harassment to speakers to participants to volunteers of any form. But in case you are a recipient of any harassment, please, you can reach out to the global team by emailing safety at World IE Day dot org. There is also a global keynote session which we are not playing out as a part of the New Delhi edition right now. It's a talk called ways of unknowing called by Cassini Nazir. You must watch this talk when you get a chance. The full keynote is available on that link, YouTube link that is shared and Cassini Nazir is also going to be available for a live Q&A in case you're interested. Please visit World IE Day dot org slash 2021 and there are three sessions, live Q&A sessions at three different time zones as you would see on the screen. And in case you want to visit one of those live stream sessions, live Q&A sessions, then a live streaming of the this talk video will happen right from 30 minutes before that chat session as well. So you can watch the entire talk then and immediately after that follow up with a live Q&A session with Cassini Nazir directory. Alright, with that, I'll move on to the first talk that is lined up for today. I would start with an introduction for Mr. Ganesh Baggler. So Professor Ganesh Baggler is an interdisciplinary researcher and is considered the pioneer of computational gastronomy. It's the emerging data science that blends food with artificial intelligence. He has an audacious dream of transforming the global food landscape through data driven innovations. And when I'd heard of this work that Professor Ganesh Baggler is doing, I was immediately excited because it's so hard to connect food with data or information and I felt that it's going to be an extremely relevant talk for our World IE Day New Delhi edition. And with that, I would request Professor Ganesh Baggler to please come on stage and share your presentation with us. Thank you for the introduction, Sovik. Let me get started. Food is extremely personal. Food is subjective and it's a creative endeavor. So I'm sure most of you are wondering how food possibly be connected with data, data science and information architecture. So my endeavor in today's talk is to unravel how computational gastronomy as a new data science is unraveling this area where we are trying to blend food with data and the power of computation. So I'm Ganesh Baggler. I'm affiliated with Center for Computational Biology at Triple I T Delhi. Let me see whether my slides are transitioning. Yeah, so I am affiliated with Center for Computational Biology at Triple I T Delhi and we at Triple I T Delhi are dedicated to the core practice of computer science and allied areas of which computational biology and now computational gastronomy is one of them. To tell you very briefly about myself, I was an aspiring astronomer as a teenager, I always wanted to be under the sky and discover new stars and do astronomy and astrophysics, which I could not. But it has been a long journey from being trained in physics, then in computational techniques, all the way in PhD in computational biology to where I am today, where I'm doing computational gastronomy and investigating food. So it has been a not so short journey from astronomy to gastronomy to put it in a succinctly. Having said that, let's come to the definition of what is computational gastronomy. It is a data science which blends food data with the power of computation for achieving data driven food innovations. It's a mouthful of a definition but it's succinctly puts together all the key terms which are relevant in the context of computational gastronomy. You would have noticed already that the word data has been repeatedly coming in this definition, telling us the relevance of information and data while we are looking at food from a data science perspective. When we think about food, we don't think about it from a data point of view. As I said, it's highly subjective. It's a creative endeavor. In fact, it's a magical phenomena when you look at food and cooking, it is about transforming raw ingredients into cooked delicious recipes. And that is a magical phenomena and cooking is a very uniquely human endeavor apart from ability to speak. This seems to be one of the abilities that homo sapiens have developed over a period of time, and which is rather unique to us. In fact, considered to be central to the evolution of disproportionately large brain sizes that we have, which have again contributed to the creativity and the artistic endeavors and mathematical abilities that we have are even abstract abilities that we have. To an extent that it is considered that cooking is very, very central to being human itself. Having said that, we have evolved from being a very primordial species to what we are today, species which dominates the face of the earth by way of technology, by way of all other aspects that homo sapiens can be attributed with. But at the same times, ironically, we have reached a stage where we are plagued with the diseases of lifestyles, wherein we seem to be plagued with these diseases such as cardiovascular disorder, obesity and type two diabetes, which can be attributed to many factors, but food is one of the important factor diet seems to be one of the key factors, and thereby it looks as if looking at food in an intense manner from a data science perspective is important for us to be able to investigate it to understand it better and to make our lives better. That is where information architecture data information science plays a huge role because food has not been seen until recently as a data science from the data science per view. It has not been seen. It's a very complex theme. It has multiple fact components to it. I'll be talking about those what are the, what are the different aspects that the food is characterized with, and they can be seen from a data science perspective. So in my in today's talk, whatever I'll be presenting essentially provides you a view by which we are looking at food and trying to make it understandable. We are like trying to look at food from a data perspective and making it computable in a big way. And this has not happened until recently only only in the few recent research articles in the last five to six seven years back, people have been looking at food from a data architecture and information science perspective. One of the components that characterizes food is that of the traditional recipes. Those are the capsules, traditional capsules that have been passed on to us from generations to generations. Then we have ingredients that go into the recipes, their flavors, the nutritional value of the ingredients, as well as the health impact of these foods by virtue of consuming them. So all of these components are one of the easy way by which we can think of organizing food, organizing the cooked food that we end up consuming on a day to day basis. In my lab, we have been building data structures and algorithms corresponding to each of these elements of food and trying to put the six or pieces of food puzzle together, so as to make data driven food innovations possible. So this is the story of data driven food innovations that I'm going to tell you. But before I get into telling you what is the future of the food when seen from information science perspective. Let me tell you a story that began in IIT, Jodhpur where I was working as an assistant professor in 1914-15 where we started asking a curiosity driven question. One of the themes that the key theme that today's IIT in 2021 is being here is that of curiosity. So it's very interesting that this was the question that we were asking when we began our journey into computational gastronomy, which was not named so at that point of time, only in retrospect it is being called as computational gastronomy. The question that we asked was why do we eat what we eat? This is a deep philosophical question and a very tough one to answer if I may say the reason being that well all the components of food related matter of what we eat are not easily available. But at the same time we know that what we eat are recipes which are nothing but combination of ingredients. By making such a reductionist approach by taking that approach, we can now transform this question into this one. Why do we combine ingredients the way we do? Is there a rule? Are there formulae by which we are using certain combinations of ingredients more frequently in our recipes as opposed to other combinations? Well, that's an interesting question to ask. And you can think about if there are answers to this kind of a question. It turns out that a mature chef who has been practicing food and cooking for a while has had an answer to this question. And this answer came from a chef, a chef Heston Blumenthal from United Kingdom proposed a hypothesis called the food pairing hypothesis, in which he suggested that ingredient that tastes similar tend to go well with each other. And that is what was the proposition that was given by the food pairing hypothesis. It essentially says that if there are two ingredients and if they have similarity in them and we'll see how we define similarity very briefly. And if the similarity is more than the propensity of they being used in the recipes would be higher. That's what was suggested. And in one of the research articles published in 2011, this proposition was in fact shown to be true for the western cousins. The Latin American, Northern American, Eastern and Southern European cousins, it was shown that indeed food pairing hypothesis tends to hold true when you look at recipes coming from these cousins. What was not done was to look at cuisine such as Indian, India's Indian cuisine, which has got a long history and is characterized with diversity. It is culturally rich and has got health-centric dietary practices. So it had not been seen in the context of Indian cousins, which is what we wanted to do when we started our investigation in 2014 by collecting the data of various aspects of Indian cuisine. So how do you go along and start collecting the data of recipes, ingredients and create a repository which will help us answer the question on food pairing in the context of Indian cuisine. Well, we had to find repositories that are already available online since we had to collect it from online repositories. So, Tarla Dalal was one of the repositories that we fixated upon after looking at various other repositories such as NDTV.com, Sanjeev Kapoor.com and others, because Tarla Dalal had a very structured database of recipes coming from a length and breadth of the country, which are pinned down to various regional cuisine. So we collected a diverse set of 2,543 recipes, which were coming from Bengali, Gujarati, Jain, Maharashtrian, Mughalai, Punjabi, Rajasthani and South Indian Cuisines. As you can see, these are, while they are not comprehensive in terms of representing the whole of India, they are definitely are representative of most of the Indian Cuisines that you can think about. So, these recipes were composed of around 193 ingredients and each of these ingredients can further be structured into one of the categories, can be classified into one of the categories exclusively such as that of cereal, dairy, flower, fruit, herb, spice, pulse, vegetable, etc. As you can see that all of a sudden what was otherwise qualitative, what was otherwise personal has been transformed into a data science. We have created a structured repository of recipes wherein we have recipes, ingredients within them and the categories to which they belong to. But incidentally, to be able to answer it from the food pairing perspective, we need to look at their taste and order because taste and order are the primary factors because of which a given ingredient is chosen to be a part of a recipe. And these ingredients act via what are called as the olfactory and gestatory mechanisms of human body by which they trigger sensory mechanisms and thereby giving rise to their selection by the cultural mechanisms. Well, what we did further was to collect the data of flavor molecules which are the key contributors towards the order and the taste in a given ingredient such as onion, chili, tomato, cauliflower, tamarind, etc. And created a tripartite data repository wherein recipes was one layer of the data, ingredients of which the recipes are composed of was the second layer of the data and further flavor molecules which are found in these ingredients was the third layer of data. And by that way we were now able to get a tripartite data structure which is representing Indian cuisine. All of this was done to ask the question on food pairing that Heston Pliwanthal was suggesting. And for that reason, we had to look at individual recipes such as the caricature recipe which has been shown here, break down the recipe into its constituent ingredients such as these four ingredients that we are looking at and come up with all kind of pairs for example, onion and coconut, coconut and chili, etc. Or pairwise combination of ingredients and find out how many flavor molecules are shared or how much is the similarity between the two ingredients. Here we are considering similarity between two ingredient is represented by number of shared flavor molecules with the assumption that the taste and order or the flavor of an ingredient is made primarily represented by virtue of its flavor profile. That's an assumption we are making here, but that's a fairly good assumption to make at the first stage. Having collected these numbers, we can take a simple plain average and get the first number which is nothing but the food pairing index, which represents for a given recipe, the average number of shared flavor molecules in that recipe. Having got this number, remember, this is the first time a recipe has been quantified, which is otherwise extremely qualitative in terms of its taste and flavor. And this has been done by using the molecular basis of each ingredient, the flavor basis of each ingredient that goes into the recipe. All of this exercise was done in the context of the fact that what would happen if in these ingredients were being put together randomly. As opposed to being by product of cultural evolution, recipes have evolved as part of cultural evolution by impact by virtue of being impacted by geography, climate, culture, genetics, and other factors playing a role. That's how the recipes have been shaped, but had these factors not being playing a role, what would have happened to the recipes if you ask that question turns out that the recipes would have had a random architecture. They would have been put together in an absolutely random fashion, that's what we assumed and we call it a monkey cuisine. Compared to a monkey cuisine, the western cuisines tend to have positive food pairing by virtue of having ingredient pairs which are very much similar to each other by in terms of their flavor profiles. What we observed in the context of recipes that we had compiled from the data was that these recipes are characterized with a contrasting blend of ingredients. These recipes tend to have ingredient combinations, opposite to what has been found in the case of western cuisines, a blend of ingredients that are very, very different in terms of their flavor profile. They are having less and less sharing of the flavor profiles. That's what we ended up observing and this is what we characterized or called it as contrasting blend or negative food pairing. The other experiment that we did, which will be the last experiment that I'll be telling as part of this story was to find out what is the role of placement of ingredients of a given category. If I were to shuffle ingredients of vegetable category, what would happen to the food pairing index? This is the food pairing index originally of the cuisine. That's what we just saw. It is on the contrasting side as opposed to the monkey cuisine, which is somewhere here. What would happen if I were to look at the whole cuisine and were to shuffle vegetables within vegetable category? Every vegetable will be shuffled and will be kept with some other randomly picked vegetable from the basket of vegetables. Everything else being the same. All other ingredients are being kept as it is. If you do this experiment, turns out that independently when this experiment is done for each category, the food pairing index changes only marginally. Suggesting that shuffling within category, the so-called intra-category shuffling of ingredients doesn't make much of a difference in terms of its food pairing index. Possibly that has some bearing on the taste and the flavor of the recipe itself, but we won't make such far-reaching conclusions. We shall only consider the food pairing index. And we find that indeed these categories seem to be rather indifferent to the food pairing index. On the contrary, there is one category which seems to be rather critical in deciding the food pairing index. And you can guess what category that would be in the context of Indian cuisine and Indian recipes. Turns out that if you were to do the shuffling, I mean you can think about it. The audiences can actually pop answers in case you have any answer to this. But incidentally, the one category that seems to be rather central and critical in deciding the food pairing index is that of the spices. If you were to shuffle spices and where to put cardamom instead of clove and some other cinnamon instead of clove and such kind of shuffling if you were to do it randomly in the Indian recipes, then the food pairing index dramatically changes from what it was to what it would become more like a monkey cuisine. Suggesting that spices are the molecular fulcrum of Indian recipes. Right, so that's the conclusion that we drew and this conclusion is extremely robust, regardless of which recipes we end up taking. Whether we have expanded our repository from what was only 2543 to close to 10,000 recipes and despite that this particular observations remain extremely stable. That's how we concluded that spice is the taste of India. That is something which characteristically defines what Indian cuisine is all about, and we went ahead and created what could be called as a culinary fingerprint. Like a DNA fingerprint which characterizes a human being, culinary fingerprint characterizes a cuisine and tells us the quintessential aspect of a cuisine, which category of an ingredient contributes most to the food pairing, etc. And we found out that spices and dairy products are key contributors to the