 Okay. Hello everybody and good afternoon. So first things first, my name is Jean and I work with NUS. You will hear my story later, but first I'd like to start with another story just to get us warmed up and to set the tone for this two-hour session. So to start with a story, this time last year I was actually seated in front of a room like now. Only I was sitting in front of a whole bunch of Japanese military cadets and they were all guys. And over the next few months I was going to teach them English in Japan and my Japanese was almost non-existent. But the point is that we started off learning how to say really simple things like good morning, how are you, how are you today, how was your weekend. And then as time went on, they started opening up and we started to get to know each other and then they came up with special requests like a really popular one was, this is all very useful English but you know we're not really going to look at going to the doctor or going on a trip but what we really want to know is how do I pick up an English girl in English. So that to them was really important and the point of me telling the story is that in any sort of a day workshop or even over the course of a few months you go through a journey with your peers, with the person sitting next to you and also with the teacher or the facilitator or what have you. You go through this learning journey and I feel it's always most fulfilling if you think of it as a journey and the fact that the journey doesn't stop there. So to end what happened with the Japanese military cadets, we did go through the whole syllabus over three months, it was really nice. I did teach them how to pick up girls in English because honestly they were quite clueless. The first thing that I asked them was, what would you say? And they were like, how old are you? Would you say that on a first date? That is classic dating 101, no no. So yeah they had a lot to learn, it was a great journey and this is what I hope, we're only going to be here for a couple of hours but this is what I hope the day will be like. So to get started today is all about you. So this is about you mapping out where you want to go, what we're going to explore. I'm going to share a bit about my personal journey just as a few lessons and tips and tricks to take away and then we're also going to look at what is data because it's a simple question but it's also not a trivial question and it's actually quite nuanced and if you want to get into data, firstly you have to know what is data and then we're going to have a little exercise into gathering data and dealing with it just to get you comfortable with the idea of what data is and what you can do with it and then we're going to look at some browser-based resources that you can look at and I'm going to point you to more resources that hopefully after today you can then go away with. So like I said, today is all about you and I've been speaking for long enough so I'm going to get you guys warmed up with an icebreaker. So you are all in pairs. You get together in a pair if you are not already in one and ask each other this question. Go. Has everybody had a chance to share their story? Okay, not yet. Okay, I think everyone has had a chance to speak. We are now going on to the next question. You are now all nice and warmed up. So again, it will be really nice if we have a nice supportive atmosphere and another thing that I would like to practice to get us in the mood is later on we're going to have a lot of sharing and when you stand up and speak after that I'm going to ask everybody to do what's called a stomp clap. So after somebody finishes speaking you want to go stomp clap and then you go stomp clap. Okay, let's practice that. Give her a stomp clap. Yeah, okay, and again stomp clap. Okay, so after anybody speaks we're going to not go the that's not what we want. We want a full enthusiasm stomp clap. Okay, so this after the icebreaker I am going to talk a little bit about how I came to do what I do and later on because I get this question a lot so I thought I'd put it out and share my story first of all and then later on if anyone wants to ask about again what is data what are the roles out there, how can I get involved then by all