food pairing across all the regional cousins that we had investigated. And the nature of contribution changes from one regional cuisine to other regional cuisine, thereby telling us something unique about each of the cuisine that we have seen. This is what we characterized as or named it as culinary fingerprints and have published it in a scientific journal. Going beyond Indian cuisine, we have also characterized the culinary fingerprints of more than 26 world regions and around 84 countries across the world in terms of their culinary fingerprints to find out what is so unique and characteristic of each of these regional cousins from across the world. This research which was done in 2015 published in 2015 was declared as an emerging technology by MIT Tech Review and was considered to be best of 2015 among the research articles that were published on archive. We saw much publicity from the media, given the fact that we are looking at one of the unique cultural heritage from coming from Indian recipes and Indian cuisine and has been highlighted in rude food by Veer Sangvi and various other news outlets, more than 200 news outlets have published it including scientific portals such as Chemistry World have highlighted it and chefs have been inspired by the research that we have done to consider how to tweak the recipes by looking at the food pairing pattern that we have observed in various world cuisine, Indian cuisine being the pinnacle of those and this is one of the examples of that where Chef Darima Arora a Michelin star chef has used our research for shaping her recipes in a restaurant. It is also considered to be the sweet spot of the food given the fact that this is one of the earliest studies which has looked into food from a data perspective and has tried to blend food with data and the power of computation. This is where when we looked back I decided to name this area as a computational gastronomy and we went further and invested ourselves in my lab to investigate the computational gastronomy various aspects of computational gastronomy build various databases around computational gastronomy arena and algorithms which can be used to ask interesting questions in computational gastronomy. Having told you this story. Now, what I would like to do is to tell you about the future of computational gastronomy and how information science data science can play a big role in shaping what the way we will be looking at food in decades to come. This is not going to happen in a day or two. Computational gastronomy is where physics was at physics was in 16th or 17th century, there is a lot of potential, but there is a lot of work to be done. Huge amount of data need to be organized around food, around recipes across the world, the ingredients in them, the flavors in them, the nutritional value, the health impacts of the ingredients and various other data which can be collected about the food. But before telling you about the future of that future of computational gastronomy, I would like to tell you about what are the shortcomings that we have encountered in the study that was done in 2015. What we realized is that we have fallen prey to the reductionism that typically physicists tend to approach any given complex topic. Given a topic of modeling a cow, a typical mathematician or a physicist would start with an assumption by saying that let's assume that the cow is spherical. And that obviously is not the truth, but that gives us a very good assumption to get started off with depending on what is the question that we are answering. And that is how we started by assuming that a recipe is nothing but a set of ingredients, which it is not. The order of the ingredient matters, the kind of processing that is being done on the ingredient matters, the kind of flavor molecules that go into each ingredient matter. The concentration of the flavor molecules also is critical and the potency of the flavor molecule is also of value that makes a huge difference in how the flavor is going to come out in a cooked recipe. All of these factors were completely ignored. And we would like to point out that these need to be incorporated while we are building this area further to ask more interesting questions and deeper questions in computational gastronomy. And I'm sure many of these questions would have come to your mind, as I was telling you about the story that unfolded in 2015 in my lab at IIT Jodhpur. What we are trying to do in the future, starting from 2016 onwards, is to build databases and algorithms to make food computable. And that itself is a major challenge, because so far food has never been computable. The data of food has not been compiled in a systematic manner to make it computable. The idea behind this is to achieve data-driven food innovations. Can we come up with applications of databases and algorithms such thus devised is the question that we are asking here. And why do we do so? There are three dimensions that we're looking at. The first dimension is that of the recipes and the data of recipes. The second is that of the molecules' data coming from each of the ingredients. And finally, the dimension of nutrition and health is what we are looking at. Each of these, I will be spending a little time to explain you about how we are working on each of these dimensions and tell you about the future of each of these aspects. The first thing is that of the recipes space. Going beyond only around 2543 from Indian recipes that we had compiled, now we have a huge repository of a few hundred thousand recipes which have been compiled in a structured manner. And this one has particularly 118,000, 118,000 recipes from across the world, which includes around 26 world regions and 84 countries, which have been compiled in an extremely structured manner. And when I say structured, I'm talking about recipe, its name, the cuisine from where it comes from, and when I say talk about cuisine, it can be the continent region and the country to which it belongs to specifically in a systematic manner. So the dietary attribute that the recipe has in terms of five dietary styles, what kind of cooking processes such as boiling, frying, sauteing, etc. are going into making of the recipe. The utensils which are being used for making of the recipe, the ingredients which are of course part of this recipe, the order in which the ingredients are being put into the quantity, unit state, temperature, whether it is dry and fresh, or the size it is being put into the recipe, and further the category of each ingredient. All of this data in addition to the nutritional profile of each ingredient, which has been compiled from United State Department of Agriculture's comprehensive database of nutritional values. This particular repository is a wonderful archive for anybody who would like to explore recipes from the cultural perspective, from a nutritional perspective, or from dietary attributes perspective, or other values for that matter if you would like to investigate. So this is one of the first repositories that has been published in a scientific space, and is being made available freely for investigation for those who would like to build applications on top of it. Going beyond compiling the data of recipes from various world cuisines, we have also asked an interesting question about the evolution of cuisines and how they are interrelated with each other. So we have built information theoretical majors about relatedness of cuisines, about how similar or dissimilar they are from each other, and have tried to find out, try to build a tree of world cuisines. And this particular tree is one of the earliest tree that we have built, we are refining it further, telling us about similarities in terms of cuisines and how they are matching to each other in terms of ingredient composition and use and less of less use of ingredient or more use of ingredient in various cuisines. This also tells us primarily that geographic proximity have played a huge role in shaping similarities of the cuisines, apart from the fact that there are certain cuisines which while being away from each other are still very similar to each other. So this is the tree of cuisines similar to the language tree that must have seen must have encountered at certain point of time. One of the most interesting question that we are asking in our lab is that of how many recipes are theoretically possible given any composition that you would like to create given all the ingredients that are available in the world. Turns out that the number of ingredients can be estimated to be of the order of 1000 and this number is a very rough estimate and a conservative estimate, given the fact that we are bundling all the onions into only one onion category. Similarly, all potatoes are considered to be a potato and similarly other ingredients are bundled despite they having a lot of different varieties in there. If you consider all specific varieties then probably the number will come into few tens of thousands. Our estimate from recipe DB is that of around 20,000. But then we are going by the conservative estimate of 1000 and a typical recipe has got around 10 ingredients. Well, then if I were to make all kind of permutations and combinations, then the number of recipes that can be theoretically be possible is of the order of 10 to the power 30, which is more than the number of stars that are there in the whole universe. I mean that's a huge number that we are encountering here. Of course, a lot of these recipes would be unpalatable and therefore would be discarded in the process of evolution, but nonetheless, we would still have a large number of recipes which have not been encountered by cultural evolution. Question is, in 21st century, with the power of data and with the power of information architecture, the data that has been compiled, can we now create new recipes is the question that we are asking. And that is the direction that we are taking in our lab with the tools of machine learning, text mining and natural language processing. I don't know how many of you have seen this movie Ratatouille, maybe we can have raise your hands who have seen this movie I'm sure many of you are smiling in your cheek when you're looking at this, this particular picture, because it's a humorous movie about an adventurous rat which wants to, which wants to cook, and the catch phrase, which keeps running through the movie by from chef Gustave is that anyone can cook. So we want to take this adage further and ask this question, can computers cook, can computers be taught by virtue of structured databases and repositories to cook by creating repository of world ingredients, ingredients, flavor molecules, nutritional value, can they be taught how to cook and can they be allowed to capture the intuition that otherwise a human chef or a human culinary enthusiasts would have. Well, that's the question we are asking. And towards this, one of the early articles that we have published is here, published in a conference called Colling, which deals with natural language processing, wherein we are trying to create new recipes with the help of data repositories that we have built, including recipe DB being one of them. And remember, this creation of such recipes is not very far away from what has been done in the context of music in the context of literature, where people are trying to imitate Shakespeare, or people are trying to imitate Mozart or Beethoven symphonies and computers are to an extent very successful in trying to imitate these masters. So say in a same manner we are asking this question, can intelligence culinary intelligence itself can be imitated. And the answer is far too away, we believe that the first step will be that of creating recipes, which can be fooling a chef, we call it a tuning test for a chef. We are running this experiment very soon by the end of this year, where we'll be challenging the chef to recognize a recipe which is computer generated versus the recipe which is human generated. That's the challenge that a chef will be passed on to. And if we are able to fool the chef statistically speaking in a significant manner that a recipe which is generated by a computer is actually a human generated. That's what the chef would think, then we'll be partly be successful. The next test would be actually cooking these recipes and seeing how palatable these are. And that I believe is going to be a bigger challenge for us, because not all the attributes would be available to us to make a palatable recipe as of now. So this might take a few years before we can become successful towards that end. Well, that was about the recipe space. The next phase is that of the molecule space. Given the fact that molecules are key contributors to the taste and order, we took up the job of organizing the molecules, the flavor molecules that are found in all natural ingredients that are going into recipes. And this I'm very proud to say flavor DB is the world's global standard for flavor molecules is repository today having published in 2016 17. And this compiles more than 1000 around close to 1000 ingredients and their flavor molecules which are a few few hundred few thousand flavor molecules, of which these ingredients are composed off. And these are experimentally validated flavor molecules using GC MS gas chromatography mass spectroscopy studies that has been organized in an extremely structured manner, starting with ingredient, their natural source from where they come from which animal do they come from the flavor molecule of which are which are found in the ingredient, the kind of flavor attributes that the molecule has got and the chemical properties that the molecule has got and made various other interlink properties are also available as part of the flavor DB as a database. We have gone ahead and publish this as an Android app to make it available to culinary enthusiasts who would like to do a bit of playing around with food pairing, because it allows you to find out ingredient combinations, which are very very similar in terms of their flavor profile. This is under right now it has been taken down because we are updating this flavor DB database. We have gone ahead and remember that creating repository is not only for the purpose of searching and archiving it and then searching it searching this databases, but also towards coming up with predictive tools. This is a predictive tool that we have built for bitter and sweet taste prediction, where in starting with a bunch of flavor ingredient in molecules, which have been characterized with bitter and sweet taste, we have come up with ways by which machine learning algorithms we have built to come up with prediction of what kind of a taste it would have. And this we have done only for bitter and sweet, because the data was not available for all the other taste such as salty flavor and umami, not enough data was available. And this data was good enough for us to build algorithm, which is a state of the art algorithm. And again, this data of bitter and sweet molecule, the algorithm, the software which has been built, as well as a web interface, which has been denoted at the bottom are all available freely for playing around. And remember, all of these takes us towards a very interesting application for designing molecules, which are having expected level of sweetness, but doesn't have enough nutritional value. This is important given the epidemic of diabetes that we are facing to have enough number of sweetness which can, which which don't add to the nutritional value. And that's one of the challenges in the technical space that is that we are facing. And this tool allows us to play with that and address that question. Moving to the last space, last dimension that I was talking about is that of the health dimension. Incidentally, we keep coming across various news pieces and articles, which talk about whether oil is good for you, egg is good for you wine is good for you or bad for your health, which aspect of health it is impacting etc. And once we come up with contradictory assertions about the same ingredient being good and bad for the same disorder. And this, in my opinion comes because the way food interacts with human body, giving rise to health consequences is way too complex to comprehend through simple equations, such as the one which has been in front of you, right, it's not so straightforward. And that is the reason why we need to put together data repositories of health impacts of food ingredients based on empirical evidence to be able to come up with interesting assertions, which are coming from evidence based studies. And that's why we ended up building something called diet rx. This is under publication under review at the moment it is not yet published, but the database is available for exploration for anyone who would like to play with it. We have created in house repository of food and disease associations, based on close to 38,000 research articles published in the last 90 years. And these data are created these repositories created by looking at the title and the abstract of a given research article, and the mention of food that has been made in the research article that we have put together and have come up with these food and disease associations food. We have identified using something called named entity recognition methodology. And similarly, diseases have been identified using any are the name entity recognition of diseases. So we have built sophisticated algorithms for identifying the name of a food or a name for disease in the title and abstract that's what we have been able to do. Further, we have integrated this data remember it's not all about creating new databases but also incorporating existing data repositories, which are relevant in the context of food, so as to create an integrated database. So we have integrated that with chemicals that are found in the food and genes which are linked with the diseases, thereby creating a quadripartite database of food chemical gene and disease associations, and all of these associations are empirically proven empirically to create barring food and gene association which is not available to us as of now that's been only shown for graphical purposes but it is not available with us. And that there by provides you a means by which you can explore this repository for finding out what food is good against which disease, and what food is possibly impacting or causing a particular disorder or a disease as per the medical subject heading, so go back and find out which literature is being cited corresponding to that particular association identified using test finding protocol. So that's DietRx for you, providing an extensive repository of food disease associations interlinked with chemicals and genes. As an application of DietRx, we have investigated these data of food disease association in the context of spices, and the reason we did this was twofold. One is spices are coming out to be critical, not only in the context of Indian cuisine, but many of the cuisines from the world, including the Oriental cuisines, where it seems that the herbs and spices seems to be playing a critical role in shaping the food pairing index. That was one reason. Second was the question which is generically being asked is, why do we use spices in our food is one of the important question that we asked. And one of the interesting answer to this is in terms of the flavor and the fragrance that the spices offer in our food. And given that one of the one of the question that has been asked is that the nutritional value is none for the spices so why do we use spices. So data of food disease association of herbs and spices culinary herbs and spices was investigated, and we found that indeed spices contribute towards various beneficial health impacts, and we found that there is a broad spectrum benefit of culinary herbs and spices. And that's the conclusion that we drew from the investigation coming from dietetics, and thereby suggesting that herbs and spices have a role to play beyond being value of value for their, for their impact for towards flavor and fragments that they offer to us towards We'd like to close in a couple of minutes if that. Yeah, thank you. So towards conclusion of the talk, I would like to tell you that computational gastronomy is a science which tries to blend the data science with food and the power of computation. And for doing so, we are asking the hypothesis given questions. We are building structured databases, and we are coming up with applications based on these databases and knowledge knowledge that we are creating as part of the endeavor of computational gastronomy. And one of the possible direction that we see is that of personalized nutrition, which seems to be one of the potential space that we are moving into, where in with the help of machine learning data science and computational gastronomy, we can build algorithms which can tell us which food should be consumed to improve our health. So this is a particular research article which tells us about how do we eat a food which will help us decreasing the postprandial glucose level the glucose level after eating consuming a moon meals, which is critical towards indicator towards type two diabetes. And that's one of the applications that I would like to point as I conclude and would tell you about various things that we are doing in our lab, including food beverage pairing prediction of taste and order culinary fingerprinting dietary interventions, design of food and beverages, and all the way up to moving in the direction of creating sustainable food innovations. And all of this has started only six years back so I would like to emphasize this that computational gastronomy is a very young area, and many of us who are data scientists and computer scientists have a lot to contribute those who have keep any interest even peripheral interest in food can actually contribute to this area by way of investigating the data that is being churned out from our lab. So towards the end I would like to stop with this quote, which says that the discovery of a new dish converse more happiness on humanity than the discovery of a new star. And I have not been able to discover new stars by being an astronomer which is what was my aspiration as a as a teenager, but I hope I'll be able to discover new dishes to make humanity happier and healthier with the help of data structuring and information communication of information architecture that we are doing in the context of food. I would like to thank a large number of my students PhD students m tech b tech and interns, who I can't even name because of the large numbers that are there, and who have contributed towards these research and would stop with that for question Thank you. Thanks a lot Professor Ganesh. This was an amazing conversation, and I have definitely learned today that every time I go to the kitchen I produce something from the monkey cuisine domain. So I think this is a great story we have about so I would like to say to the audience. If you're watching live on YouTube, you can post your questions on on the live chat. If you're on zoom, you can use the Q&A panel to pose questions, and we'll try to take those questions to Professor Wagner. There are we have about eight to nine minutes available to do questions before we move on to the next speaker. So I think one of the great things that Professor Wagner is doing is if you've taken a concept that is food which is an everyday occurrence for us. You have used the concepts of information architecture to make sense, extract the tacit knowledge that is there or tacit information that is there in food, made sense of it, and you're showing us what is possible once you've made sense of it. But tell us, how was it when you started off? What were the challenges in discovering the fact that food which is often considered a cultural thing or a common thing, never thought from a data lens or an information lens. How was it to start it started off and what were the challenges you might have faced at the beginning? Okay, good that you asked that question. Even as a professor in IIT, it was extremely challenging where it is considered that freedom is the first thing that is given to you. It was extremely challenging. Because when I started this investigation, we started we created the first compilation of recipes, and we did the first analytics and I presented this to a couple of my colleagues who are coming from physics and mathematics background. They were extremely critical. They said that why are you working on food and recipes? Why are you looking at these data? Compiling this data itself was challenging because it was not easily available. Names of ingredients were coming in different, they were different avatars, the vernacular languages were being used for naming the ingredients just to give a couple of challenges that we faced at the beginning. So right from challenges of the data to the challenges from external externalities like people not accepting the analysis as a meaningful science, we faced quite a bit of tussle here to take it forward. Right, so you're in many ways challenging common notion that you always take for granted that food is food, it's completely subjective and there is no objective quotient to it in a way. And if you can break it down and structure it so well that kind of gives this new dimension. Now, the question I have is what do you believe is the role of curiosity and experimentation in order to come up with these dimensions that you're talking about or these ways of looking at food that you're coming in. What's the role of experiments and I'll make it a question a little larger because of course you'll answer it from a food point of view. If someone is taking any other subjective domain like from food it can be any other field tomorrow, any other field of art as a matter. What is the role of experiments and how do you start to begin to do those, get an understanding of this art form. Right, I think any creative domain or qualitative domain that we would like to deal from a data science perspective or from an information architecture perspective, one need to respect that at some level there is there is an intuition which goes into it whether it is music, painting, food or similar domains we need to respect that part and it will remain there for a while because we do not know whether we will be able to break down it into entirety in its structure in its constituent elements to be able to apply the tools of computation. We won't know that but we can start with that hope that yes it is possible to break it down at least at its first level we can break it down into its constituent elements like we did in the context of food in terms of ingredients, flavor molecules, nutrition, taste and order etc we can do that whether this can be used for challenging human intuition and creativity is a completely different ballgame altogether. So that is what I would like to suggest that people should not shy away from experimenting and collecting the data and trying to break down the qualitative arena into its constituent components one should do that definitely. Right and how do you figure out what are the constituent components in the first place like if you would catch hold of a common person on the street and say that break down the constituent components of food. No one will be able to say oh there are pulses, there are meats, there are foods like these categories of food that you have also specified which is very core to information architecture. How do you come up with these? Well these were not easy to come by partly we had to deal partly we had to rely on the scientific literature about what has been talked about these attributes earlier and partly we had to talk to an expert for example a chef we had to connect with them for just to give an illustration. I will say that when I was trying to find come up with cooking protocols what are the different cooking protocols which are being used I could not rely on my intuition I created a list of one I think it was 270 different cooking processes which going to cooking and we presented it to a chef expert chef to ask his or her to find out okay to tell us what are the outliers here what are similar processes and which could be called as distinctly independent cooking protocols so that is a protocol that is the process that we need to use. Right so basically go to people who are doing it and get as much learning from them as possible. Okay let me take a couple of audience questions at this point of time before I move further. We have about four minutes left for the question and answers. Okay so the first question that was asked by Kiran Sethi it is why do we need to predict bitter and sweet taste it's very obvious that having a bitter component in a recipe will make it bitter only and same goes with sweet taste prediction so can you answer this question for us. Okay this is more from the context of designing a sweetener so there is a sweet design sweetener industry which is trying to come up with synthetic molecules which are sweet in taste but at the same time don't have nutritional value and that is there is a possibility that computationally you can design lots of molecules or million molecules predict its sweetness value and identify only those which are sweet enough which can be taken further for their experimentation that is where predicting sweetness filter in computationally is important. Got it. So Kanendra asks how personalized nutrition will be different or better than personalized medicine is food can food molecules be used as drugs as well. I won't go to the level of talk saying that food molecules themselves be used as drugs but I will talk more about how food itself can be used as a potential mediating agent for treating a disease like in the context of diabetes. What you eat is going to trigger your post-prandial glucose level and thereby if you can design or if you can find out food which triggers desirable response in me vis-a-vis Kanendra that would be extremely wanted it will be highly desirable. I would like to find that out that is what personalized nutrition is expected to do it is not expected to go to the level of molecules to find out personalized drug as such. Okay, so here's one question that I briefly touched upon but this has come from Nitya food also has a political cultural and religious aspects that are hard to dissociate from. So if you want to quantify such things or create classifications for them how would you go about it so you briefly touched upon that anything you want to add for this. Nitya has touched upon a very interesting aspect of food food has got extremely so a science social aspect of it political aspect of it. It is politicized heavily think about beef if you want you know. So yes, it's very difficult it's very very challenging what we are trying to do is to ensure that the recipe compilation that we make is not only allied representation which is what we were alleged in the beginning that all the recipes which are documented are coming from allied communities and they don't represent Dalit recipes and tribal recipes etc. So it is when I'm declaring a cuisine to be Indian cuisine that I incorporate enough recipes that are also coming from such not not a light communities as well. So we try our best to ensure that we don't do any injustice you know to communities which are not represented enough but to account for political aspect is very difficult I have not even. Okay the last question from the audience that I'll be taking is what determines whether a molecule or a collection of molecules will have a particular flavor or a new flavor will emerge out of it. Very good question. First of all, individual molecule has got certain taste or order based on the functional group it has got or combination of functional group it has got. Although there is no clear science available about it there is there is artistic people who they are they are flavor scientists they are called but they are flavor artists more less of a scientist. They know how to put molecules together to come up with a desirable taste of a mango or a desirable taste of a tamarind etc they know how to do that. Right, so flavor arises out partly out of individual molecules is functional groups and partly because of multiple molecules coming together and giving rise to an emergent property. Both of them are contributing towards the final taste and order. All right, I'll ask the final question as a closure question at this point of time. After doing so much work have you managed to create a unique recipe by yourself. Good question and a pinching question at the same time. No, we have not been able to do so and I believe that to crack human intuition or rather to crack something which human intuition does is going to take some time. So that's a very tough one to break in our lab I must say but we hope to do so in the coming times. Now that we have algorithm which can design recipes we hope to at least handpick some recipes which are palatable and present to the world saying that this is a recipe which is coming out of a computer. Right, all the very best to you Professor Wagner I hope you do come up with a few of them and I would love to tell you whether they are palatable or not. And for all the other audience if you have further questions you can use the updates and the discussion tab in the Hasgeek channel event channel where you can post further questions and we can request Professor Wagner to come in and answer those questions for you. Thanks a lot Professor Ganesh and now I'd ask Manish to take it for the next talk. Thank you Sovik. Thank you Professor Ganesh. It was truly an amazing talk. So we'll break for a few minutes. We'll resume the session at 1145 with Ganesh's talk. Hello everyone and welcome back. So there will be a quiz at the end of the session. The link will be shared to you with an email and I hope you'll participate in that and definitely there are some exciting prizes there. Amazing. So let's move on to our next speaker. Asad Ali Junaid. Asad Ali Junaid is an experienced designer at Adobe. Junaid has about 15 years of industry experience as an experience and interaction designer contributing to companies such as Oracle, IBM, Intel, among others. Junaid has a multidisciplinary background with masters of science degrees in human factors psychology and in electrical engineering as well. He is also an entrepreneur and an author of an epistolary novel. He has several teaching credits and regularly teaches and conducts workshops on human computer interaction and creativity. Today Junaid will be talking about choice architecture and interaction design. Over to you Junaid. Thank you Manish. Thank you Professor Ganesh for the enlightening talk and thank you Sovik for making this happen. So let me share my screen and start it off. You guys can see my screen. No objections there. Okay, so basically a couple of years back, right? What happened? I set it out to control my phone usage. So I wanted to control this by saying that, okay, I will charge only once a day. Morning for one hour 45 minutes or whatever. So I should charge early in the morning before I used to go to work and then charge it again the next day morning. So that was how I wanted to control my cell phone usage. So what used to happen for some days was the phone battery used to drain out by the evening and I was using the same amount of phone. I used to use the same phone. I never used to use it as much. I never started using it more. So I couldn't figure out what was happening. So the crux of this problem, the crux of my introduction to the whole area came from that single problem. So I mean it's again the theme of this talk is curiosity and that's where it started for me with curiosity. So I'm just trying to, yeah. So I have an iPhone. I'm sure people who have the iPhone are familiar with this menu. You just scroll up on the iPhone and you get this menu. There are different options for it. You need to turn on the data packet, the Wi-Fi, Bluetooth and there's music options and all of that. So we'll just focus on the top section here where you see the data on and off, Wi-Fi on and off and Bluetooth on and off. So we'll focus on that. So what used to happen was I figured that the Wi-Fi of my phone used to automatically turn on. So I mean though I didn't turn it on by myself, I used to switch it off and then I used to go to office by the time I reached office it used to be on. I was like, I mean, what's happening? So then I researched further and then I discovered that there are two different states for turning the data package on and the Wi-Fi on. So for turning on the data packet, I mean the 4G data, you have the one on the top there. You click on the off state goes to the on state goes in a green color. But for the Wi-Fi, you click Wi-Fi has a cross across it. And when you click it, it turns on using the, I mean it chose on in the blue color. I mean, I didn't pay much attention to it. I didn't know what was going on. I mean, as long as I'm getting the Wi-Fi, I'm fine. As long I'm getting the data package whenever there's no Wi-Fi, that's fine as well. Right. So I don't, I didn't care as much then. But I started noticing this was what was happening. And the on state for the data packet you see was green, the on state for the Wi-Fi and the Bluetooth was in the blue. Similarly, I mean, but the surprising thing was when you turn off the data packet, it went back to its previous state. The off state as what I thought was it wasn't going back to the earlier that state which was there, but it used to be in this like light green shade, what you see here at the bottom. So I'm like, why is this happening? So I wasn't sure what was going on. So when you turn the Wi-Fi on and if you want to switch the Wi-Fi off, it didn't directly go to this state. It went to this intermediate state below. And from this intermediate state, the Wi-Fi of the phone used to turn on automatically. And I was like, what's going on? I mean, I mean, I didn't, I mean, it's not in the control. I mean, what am I doing wrong? And why is this happening with the iPhone? I'm sure people who have an iPhone might have experienced this as well. Similarly, with the Bluetooth, there are three Bluetooth states. There's a Bluetooth off, there's a Bluetooth on and there's an intermediate Bluetooth state where it remains there and it can turn on on its own. I'm like, what's going on? So even in the control panel, right? So whenever the Wi-Fi and the Bluetooth are in this intermediate state, you have this saying that it's not connected. Though it seemed to be on, but it says it's not connected. And the mobile data, I mean, it's off there. So then I figured out that if I have to turn the Wi-Fi completely off, I have to go to the settings and click on this. It says it is on and then I have to turn it off from here. So it's like, I mean, three steps, I mean, you have to go to the settings and you have to do all of this for turning the Wi-Fi off. And whenever the Wi-Fi was, Wi-Fi was in this intermediate, yeah, this is how it is when it is, it clearly says it is off when the Wi-Fi and the Bluetooth are in this state. It clearly says it's off. So, I mean, is it such a big deal, right? I mean, why is this happening? And I wanted to figure out because it was, if the Wi-Fi is in this state, right? It was draining my battery. It was continuously searching for a signal. It turns on its own. It's draining my battery and it's not lasting the entire day like I had imagined it to. So, I mean, this was like a big deal for me specifically and I thought, okay, is it a big deal for others as well? Then I start Googling for answers and this is where I ended up. And this is like a little dated post, like 2017 post. I'm sure there are a lot of other posts now, but a lot of people have asked, how do I stop my iPhone from connecting to Wi-Fi automatically? And how do I... So, I mean, this was a real problem, right? So, it is indeed a big deal for people like people in India, people like me at least who don't want to keep charging their phone multiple times during the day. I mean, we might be going out, we might be in a different... I mean, you can buy a bulky, you are charging this thing, but why do you want to do it? I don't want to carry that big bulky charger with me and all of that. So, I didn't want all of this to happen. So, that's the background to it. So, then I figured out that if the iPhone is not able to connect to any Wi-Fi network when it is on, it will keep searching for networks or hotspots. And this drains the battery. And it will continue to search for networks for as long as the iPhone to authenticate the Wi-Fi network to connect. So, this was, I mean, just draining the battery here. Then this is the official thing which happens, okay? So, in iOS 11 onwards, basically, the iPhone, it is hard to switch on by itself. So, when does it switch on? The user walks or drives to a new location. Whenever the user goes to a new location, maybe from home to office, it turns on by itself. And in between, for me, my commute is more than an hour. And in between, in this one hour, it continues searching for this Wi-Fi signal until I go to a new location. Or in the second case, it's 5 AM local time. So, if you turn it off when you're asleep and by the time it's morning, the Wi-Fi is turned on by itself. Or when the device has restarted. And to completely switch off the Wi-Fi, you have to go to the settings and turn the Wi-Fi on and all of that. So, and this is, again, there are different articles, how to fix the Wi-Fi problem, how to turn off the Wi-Fi, all of this. This is like, this was an actual problem. In fact, Road to Apple tweeted them and all of that. I didn't hear back from them. But for me, this was a problem. And basically, by doing this, there are three design principles which the iPhone pilots, these are user control and freedom, consistency and standards and visibility of status. But user control, it's not, basically, it switches on by itself. The user is not in control. And it is not completely switching off the Wi-Fi. Even if I want to switch it off by doing a swipe up, right, then consistency. This is not consistent. So, for me, the data packet and the Wi-Fi are ways to access the internet. So, by following a different convention for switching on and switching off the data packet and the Wi-Fi, it's basically violating the rule of consistency here. And a lot of people, I mean, in terms of visibility of system status, right, I mean, a lot of people, a lot of users didn't know why their phone was automatically switching, so Wi-Fi was switching on. I mean, as seen from the Google results, right, a lot of people didn't know. And I was like, how could Apple do this, right? Apple is supposed to be the gold standard of user experience and they were the guys who actually coined the term user experience. I mean, Don Norman was working there and he wanted to, he set up a team and he was the first person who with the title of user experience designer or user experience architect. So, this was the company which was, I mean, known for its user experience and how could they even think of putting their users. For me, this was like a genuine problem. I'm very curious to figure out why are they doing this, right? So, it was then that I came across all this search about choice architecture and what they might have thought. Basically, choice architecture is the design of environments in which people decide. It's basically a decision making thing and it's people decide based on the choices available. And basically, choice architecture, the commonalities are choice architecture, information architecture, interaction design. It's all about presenting information to consumers or users, right? So, a choice architecture has a responsibility for organizing the context and furnishing the choices using which people nudge towards making favorable decisions. Okay, information architect is helping users find information with the primary goal of completing tasks and interaction designer designs for user experiences, the interactions. So, these are very loose definitions which I thought would serve the purpose of this presentation. So, I put it here. I don't want to get caught in debates on defining all of this. But basically, for the purpose of this presentation, I have this. And the bottom line is information architects, interaction designers, user experience designers. We are all choice architects in a way because we are presenting choices for the user or the consumer to be able to select among the choices available and make decisions and get information basically. So, there are different environments, right? Environments designed to present choices. So, I mean, you go to a restaurant, there's a buffet there, there are different choices. You walk into the supermarket, there's so many choices. I mean, consumers go in and are able to pick what they want. You go into Amazon or Flipkart, you have so many choices there. You have different ATMs, how ATMs present choices are different. People design software applications, you have different choices there. You go into like a social media setting, Facebook, Twitter, you have again choices there. You pick up a newspaper and the news you see is based on choices which somebody has made for you to see. I mean, even a TV news channel, right? What kind of news is shown to you? It's again, a set of choices and which can lead to further debate. That's a different whole discussion, that's a separate discussion to happen. I don't get in there right now. Let's talk about Nudge here. People's decisions can be influenced by small changes in the way how choices are presented to them. Again, these kinds of choices which enable people to make favourable decisions as decided by the choice architect are called Nudges. People can be nudged into making a decision as decided by the choice architect. Nudge is, I'll go into examples. For example, in your menu, if you decide to have lots of healthy food options, lots of fruits, lots of veggies, this is a nudge. It nudges you to basically eat healthy. Totally banning junk food is not a nudge because you are completely eliminating all those choices. That's not a nudge. Putting lots of healthy food in a menu is a nudge. Similarly, in the environment, these environments are ripe to have these nudges. For example, in a buffet, you can go to a Chinese buffet or you can go to a buffet which has lots of healthy food items. You go to that healthy food buffet, at least in the US, you had lots of healthy items such as lot of veggies, like soups and salads and all of that. You can have nudges in these environments. In a supermarket, how do people present choices? Maybe they put in the most expensive items like a friend, they'll hide the less expensive items back. Even in an e-commerce site, you go to an e-commerce site. The first thing you see in that top section is all these expensive phones there. At least the HTC Bank ATM which I use regularly, it's narrowed down the choices. I can just quickly get in and quickly get out taking out the specific amount of money which I want. It has listed a set of favorite items in the list of choices which present to me and I can get out as quickly as possible. A software application as well. You can have several nudges there, social media setting. I'll go into examples of these settings. Even in a newspaper and a TV news channel, you can have lots of nudges to influence basically how people think about maybe this nation, maybe a political party or things like that. There are a couple of misconceptions about nudges. It's possible to avoid influencing people's choices. In a setting where you have lots of choices, it's almost impossible to avoid influencing people's choices. You have to present information in a certain way. You will be biased in some way or the other. Maybe you are biased based on the revenue you want to earn or if you want to have the greatest of experiences or putting in most number of choices, all of that. The second misconception is a nudge always involves coercion. Coercion is a negative term. It's like forcing the user to do something. So that's a misconception. You don't have to have coercion in a nudge. Another misconception is nudges reduce ability to solve problems. So you're continuously nudging people to choose the right thing or decide by themselves. So nudges, I mean, they're not necessarily, they don't necessarily reduce the ability to solve problems. So there are two extremes here. One is called libertarianism. I mean, it's a hard term. So a critical responsibility of choice architects like this is to protect the people's right to choose. That's one critical responsibility there. And this is known as libertarianism. So in an ideal world, people have all the choices they want. They are not biased choices. They can make the perfect choice without being influenced by anything or anybody. So that's a libertarian concept. The other extreme of this is called paternalism. So it's an action limiting a person or group's liberty or autonomy with the intent of their overall good. So again, the important thing here is the intent is their overall good. Example, I mean, you have laws against seatbelt. I mean, if you don't wear a seatbelt, you'll be fine. If you don't wear helmets, you'll be fine. And I mean, the iPhone Wi-Fi setting, right? So, I mean, the Apple thing is, okay, maybe this is good to people in some way. So that's, I mean, maybe in a different country, it works out because the data charges are very high, maybe in the US. And that's why they have this option of wherever they go, the first option would be for the Wi-Fi thing. So maybe they thought that this would work universally as well, which probably doesn't. So that's what I have taken out from that. So interaction designer and information architect, we all walk this thin edge when designing applications, whether we guide users down the right path, or we have to force the user down this path. So sometimes it so happens, right? If the user doesn't feel that he's in control, right? So users have to feel that they have to be in control and they have freedom to do what they want to do. So if the user is not aware of what the system is up to, what should happen if he initiates action and that he's in control. If the user is not able to do all that, then he'll be pissed at whatever we have designed. So basically that's what I was trying to say. I mean, the expected user behavior is against or regardless of the will of an individual. So if the user is not willing to do it in certain ways, it's against his will and all of that. So it's important to note that no matter how much if you are in an environment where there are design choices, there is no such thing as a neutral design because you have to present choices in a certain manner, regardless of it. So small and insignificant details can have major impacts on human behavior. So this is the whole theory behind it. And what is the need? What is the need actually for? Why do we have to architect choices? Why do we have to nudge people? So this is some statistics. 13% of the world, people in the world are obese. 39% of the world's adults in the world are overweight. One in five children and adults globally are overweight. So if you're saying that people are able to make choices by themselves, right? And this eating this food and this quantity is bad, then there wouldn't be a, then in an ideal world, there won't be any adults who are overweight. People won't pick up bad habits like smoking and people, I mean, though marriage rates are declining, I mean, you have a lot of living relationships and all of that. But even then, right, among people who actually get married, 39% of marriages, I mean, we have stats for the US, the end in divorce. And another... Sorry, we have four minutes. Four minutes, we started a little late, but yeah. Anyway, so share of children, I mean, this is, again, another example. Why? This is, again, another example to show that people make bad choices in terms of people who are born outside of marriage. Again, one reason why we need this choice architect, actually, is to avoid this biases. So there are plenty of biases in the world. And it's a deviation of the norm and people don't make rational judgments. So that's, I mean, the problems which the bias help us to address. I mean, bias are there for a reason. And these are the problems with the biases help us to address. So some of the examples of nudges we see in everyday life are this, I mean, if you have a car or a bike, you see this very often. There's an engine check, there's a low fuel check. Even you have a phone, you have, if there's low battery, it shows up there, that's a nudge. You have all your notifications saying that, okay, I mean, I wouldn't check my Gmail unless I have a notification that, okay, there are messages there, or I don't check my WhatsApp unless it shows that there is, it's a nudge to help me start looking at that. Gmail has a nudge. I mean, you would have got this if you are sending something and there are no files attached. Do you want to send it anyway? Notifications, I mean, there are some subtle notifications, kind of notifications as well. So if in the pages it says nine new messages or in a group, there are like two new messages or there are so many comments on a message or there are so many people who have reacted to this. I mean, it's a nudge, it's a subtle nudge to see there. And when you are trying to shop for a plan, right? I mean, there's a most popular target as it's nudge for people to actually pick that plan. This is one very common thing which a lot of people, including my wife, my wife is a PhD, she falls for this. So the anchoring bias uses a higher price to anchor. Anchor, it says, okay, this is so much, you're saving so much. And this is what the price is. I mean, people will go for this. This is a lot of people will go for this. And there's this thing called the decoy effect. Decoy effect can be used to, it basically says that you will compare the item which you choose to less attractive options. So you're getting a 12 GB RAM and 256 GB storage for 29. When you have the same 80, 12 GB RAM to 256 GB storage for 45. So this is so much better option than that. So there are similar ways it'll compare. And this is in Amazon. Amazon actually doesn't allow us to compare phones which we pick. It just shows whatever it wants to. But I mean, Flipkart does allow us to compare. And I mean, Amazon is definitely taking advantage of the decoy effect there. Another same thing, the decoy effect is the not 5G has two options, right? And the price difference is hardly anything. It's like 2000. For the 12 GB RAM, it's like if you paid 2000 extra, you're getting a great value for the 12 GB RAM. So again, it's a decoy effect. The sales nudge. I mean, do you remember all your like clothes jobs, kids clothes job, the food court is all on one floor. And the nudge here is you are going there to decide where to buy from. Rather than from to buy or not. Similarly, you go to a location. There are lots of other eateries available. And again, it's to buy or not. It's where to buy from, which evil to choose. Similarly, you have all these festivals, right? They come together. All these festivals sales come together. And again, the choice is you're not buying or not. That's not what's in your mind, but you decide where to buy from, which is another choice. So if you want the best players to reach the finals, you have the seeding system. And sometimes in tournaments, world ranking is not equal to seeding. You get the, it is designed, the whole system is designed so that the best players come there. In the context of time, I'll just skip this. You have affordances as nudges as well. And there are false affordances and hidden affordances. And there's default settings as nudges, which are very powerful. The example which I wanted to give was how many people have actually changed all the default settings on the phone? So not many. I mean, I haven't changed all the default settings on the phone. Maybe for the ringtone, that's it. And even the organ donation part. So if you are given option to opt in, maybe 30% of the population would opt in to donate organs. If you are already automatically enrolled and you have to actually opt out, only about 10 to 15 people bother to opt out. So you have about 80 to 85% of the people who actually are willing to donate organs there. And the worst, worst kind of default setting which I've come across, it is even worse than the iPhone settings is for the WhatsApp auto download. So I have come across multiple locations where my parents' phones are like actually hanging because all the videos and photos, they're just downloading and it's the phone hangs. I mean, it's like the first kind of default setting. So basically the key takeaways are while presenting choices, it is extremely difficult to avoid influencing people's choices. Humans are biased and terrible at making choices and therefore nudges are there. They're needed to overcome flaws in decision making and cognitive bias. Default options act as powerful nudges and people's decisions can be influenced by small changes in the context of how choices are presented to them. So that's, and for further reading, this is an amazing book. I recommend, highly recommend that you pick this up. Thank you. Thank you, Janet, for an amazing talk. I apologize for you had to skip a few slides there, but definitely it was a very interesting talk. In the interest of time, we'll take only two questions for now and definitely one of the questions you can stop sharing now. Okay. Cool. So in the interest of time, we'll take about two questions and this one is coming from Sovik and it is definitely an interesting one. So he asks, is a libertarian and a paternalist likely to come up with different information architectures or understanding of information? Yeah, probably. So if, again, it depends on the individual, but based on what they're intending to do, right? What the intent is, whether they want to give all the choices they want or they want to reduce the set of choices to be to make a person to come to a decision, which is beneficial to him. I mean, they'll probably, I mean, I strongly think that they would take a different path. Cool. So the second part of this question actually talks about as a designer of these technologies and these platforms, what is something we can deliberately do so that our work comes off as unbiased? It's very difficult. I mean, to come across as unbiased, I mean, if you have the right intent and all of that, yes. But I mean, if your intent is to benefit the audience the most, but I mean, if the organization you work for, example, you work at Amazon and you have to present choices to the consumer to maximize profit. I mean, there's only so much you can do, right? So that's where it stands. Yeah, definitely. So interestingly, I recently started off learning about this topic called behavioral economics. Yeah, this is completely connected to all of that. This is an overall umbrella of behavioral economics and underneath this, in fact, the book which I mentioned, the authors have won a Nobel Prize in that field. So this is like a whole jungle in itself. So let me, I asked a question from the technology designer side of it. Now let's move to the consumer side of it. As consumers of technology are impacting the physical world, we come across these nudges that beat the decoy effect, beat the higher markup on prices, you know, on products, which is like, it talks about the anchoring bias that it tries to induce with that. So what do you think we can do as consumers now, you know, to just be aware of the irrational decisions we make while we come across these biases? I mean, as you mentioned, it's awareness. I mean, if you are aware that, okay, this is how the choices are being presented to you and you know that what is the science behind all of this, right? What is science? What are thought behind all the choices which are presented to you? Then you can decide appropriately. I really wanted to buy the OnePlus Nord 5G. I mean, I have a, my iPhone is like seven years old now and I mean, it's still going great, but I still really wanted to buy the iPhone Nord 5G. But when I look at, I mean, the choices which are presented to me by Amazon, right? Comparing the iPhone Nord 5G, 12G, 12GB RAM with the 8GB RAM and the price difference. And I'm aware what they're trying to do. They're presenting this decoy to me and even the, so because of the awareness, I'm thinking, should I really buy this or is there any other better option in the market? And I'm used to the iPhone now. Should I just go ahead and buy the iPhone 12 or iPhone SE 2021 whenever it comes? So I'm aware of what is happening and why they're doing it. And that helps in a lot of ways. Perfect. All right. That was an amazing talk. I'm sorry. We could not take more questions. Definitely. All the attendees, if you have any questions, I think you can reach out to Janad. You will include the profile link in the follow-up emails. And thank you so much, Janad. It was definitely an enlightening talk. Thank you. Thank you, Manish, for the opportunity and Sovik, for the opportunity. Yeah. Let me ask Sovik to get started with the next speaker. I hope I'm audible. So thanks, Janad. And thanks, Manish, for doing a wonderful session on choice architecture with us. We were a little short on time, but I'd like to announce that if you have any follow-up questions for Janad, please post them on Hasgeek website. There is the updates channel where you can discuss that and post it. Also, I take this opportunity to remind everyone that at the closing session, which is scheduled for 1.30 Indian time, we'll be announcing a quiz link, which will enable you to be one of the four different prize winners. We'll announce the quiz link at 1.30. So please be around for that as well. So coming to the next speaker that we have. Our next speaker is Anshul Tiwari. And he's going to talk about using information architecture to garden better. This is just like the first talk where food and information architecture, you don't relate it. It's very unlikely that you will relate gardening and information architecture to introduce Anshul. Anshul Tiwari is a social entrepreneur and the founder of Youth Ki Awas. He has worked across media and citizen advocacy. I had to truncate his introduction a lot because there are a lot of things to his credit. Among them are, he's a Ashoka fellow for social entrepreneurship, was named the young innovator by the United Nations ITO in 2012. And he's also a Forbes 30 under 30 media influencer in 2018. Now with the feeling of being a complete underachiever myself, after hearing that introduction, I hand over the stage to Anshul. Thank you so much, Sovik. That was a very kind introduction and I am delighted to be here. Thank you for having me. It's a very interesting topic. Honestly speaking, I'm a bit nervous because I'm talking for the first time publicly about my passion for plants and gardening. So big, big thanks also to Burman for recommending me for this. So, you know, my background actually like Sovik mentioned is very different. I started and I run Youth Ki Awas, which is, you know, completely crowdsourced media platform that gives space to lots of young people across the country to speak up on issues that they are passionate about. And I've been doing this for the last 13 years. So I started this when I was 17. I was still in school at that time. And I think one of the things that I did not anticipate was how the platform would develop. And I think my first brush with information architecture or rather the lack of it came in when I started realizing that there are just so many people who are writing on the Youth Ki Awas platform or sending in their content or sending in their stories, sending videos, photos, but I did not have a very organized structure of managing their information, where they come from, what they are passionate about, you know, what motivates them to talk about something, what really gets them going when it comes to social justice issues, politics and so on. And I think early on, closer to 2010, I started organizing information in the best possible way to really start understanding who these users are. And over the course of time, over the last 13 years today, I think just organizing information around people has become something that we do a lot. And that I have become extremely passionate about personally, whenever I find out about platforms that exists out there, I always want to know more about what is, you know, who are the people behind it, what's the information that exists out there, and so on. So when I first started gardening and, you know, gardening really here is more like a metaphor of sorts, it could be anything that you're passionate about. You can see my photo on your screen, I look extremely happy, you know, in my garden, so I'm super passionate about it. I, you know, I have grown a deep sense of curiosity as well as a passion for a particular kind of a plant. And this plant grows primarily in South Africa. It's called a hewarthia, broadly classified as a succulent. But what's beautiful about this plant is that you can create, it's probably the only plant in the world that lets you create endless hybrids of it. So you can literally hybridize, you know, n number of genuses and species that exist out there. And you can create your own hewarthia, you can have your own plant that is very unique to you, that does not exist in the world that you have invented pretty much. And that's something that pulled me right into it, because when I started collecting this plant, I started seeing people and, you know, master growers from Japan, from all of Europe, doing this. And, you know, as is the case with me, I went right into it. And as you can see in the photo, I have a massive collection, more than 2000 of these plants in my house. And I've been growing them for a very, very long time now. It's been more than, I mean, very long as in it's been more than three years for a person to focus on one specific, you know, plant. So I've really gone very deep into it and I've been able to gather the plants from across the world. So from everywhere from South Africa, where the plant originates from to Japan, Thailand, Portugal, and so on and so forth, getting licenses, getting certificates to really bring the plant in, making sure I have all the approvals and so on. So it's very safe to say that I went very deep into the hobby. And one of the things that I started doing about a year and a half ago was I started trying to create my own hybrids. And, you know, the process of hybridizing a plant is very simple. Plants put out flowers. Flowers have pollen, you take pollen and you put it in another flower. It leads to something called a seed pod or a fruit. And that typically has seeds. You sow them when they're dry and you get a new plant. It's a very simple process for somebody who's really into gardening. And if you're not into gardening, it may sound complicated, but when you get right into it, it's actually pretty simple. So I started really going quite deep into it and I started figuring out that I do want to create a hybrid of my own and not just one, I want to create several hybrids that I can say that, okay, I invented this plant in my house, in my garden. So as I started going deeper into it and started learning a lot more about this particular plant, I realized that there are three very critical pieces of information that exist across this particular plant and actually any plant if you're gardening. One is the genus of the plant, which is the, you know, it's like the top level family of the plant that it comes from. So within the particular plant that I grow, which is called a havertia, you know, earlier about more than two decades ago, there used to be a massive classification called havertia. And havertia is the genus that we're talking about over here. Over the course of the last many years that genus was then further broken down and, you know, researchers and botanists they realized that different havertias have different kind of features that they put out. So the genus actually got broken down. So there came a havertia, there came a havertiopsis, a tulista and blah, blah, blah, like many, many more genuses came about. Now within the genuses are the species. So in front of your screen, you'll see two very different kind of havertias. The one that's a bit more red is called a, it's actually now classified as a havertiopsis. It's called a havertiopsis coelmeniorum. And the one that's not so red is actually called a havertia wimmy. And that's the species of the plant that exists out there. And the third very important part of any single plant is the location that it comes from. So many plants are of course born in nature. And you can literally find out the location where it was, you know, sown or where it was grown, where it was growing in nature, etc. Many of them also come from the gardens of different hybridizers. They come from the gardens of different collectors. And this hobby by the way is vast. So when I got into it, my first 20 plants and I thought that I have the world. But then I started discovering people who had thousands of them. And just like I'll give you an example. There is a group on Facebook called the International Havertia Study Group, which has more than 15,000 people. And that's just one of about 30 such groups that I'm a part of, which has thousands of people growing this plant. So there's really a very solid community out there of people who are passionate about this one particular plant. So when I started hybridizing them, initially, I would just pick plants at random and I would try and cross-pollinate them and I would try and see if a seed pod would emerge. For almost a year, I was completely unsuccessful at it and I could not really wrap my head around what was going on. At that time, I also didn't care as much about the identification of the plant. I didn't care what species it was. I just looked at it, I liked it, it looked great, I got it. But over the course of time, as I kind of realized that if I really want to create hybrids, I need to understand the plant better. I started organizing the information a lot better. And I did something super simple. I just put together an Excel sheet. Like the one that you see in front of you right now. So it's very simply classified. There's the genus of the plant, there's the specie of the plant, there's the locality of the plant and there's a collection number. So I even went down to understanding whose collection it comes from and the number signifies when the plant was born, what level of parentage it has, etc. And if you Google some of these numbers, you'll get all the information because the people who have collected them, like the letters that you see, MBB, IB, they're actually names of collectors, they're initials of the names of the collectors. You'll find all the information that they've put out publicly about their collection and you'll know exactly how they're growing. So when I started putting this information together, I started understanding a lot better which of my plants come from similar locations, which means basically that if a plant is growing in nature at the same location, if two different plants are growing in nature at the same location, there's a very high chance that there is natural pollination occurring between the two of them, which gives me the indication that I can go ahead and do it as well. Similarly, I started realizing that if there are two plants that look similar, they're from the same species but are different in the way they look. So for example, if you see the list that I have in front of you, the second plant over there is called Havorthia Atenuata and the third is called Havorthia Atenuata Virradula. They're actually two very different looking plants but they come from the same family. So that started giving me the indication that, okay, this is actually related, these two plants are related, I can actually start hybridizing them as well. And slowly and steadily as I went deeper into it, I actually was able to very successfully create hundreds of hybrids all of a sudden. I was getting a tremendous amount of seed pods on my plants, I was able to grow my garden from I think about 1500 at that time and suddenly added 500 more to my collection. As of right now, I have close to 2000 like I said and I have almost probably more than 800 seeds that are waiting to be sown so my garden is growing at a tremendous speed. And the place where this helps is not just to help me make the decision about what I want to hybridize, it also helps me strengthen my collection. And this applies not just to plants, it applies to everything, it applies to art, it applies to any hobby that you're passionate about, any collection that you're trying to build, it applies to philately, it applies to collecting coins for example. If you're able to structurally organize information around the things that you're passionate about, it helps you also find the gaps, it helps you find out what you're missing in your collection and where you could possibly get it from. So a lot of the information when I started organizing it, it started helping me understand in a very structured way what was the next plant I wanted. So I stopped making random purchases because the one thing about collections is that it can really burn a hole in your pocket and you don't even realize when it does that. So I think the moment you start organizing the information around your passion or what your hobby is, you start realizing what you want to say no to and what you want more of. And I think that's the best thing that organizing information around this particular plant helped me. And as you can see in this photo, I have a crazy collection and this plant that you're seeing in front of your screen right now is actually a hybrid of the two plants I showed you a little while back. And this hybrid was created in my garden, so I was able to successfully create this hybrid as well. And I just want to show you my first hybrid that I created, which I call the Orion, which actually doesn't exist in nature. It's a beautiful hybrid between two different species that I put together. And now I'm slowly and steadily getting into intergeneric hybrid. So not just hybrids within the Havaltia family but related families as well. And that's what, honestly, organizing information can really do for the things that you're curious about, the things that you're passionate about. It can help you build a stronger sensibility around them. It can help you find out more and really go deeper into your hobby and not just have it as a side thing. It can actually become something that becomes a very integral part of your identity, just how this passion is becoming for me. So with that, I'll kind of wrap up and I really hope that this information was helpful for all of you and that you'll be able to apply some of these things to your passions and the things that you're really kind of going after. Like I said, it's not just applicable for plants, it's applicable for anything that you're passionate about and you're trying to build a collection around it in particular. So thank you so much for having me and, Sovak, if there are any questions, I'll be more than happy to answer. Thanks a lot, Anshul. That was a fantastic short talk and definitely very inspiring. I think this talk should have been titled, Using IA to Save Money, because no one would have thought it that way. And also you started off with a very nice introduction to IA in the sense that you don't really realize when IA is good but when you start off with a mess or there is a lack of IA, that is more evident in a way. So which is the part you started with the youth Kiyavas bit? So could you talk a little bit more about that aspect as in how IA is invisible until it is failing? Yeah, absolutely. In fact, not just at youth Kiyavas, I think not just in terms of the platform, I think even in terms of how you manage your team, how you organize the key skills that your team has, how you're thinking of building sales processes, how you're thinking of building potential funding opportunities. I think there is missing information everywhere, which honestly leads to a mess for sure. And it's only when that mess comes in front of you is when you realize that this was right in front of us all this while and we should have just thought of organizing this better. And it's all something that honestly I have gone through, not just in terms of the platform, like I said in terms of the team, in terms of funding opportunities, revenue streams, etc. Even with my hobby for plants, the fact that I could have used IA to organize them better did not come naturally. Like I said, it came at a point where I was not being able to create hybrids and then I kind of started realizing that maybe I should just start understanding the information around these plants a bit better. And I think it's okay because sometimes just organizing information sounds and seems a bit tiring and it seems a bit very process oriented and people don't like processes, like everybody hates structures and processes. So I feel you start off not appreciating it, but you end up appreciating it a lot when it becomes very simple, very well organized and something that actually helps you move much faster. So it's only the initial moments when it's a bit tiring, but the long term benefits of organizing information properly are terrific. Right, absolutely. At this point, I would like to just remind our audience if you're on Zoom, you can use the Q&A tab to give us questions for which I'll take it to Anshul. And if you're on YouTube or any other live stream channel, then please use the live chat tools that are there, put in your questions and they'll be forwarded to me over here. So coming back to you Anshul, IA as we understand is a subset of UX. In a way, it gives you a better user experience and enables you to do things. Would you call yourself someone who is from a UX background? It's very tricky, but I like to believe that I am because I really love and enjoy understanding user experiences personally. And do you think people need special powers to be able to do IA? Not at all. I think it's just being able to put logical information together and I think you just need practice and understanding of the kind of sets of information that you have. Right, absolutely. Okay, so it is fair to say that without having organized your entire information, you wouldn't have been able to achieve all the things that you've achieved. Absolutely. But at any stage, or rather, let me say at this stage, what are the next things because everyone gets an idea, like for example, the Ganesh Bagless talk when I heard it. So you're almost presenting an information in a way in which you have understood it. But then the fact that there's a potential for many more things to be discovered beyond that is there and that is something that always skips our mind. So wherever you are today, what is next? What are the next thing that you might be able to discover? So as far as this particular hobby or passion that I have is concerned, I think my next step is to actually make a lot of this information public because there are many other collectors out there who are trying this but are not being able to be successful at it. And I think there are groups of people who are already organizing a lot of information around this and I definitely want to very actively contribute to the creation of that process. So I see that as a next step because also it helps build a stronger community. It helps bring more people into it. It helps introduce the hobby to a lot more people as well. Right. So there's this question from Thompson who asks, what role has online communities played in developing your hobby and do you have some thoughts about IA being a collaborative exercise? Yeah, absolutely. Actually, the communities that already existed around this particular plant played a very, very big role in helping me develop this hobby because when I started getting introduced to a lot other growers in other parts of the world, not only did I get access to information that I had absolutely no clue about but also I got free plants. I mean, they love the fact that I was trying to do something to this tune and they love exchanging plants and I got a whole bunch of them. And now I'm very closely knit as a member of many of these communities as well. And I do think IA is a very collaborative exercise. There are several of these Excel sheets that I have with growers from Portugal, from Canada, from Japan who are just constantly adding information. Very recently, there was one particular species of the plant that I talked about, which got completely eradicated from nature because there was a highway that was constructed where it actually grows. But a lot of collectors from across the world were able to come together and put together information around the plant because they had it and they were able to collect it from different growers, etc. So I think it definitely is a very collaborative exercise for sure. Okay, another question I'm taking, Ajay from YouTube has asked, were you able to use information mapping techniques or IA around watering and caring for these plants because as plants from different region would need different care? That's a great question. Honestly, I haven't gone as deep yet. I very perceptively just decided the care and the watering schedule. Honestly, because they also come from the same genus, 90% of their care is the same. It's only 10% that's different because they come from a different location, their flowering cycle is different, etc. But that doesn't make a huge difference to me. But actually now that you say it, I do think that there is definitely some applicability of information architecture around how you care for a lot of these plants. Right, so on that note, you already have shown us some of the Excel sheet. Are there any aspect of categorizing, classifying or anything that probably you have done and it's unique to you and the other collectors are not doing it that way? Because information can be seen differently by different people. So the Excel sheet that I showed you is a very standard way of organizing information because that's how most collectors around the world do it. But of course, when I create hybrids, this is not a hybridization sheet that I showed you. The information that I save over there is very different. It looks very different. I am able to put down a very deep level of parentage of the plant and that's something that I actually saw only one grower based in France too. And he was very kind enough to share how he organizes that information and I was able to really kind of learn a lot from him. But yeah, I think the information around hybridization is probably very differently stored with me as opposed to a lot of people who are growing the plant across the world. Got it. And another question from Misha, I think this is also on YouTube. Apart from hybridization, what other aspects of gardening or plant collection can we use? I think just the fact that you're kind of building a collection because it helps you see the gaps in terms of what you don't have and what you could or should have. And like you said, how to save money. I think that's a very critical aspect because now, I mean, if you've noticed during COVID, people just went berserk buying houseplants and spent a crazy amount of money on plants. And many of them also lost a lot of plants because the information just doesn't exist as easily. That's the challenge with the hobby like, you know, plant growing, you talk to one grower sitting in some part of the world, their conditions are very different from your conditions. And you know, you might think that okay, if I do it this way, it'll work for me as well, but it doesn't. So I feel it can really help you grow them better beyond just hybridizing the kind of plants that you have. Alright, there are two more minutes that we have with you and I have two questions at this point of time if there are no other questions. One is that after reaching where you have reached, do you still discover something fascinating when you put together information and can you tell us something fascinating that you have recently discovered? Absolutely. Actually, I can give you two examples, one from more of my work domain. One of the things that, you know, that we at Udk was have been trying to figure out is that we've been a writing space for more than a decade now and where do we go from here? What's the next step that we should take? Just looking at, you know, the very obvious information, you can tell that, you know what, people don't just want to write, people want to create videos, we don't want to share photos, etc, etc. But when you move beyond that and when you discover information that's a lot deeper than that, you realize that actually beyond writing and sharing information, people just want to talk to each other, people just want to connect with each other. And that's a piece of information that I discovered just by chance while going through a lot of the surveys that we do with our audiences and a lot of the conversations that we have with our community. Similarly, you know, in my garden as well, I recently discovered that I might actually be able to create one of the rarest hybrids ever because two of the parent plants that are very difficult to get happen to be in my collection and they come from two different, completely different parts of the world. And I'm giving it a shot, you know, I'm trying it out. My final question and in under a minute, how do you create a hybrid? Do you just lock two plants together in a room? Yeah, and you put on some candles and that's it. No, you I mean, honestly, if you really want to know, you take the pollen of a plant and you put it in another, you know, like you just pick it out like you play B. So I use a I use a very fine hairbrush. I dip it in the flower of a plant and then dip it in another flower of another plant. So that's that's how this this is definitely fascinating. Okay, thanks. Thanks a lot, Anshul, that will be the closure for this session. And if anyone of you have more questions for Anshul, then please put in your questions on the Hasgeek event page on the in the discussion section and we'll try to have Anshul answer those questions for you. Thanks a lot, Anshul for answering the questions and such a great story. Thank you so much. All right, our next speaker is going to be Gurman and it's it's 1245 time for the next speaker to come in so I'll be directly jump to that talk. Now, Gurman is someone I've spoken to before in some other forum, and she is a data and a graphic journalist and at Reuters in Singapore so she'll be joining us from Singapore at this point of time. Although I'm, I'm told that she is from Delhi and that's where and that's great to have the Delhi Association always because we are, we are, this is the first time by the way we are doing World Eye Day online, completely online. Before this we would always do World Eye Day at a physical location in Delhi. So, Gurman, while she works at Reuters in Singapore she tells stories with code. This is a combination of data extraction, analysis and forming compelling narratives around data. Gurman is a self taught coder and a designer, and holds a master's degree in journalism from Columbia University. With that, I would invite Gurman to please take the stage share your presentation. Hello everyone. I'm going to run here me okay. Yes, we can hear you fine. I'm Gurman and I today I'm going to be talking about visual narratives to feed the curiosity. If I'm going to talk about architecture. Of course this talk has an architecture and a structure. Obviously. First I will go in on say, hey, hi, I'm Gurman. I am a graphics and data journalist at Reuters. I'm originally from Delhi, as Sovik mentioned. I worked in local newsrooms in the US. I was also one of the key members of the first data team in the stunt times. I have two degrees in journalism, none in design or computer science. And I didn't know anything about information architecture or even what the term meant until Sovik called me. So, that's me. In terms of what do I do. As he mentioned, my work is a combination of journalism code and design. If you're wondering what in the world does that mean it means I take things like this such as a sentence in a press release and write some JavaScript code and turn it into a line of practice. So those are the kind of things that I do. Oh, yes. So yeah, a line of practice. So that's the work I do. So where does information architecture come in all of this. I googled you know because I don't know anything about information architecture and I got this definition that says it focuses on organizing and structuring content and effective and sustainable way. So organizing and structuring I think were two words that were familiar to me. And you know what else has structure stories stories have structure too. And today I am going to be talking about structuring stories and structuring experiences when you're telling stories. And thing is that journalists have like all these formulas or frameworks to structure these narratives. And we have these fun names for them you know we have an inverted pyramid we have a martini glass we have an hour glass we have the kebab we have a diamond. We are very creative and all so we come up with these creative and fun names for otherwise stories about depressing things like blasts and accidents and things like that. So the most typical form of stories that we would see a news report would be the inverted pyramid. What does the inverted pyramid look like you bring out the most important information in the top and as you go down the importance of the information decreases. So the idea is if I walk into a story even if I read the first paragraph I will know what happened. And then if I want more details on a dig deeper I keep going deep and deep and deep and deep. I can leave at any point in time and I would still know something I will not walk away with the without knowing what exactly happened. Here's an example of a visual story like that. This is from my time at HD. If you walk into the story there's a headline that tells you something that has happened. And then with the tease there's a graphic that just shows that that margins have been decreasing so elections have become less competitive and sorry, more competitive. And then you say that in a paragraph you explain what the story is right at the top. And then I tell you more bits and pieces about it. I actually show you how the median has shifted lower how the breakdown how the spread and the distribution has changed. And then I tell I talk to some, you know, candidates who have one with low margins and get their quotes and pick out more instances about it. I talk about how it's different by party. And towards the end I maybe pick out a few consequences that have very low margins consistently and say why that might be happening. So like, it's a very straightforward structure even if the person left at the first scroll they would know what the story is about. And as they keep going deeper or the information, the depth of the information just keeps going forward. And now the story structure is a martini glass, but unfortunately all I have is water with me right now. The martini glass structure. You is kind of like the inverted pyramid, but then after that to give a good chronological order of events, and you have a strong ending. And this is great for like events and like things like accident or a bomb blast or a riot because there is a timeline that is important in the midst of all of this. So here's a story that we did after the tractor rally on Republic Day this year. And you walk into the page, we talk about, we're going to talk about you're on a tractor, you're in the rally. And then we tell you, you know, this, this happened, there were clashes between the police and farmers, they're protesting because of this. And while on one hand you had the Republic Day parade on the other, you had this tractor rally. So we just base out the premise of what happened. And then I start giving you the chronology of events. I tell you that this is what the roots were, this is where people started. And then people broke barricades and went away from those roots. And then they were clashes between the police and the protesters. And exactly how did people get to the class site. And once they were on the class site, it was this whole thing about flags. So we talk about the flags at the class site. And after all of this towards the end, I come back to my tractor line, because my tractor line is a strong ending to this otherwise. Otherwise, otherwise very chronological story, and you leave with an impact. That's the idea. So that's the martini glass. Now, another, you know, bit is whether do you explore or do you explain how do you interact with the reader, and how do you let the interactivity flow. So that he spilled the martini glass. So if I spilled the martini glass. I'm talking more in terms of the usability of the page now. And we start with an intro, and then I walk the reader through an experience, and then towards the end, I let the reader explore for themselves. So this build martini glass. If I this is an example that some of my colleagues did about how herd immunity works. So we, this is this whole explainer we have the simulations running that show you that according to different vaccinated populations, how does the spread occur. And you're looking at that we compare it. We add a layer of social distancing that if people are not moving, and they are not vaccinated what happens, as opposed to when they are moving. We add a layer of vaccination to that and say that you know, ideally people don't move while a certain population is infected vaccinated how effective that is in stopping the spread. And after explaining all these mechanics and things like that, we let the reader play, and you know change these parameters we let them change what percentage of the population is vaccinated we let them change the are not we let them change how effective the spread is. And how those parameters reflect in the spread. So we have explained the first and then towards the end, we are letting the reader explore. Another example of that is the story I did about government schools in Delhi, several, several years ago. And if in the story, I tell you something about government schools in Delhi, I tell you we have these many schools, these many like what was like the size of these schools. Because they are in more populated schools and more popular neighborhoods. And then, but only few have class 1910, and how many people get from class nine to 10. And then how many secondary schools are there in the, it's like in the number of schools keep getting smaller, as you keep going down. And then only these many schools offer the science stream and see how your choices are limited. And if you're a girl it's even harder to go to a school that offers science. And then I pick out one girl if she lives in Sangam Vihar, how easy would her to be to access a school that where she can study science and the secondary level. And after explaining all of that, I let the reader pick these choices and see what their choices are like. So if a boy is studying commerce, as opposed to a girl is studying commerce, or if it's within science, it's a girl or a boy, what which schools are left you see which pockets remain devoid of choices which pockets have choices. So it's towards the very end I let the reader explore. And after that I have explained the problem. Then the next story structure is the diamond. And the diamond starts with like an anecdotal example, then you tell them something about what the story is about to give a lot of detail and background and then you end with an anecdote again. So, this is the story I did about Bollywood music. I am a big fan of Bollywood songs. I think one of my very hidden talents is my entire skills. So this story was something I was fascinated by I was watching an interview of Anupama Chopra with some singers where they mentioned that you know how every other song is now sung by Arijit Singh. Whereas earlier you would look at music cover and would all be Lata Mangeshkar and Asha Bhusli and it was interesting that you know here's a feel that was actually dominated by women and now is nominated by men which is like playback singing when it comes to solo songs. So for that I start with an anecdote I pick out 2017 and I show what the split is like with the songs that are sung just by women and songs that are sung just by men. Then I tell you what is happening. I tell you what the story is about this was not always the case that this is a shift that has happened. And then I show that in a chart. I show that shift in a chart and then I get into more details about it you know when exactly the shift happened how it was because more and more singers entered the market who ruled the market when it's all like background information. And then I ask the question that what led to this explosion of songs that were sung only by men. And I do some old school reporting where I talk to some experts and some music directors and figure out what shifted and they gave the series in such as how the lip sync faded in movies how the story line shifted and that changed what songs ended up in the movie. So you contextualize your findings with stuff so again more background information. And towards the end I compare again 2016 and 2017 I pick out select movies and show talk about you know again I am with an anecdote. And there's also this exploratory bit where you can look at different decades you can pick out movies and see what the gender split is like. But if you're talking about exploration, the assumption might be that dashboards are just exploratory and interactive. I don't think so that may not be the case that was not all about clicking buttons and picking out with charts and things like that dashboards are also about telling stories. And that I want to talk about our coronavirus dashboards. So if you come to the Reuters coronavirus dashboard, you don't see global figures at first you see three random headlines. These are generated by algorithms that recognize what's interesting happening where where there's an uptake where there's a downtake who's reporting highest numbers in XYZ day where there are big rises. So things like that. And we just give you plain English and tell you what's happening. There's no fancy charts is nothing that's just a sentence telling you something interesting. And then we come down you know here's like these countries have the highest cases these countries have the highest deaths. Then because vaccination is the most important thing right now everybody's concerned about we tell who's vaccinating the past is with respect to their populations. Then we have this metric where we talk about the peak instead of comparing absolute cases we wanted to show where different countries might be with respect to their own pandemic. India for example, if you right now the cases that you're recording is only 15% of what they were when it was the highest in the entire thing. So, you know, it's like slightly interactive and all of that but at the same time, we are explaining what it means in pure English. We are telling you why this matters we are telling you what this means why this is a better metric than comparing absolute cases and things like that. And then we, you know, show our trends by region but even here before a chart we are using words to tell you something about it that how many case how many of them are coming from Asia, if all of every 100 cases. So using a combination of words and visuals ends up being a very nice pairing in absolute in like absolute independent terms they don't act that they aren't that effective but when paired together, they can be extremely powerful. And towards the end you know we let you go to whatever country you want to explore and see if you're interested in any stats. So, springs me to the point of where to use words you want words to complement the visual elements and enhance the picture without describing what the reader can see for themselves. So, I think that's the distinction of bringing in words to add a layer to the visuals. And then there's this other bit of when it comes to information, do I zoom in or do I zoom out when telling the story. In my experience, it needs to be a combination of the two. And that brings me to the eye glass. So you zoom out, you zoom in, and then you zoom out that's one story structure as well. So for that I have this example of story about ground water in India, I show you this map that gives you an overview of how bad it is in India and you see the North India red spot and how the decrease is happening there significantly. And then as you scroll in the story, I see the how I show you how different districts line up in this experience, which districts have good ground like ground water levels, as opposed to which are depleting their resources as sized by their population. And you end up seeing, you know, how many people live in what how many people live in areas that have decent ground water as opposed to how many different areas that have beyond their available resources. It's a guy about, of course, over exploiting ground water. And then after this, you zoom out again. Now from the district level, I go back to the state level. And then after this, I zoom out further from the state level, I go to the national level. And it helps you put things in perspective, the zooming in is adding up as a very like for someone who is from Bangalore or Ghaziabad to see where their hometown or district lines are, as opposed to also get the overview of what it actually means in the global context of what the state in India is like. So if I were to recap all that I spoke about these these are like four key decisions you're making in terms of architecting different structures of the different elements of your story, when it comes to telling these visual narratives. If I were to talk about how these decisions aid curiosity, we structure in a way to hook the reader in the beginning, or to leave an impact in the end. So that's how the order in, you know, act as a very effective way to peak curiosity. Then we explain to peak their curiosity and let the reader explore to feed it. And that works great with, you know, these elements, these kind of stories that people call explorable explanations. It's a whole genre of visual stories as well. And then you use words to complement the visual elements. And that also aids to a person's curiosity when they look at the chart. Oh, this is what it actually means. This is the context behind this chart. And you zoom out for the overview in context, but you zoom in to satisfy the curiosity. And that's kind of like how these different story forms can be used to tell stories in more like something that feeds the curious soul. And that is me and my talk ends here as per my map progress bar. So I'm up for any questions that people might have. Alright, Gurman, thanks a lot. That was a lovely presentation. Trust me, I'll never see a martini glass the same day again, especially if someone spills the drink in the bar. Also, I've never seen news stories being actually described in this way, the frameworks or anything like that. Because he didn't go to journalism school. Yeah, yeah, so this is, this is, I think this is a great way of saying that if you are someone who can bring to two or three different disciplines together that automatically opens up a whole new perspective in a way, a whole new way in which to see, see things. Now, I'll start off with a classic chicken and a question. What comes first, the framework or the story? The story, then you pick the framework that best suits the story. You won't know, you have to do your reporting first and then you figure out what is the narrative in which it might suit. Sometimes you just play around as well, you try one and you show it to your editors and they're like, you know what, what if you switch this around and then you go back and see, oh, maybe that works better. It's often hit in trial, but you first need to do the reporting to know which format might suit. Got it. At this point, I forgot to remind attendees, if you're on Zoom, please use the Q&A tab to ask your questions for Gurman. If you're on YouTube or any other live stream channel, then you can use the live chats or any chat option there available to put in your question there and it'll be forwarded to me over here. All right. I'll take a question that Manish asks, which is, it's quite relevant like in a newsroom settings where things move at crazy speed. Information is never enough. How do you choose one of these narratives with confidence? I think you just get it with practice, you get better with practice. The more you do it, the more intuitive it becomes. As a newbie, I was horrible at it. My stories would not have a narrative period. Forget about a type of narrative. It would just be facts, a collection of facts, but you get better at it. You editors help you figure out as well. And like sometimes, as I said, right, like for example, the chronological events, you know that when there are events that demand a chronology that would fit better. You know places where you want the reader to explore things and where the data allows for that and the data demands that so you pick those bits and pieces. Yeah, it's just that it gets more intuitive with time. And honestly, nobody like an editor won't say that, oh, let's do a diamond for the story. That's not how people talk. These are more like how academics have driven these frameworks. Like one or two of them might be things that I've modified, but most of them are things that researchers have looked at different articles and looked at trends that pop up, like the techniques that pop up and try to give them a framework. Yeah. Oh, got it. So you're saying that at this point of time, first you write the story, you produce it and all and then later on all the academics come and they say, oh, this is the structure should be called this. This is a machini glass. So you're saying that these classifications have not come from within the journalist community that have come from external observers or commentators. Yeah, and these might also be journalists that move to academia and study journalism there. So that way, but in journalism school, they will definitely teach you the inverted pyramid and things like that. Got it. So let me also jump to other questions that the audience have asked first. So said that single asks, are these storytelling frameworks relevant only for written or visual formats, or can these be applied to verbal communication as well. What are your thoughts on how those formats differ. So, some of the inverted pyramid is also very classic thing that you would see in say TV news TV is both like of course, the combination of your visual and audio thing. But if you go back to podcast and radio, even there you would see a lot of anecdotal leads and then going deeper and telling one story. Yeah, these story techniques apply everywhere. It's just a story telling technique. I would say that if you're doing a business presentation, you could use it there too. So it's just a method of how do you capture a person's attention? How do you satisfy the curiosity? And like, you know, how do you communicate what you're trying to communicate? It's more of a communication technique as opposed to a journalist technique, I would say. I think that's a that's a gateway to fame and it's not bound to any particular medium for that matter or any particular format for that matter. Okay. So someone who chooses to not name himself or herself says, has a question, any story that was the most difficult to visualize and how you dealt with it. So can you think of any such experience very hard to visualize? I think like, especially if you're talking about debt is very hard to visualize in general, because you want you're talking about a very serious tragedy, essentially, you're talking about the tragedy. And you want people to stay people when you're telling the story and not just become numbers. So tech always is a very hard thing to cover. In the case of COVID, we did this bit where like the deaths were at their peak in your in June, July, and people were dying everywhere. And to talk about that, we spoke about the pace of death and it's very sobering number that one person in the world was dying every 17 seconds or every 18 seconds. And when you put it in that way, it becomes much more powerful. And you know, of course, that these many people are dying every day and you're reporting these numbers, but to reduce in those seconds. I think that that was death always is very hard. And to think of ways that makes people remember that we are talking about actually a tragedy. That's quite hard. All right, so anything to do with any morbid story for that matter? Yeah, any morbid story for that matter, because you don't want a chart to dilute the tragedy. Dilute the big thing. So do you feel, and I think as an extension to this, do you feel visualization sometimes falls more in the entertainment space or rather than learning an information space, or at least is often mistaken for that? I think in terms of COVID has proven that entirely wrong. It's a story that's driven entirely by visualizations and data and where cases are picking up, who's managing the pandemic, who has lost control, where older people are dying, where it's younger populations of things are fine. So like those kinds of things, those are scientific questions and things that are have been crucial in managing this disaster that's struck the world. It's entirely driven by data and visualizations, the log charts, like people got comfortable reading log charts because of that. And I think that's saying something. So it's definitely not just for entertainment. COVID has definitely broken any of those beliefs if they existed. This year has been of too many line charts. Last year, I hope this year will not be of any line charts. Sorry, last year. I'm still there. I'm still in March 2020. Let's get you back to 2021 with the next question. So this is a question from Arjit who was asking, how do you get viewers to believe your facts or sources? Or are we doomed to have only people who believe certain things, read stories that support these points? Believe certain things, read stories that support these points? No, I'm not very clear about this question, the second part. But I believe what Arjit is trying to ask is that, how do you get viewers to believe your facts or sources and how do we always trust the personality that it's coming from this place? That's why it must be true. So we caveat it, for example, when I was talking about the tractor story. The getting how many tractors were there, it's like practically not possible. Nobody was sitting there counting tractors. And if you ask the farmers, they say we had two lakh tractors. If you ask the police, they say like maybe on one side they were these many tractors. You tell the leader both, you tell this side is saying this, this side is saying this. I am visualizing what this side is saying. And if the other side is true, that means it's this. That means it would be 10 times of what I've shown, something like that. So we dealt with that ambiguity in that manner. I think data in general, there's a grave issue in data and like a lot of research has been done on that and how do you communicate ambiguity? Because all data is ambiguous. The same data is absolute is a falsified thing just because information is in a spreadsheet doesn't mean it's the gospel truth. Somebody is putting things in a spreadsheet and somebody with biases and somebody with parameters is putting things in a spreadsheet. So some information is always going to be missing. And kind of communicating the reader what information is there, what is not there, what might be missing and what we can say, what we cannot say are the things that you do. You can't make a reader believe you. The side part is that within the age of misinformation people believe anything. So the responsible thing you can do is tell them where you're coming from, what your process and methodology has been and leave for the reader to decide if they want more or are they satisfied? Got it. Got it. Okay, so another question from Lal poster, I guess this is on YouTube asks, how do you choose when to make a visual interactive with tooltips? I read your blog on this, it's clearly someone who's a fan. I read your blog on the same, but why I still ask is because always important to bring out specific details. I think, and I don't have the rest of the line that he's trying to answer. So how do you choose? So I avoid tooltips. It's a very, because it's hard work to get tooltips on and work on every device and on a breaking your situation tooltips are like you just don't have the time to code it out. We do a lot of static visualizations with annotations, because especially in a breaking your situation, like if earthquake has happened, and I'm seeing how bad this earthquake was in respect to others and I'm talking about magnitude and scale I might just do a static chart, put some annotations, put it out there. Why? Because maybe the reader need not explore what we have. Like, if you look at usage analytics, I'm not talking in particular about Reuters, but I'm talking about studies that have been done by the New York Times in practice. They say only 10 to 11% of your readers use tooltips. Press don't, especially people on mobile do not. And the thing with tooltips, but the point is these 10% people might be your most dedicated readers. Like they really, really will engage and the time spent on the page will be really, really high. So when it comes to a COVID dashboard, of course we want to have tooltips because it's a dashboard. When it comes to a story, maybe if the narrative is doing the job, maybe I don't need a tooltip. So for example, this is a story I did, which I did not show today, but it was about the Indian elections in 2019 and I showed the faces of all 8500 candidates that contested. The idea to show their face was to show the scale of things. Like, you know, this is this huge election. So many people can test these many are men, these many are women and like drive home certain points about the people who were contesting how much wealth they have, how much criminal cases and things like that. Now the idea of that is not to help someone choose who to vote for. The idea of that is to show scale. So I do need not make it in a way that I hover over a person's face and I get to know who they are, where they're fighting from, because I think that interactivity or tooltip A would be expensive, because I have to load 8500 images separately. And B is unnecessary because I am telling the story it's not a place for them to come in and see who to vote for. So I think it really depends on the use case that if the use case is more of a story and sharing the finding maybe I don't need a tooltip. If it's a dashboard and it's to help the user make a particular decision or learn something that's based on their experience, we will add the tooltips. Got it. I'll take the next question from another or someone in I think in YouTube Ashish was asked, how do you figure out the feeling to do a data story like you did with the Bollywood song story. Curiosity, you read things and you become curious about it and you ask the question. Oh, I wonder why this is happening. I wonder if I can prove it. And you get that feeling of curiosity and you go and you chase that curiosity. And I get that all the time. Reading helps a lot. The more you read, the more news information you consume, the more questions you get. And that helps. Okay. So, Lal poster who had asked the question before this has has another one coming from an anthropology qualitative background. I feel data visualization, which is stats in general reduces people to numbers. How do I deal with morality of this situation. Yeah, that's as I was talking about earlier. That's like a very tricky bit, right. But I think there are ways to combat it. So for example, if I'm talking about 8500 candidates, and instead of saying they were 8500 candidates and making the line chart of how the number of candidates have increased over time, I put on faces of these people on a page. And the minute I do that, I, it hits me. Oh my God, these are real people. And these people are not ending like these people, more people, more people, more people, more people. So, that's one way of putting making sure people remember their human. And similarly, there was this piece that my colleagues did back in 2015 2016, where we were still grappling with, you know, humanizing data and with things like we talk about data humanism. And within that, they, they were talking about the Rohingya crisis, and they decided to do crowds of actual people that one of our 19 mates drew, like illustrated people to show how the scale of the refugee crisis kept increasing. So using humans is one thing, one way to do it. Mona Chalabi, who is like excellent on Instagram, does these hand drawn illustrations to use data in a very human manner. I think using illustrations is one way to humanize things. I did another talk on like how do you create relatable visualizations and illustrations is one way to do it using actual people is thinking of data as not data, but there's in more visual forms. For example, I did a piece about pollution in Delhi, where we installed the camera on top of a rooftop to click pictures. So instead of saying PM PM 2.5 is rising. I showed two filters next to each other and showed what 40 PM 2.5 is like and 400 is like. And so you see it in a more visual manner, and it's much more relatable than a live chart. So I think you think of these other visual ways to tell a story, as opposed to just making bar charts and line charts. Right. Okay, I, at this point of time, we have about three or four minutes more with good man if there are any other questions feel feel free to shoot I'll just ask a couple of them I'll take a step back and also audience members at 130. I do the closing talk where we'll reveal the the quiz link which you can take and you might be able to win one of the four different prizes we have. Right. I'd like to take a step back from pure anthropological questions or qualitative questions and all these things that that were being talked about, and get take the conversation back towards information architecture. In, in your work and in the work of journalism. What do you feel is the role of information architecture or taking a correcting a lot of information organizing structuring it to be able to create stories now do you feel that that's a very common practice. And if and if that is or is not how does one hone that practice or improve on that to be able to find out stories by structuring information and analyzing them better. So one thing that I have been personally working on over the last couple of months is to structure my notes better. Because you're the research that you do journalists in core researchers, we talk to people, we read literature and we ask questions and we try to answer questions. So, to make my research more structured because I am honestly a very, very, very unorganized human being, and it's quite limiting to read all these things and just have the notes here and there and things like that. I've been trying because a friend of mine just would meet me every time and be like, oh man, you need to make better notes, you need to make better notes. So, I think that's one way, one thing that's very important. So story ideas that people have, they might get lost, I know a lot of reporters who will have spreadsheets of story ideas they have stories they would someday like to work on projects they would like to do. So, to track them because it's part of the reporting process, you're chasing people, oh please respond to my request, how many days has it been, which ones I have to file, which ones I am filing again, which one I have to follow. So these processes that are part of the reporting process, I think organizing information is crucial because otherwise it is extremely hard to manage and do the job you want to do. That's one thing for sure where it's useful, more organized not taking because the reporter's notebook is crucial. What about the other side? What about the other side wherein when if you have already created, I'm sure good one by now you would have created hundreds of visualization if not thousands. Then if you have to present it at one place at one platform, say Reuters for that matter, even at the point of presentation level, you will have to organize them in different ways so that the information is easily findable by people who will come in and consume that information. How does that exercise go? How do you or your team decide that this is the first story that we should keep or which is the way in which people should be able to navigate and find your stories? Is there any thought process that you can share on that piece? Because the world has hundreds of stories happening at the same time. Yeah, so one thing of course is that you are probably publishing the story while it's relevant so you would be doing something on chronologically. No, it's timeliness. I'm talking about timeliness. So for example, if fires happen in California, we would be covering it while it's happening or just after it's finished. Few people are going to care about several months from it after the event has happened. So their timeliness helps because that will help us distribute it in other stories that we are doing about the event. Socially, we get a lot of traffic when it's a timely story. So that way the timeliness helps. Then otherwise, even on a graphics, we have a graphics portal where you viewers can go and readers can go and search stories by subject or by people. So if someone's interested in the environment or someone's just really interested in coronavirus or someone's just interested in Brexit, they can go and navigate our stories that way. But I think most of the few people will go on that page and look for graphics. So most people will stumble upon it when someone shared it or when it's there within a story, linked in a story and things like that. The thing about Reuters is also because we are a B2B model. So some of our graphics will also be picked up by other news organizations. So you really don't know where and how people might discover it. So if states times has like embedded a line chart about the COVID that we made, maybe somebody is exploring our dashboard through that. So yeah, it varies and there are multiple ways you can get to it. Got it. So we've just completed our time. My very closing question at this point of time will be out of the hourglass and martini glass and all of which of them are your favorite and have you invented your own framework by now? I think the zoom in zoom out one hourglass is actually my it was a modified version. The actual one uses different words. So I would say my it's my modified framework of the original hourglass. It's inspired by the original hourglass. But I yeah, but I love the spilled martini because I love leaving Reuters explore in the end. I also love like zooming in and zooming out. Got it. Thanks a lot, Gurman. This was a lovely talk by you and thanks audience for all the questions. So now is the final part for which and this is the part when I'll be stumbling a bit because I have to switch to a presentation. Okay, so so you should be able to see the screen so we have completed all the four talks. It was thanks a lot to all the speakers. Professor Ganesh Baglar, Jeanette, Anshul, Gurman for such a diverse set of talks that you have brought in in the World Eye Day. I don't think we have ever done a World Eye Day where the diversity of talks goes from food to plants to news to choices and nudges and things like that. So, so thanks a lot for everyone to come together. At this point in time, I would remind everyone that the global keynote by Cassini Nazir is available on the link that you see on the screen. You can join for a live Q&A session with Cassini Nazir. I would highly recommend you to watch this talk. It's a prerecorded one, and it beautifully combines how information and curiosity go, go hand in hand. There are three sessions at three different time zones where in Cassini Nazir will be available to take your question and answers. And about 30 minutes prior to that, they'll also be streaming this, this talk so you can go to World Eye Day.orgs 2021 to register for one of the Q&A sessions. How can you get involved with World Eye Day? There are different projects that World Eye Day does, like state of content, content collection, transcription and translation of so many things. World Eye Day or website, World Eye Day is a nonprofit foundation at this point of time. There's a World Information Architecture Cafe that they're trying to set up. Similarly, so you can get involved in a project or you can get involved in a committee that World Eye Day is creating committee that organizes events and community committee that looks at how I can help in education. Or can be taught and also around accessibility and other things. You can also join the global board of World Eye Day. So in year 2022, every year a new board gets created and there are co-directors at different geographies. So you can for, for to know how you can get involved with World Eye Day, you can go to World Eye Day.orgs slash get involved. Okay, so then you can also join World Eye Day conversations at Discord. The link is World Eye Day.orgs slash Discord. This is a Discord is a tool wherein you can do lots of open chats and communications and under different channels. If you're a gamer, you probably already know about Discord. This is the board members that board members for 2021 global World Eye Day team. And this is the regional board for global World Eye Day team. This, here are the set of volunteers and organizers for World Eye Day in New Delhi. Thanks Abhishek for taking all the communications that goes out from World Eye Day. Thanks Manish for moderating as well as holding the fort with Zoom. Thanks Aura for looking after the questions that must be coming on YouTube and other channels as well as producing a lot of artworks and things that were used. So thank you all of all three of you. And I thank myself as well. Thank you. Here are the supporters of World Eye Day New Delhi, Miranj and Hasgeek. Hasgeek has been the event platform for us who are doing the entire recording as well as broadcasting. So now the giveaways, the goodies. If you go to bit.ly.wiad21gifts and you fill up your details there. There are on the left hand side are the set of gifts that you will get for sure. On the right hand side there are going to be six global lotteries out of which you might be eligible to win a paperback copy of, paperback and a signed copy of the book how to make sense of any mess by Abic over or our noise cancelling headphone from Optimal Workshop. So this gift is for added for all attendees. Now for specifically New Delhi giveaways and this is the quiz thing that we had been talking about. There are four giveaways that are available. You can get a ebook bundle from Rosenfield Rosenfield media. This is lift off and living in information you might get one team subscription for SEO RP. You, you could get one of the two ebooks on how to make sense of any mess by Abiko word. What you all you have to go do is go to the online quiz form and go and fill this form, which is bit.ly slash w i a d slash hyphen New Delhi hyphen giveaway. This link is going to be valid for only a couple of hours so at 4pm we are going to close this whoever gets the maximum answers correct the questions are directly coming from all the four talks. That have happened so far. So whoever gets the maximum questions right or the earliest the top four people are going to get all these the gifts that are there on the left by the way forward to mention that in the previous link this link to fill up for the global set of gifts is only valid till tonight. Tonight US Times of it for people in India you still have a little bit longer beyond midnight, but you should go in and fill in your details if you attended the event and and grab all the free gifts that are available for you. With that, see you next year for w i a d 2022 I would leave this slide open for a few more minutes, just in case you are not, you have not been able to copy this URL. Alright, so at this point in time I'm going to stop sharing my screen, and then you should be able to get this link online as well we will post the link. There as well so thanks again once again Manish thanks has geek thanks Abhishek thanks sorrow for helping us out in organizing the world ID 2021 for New Delhi. Stay well have a good year in front of you everyone and see you around next year. Thank you.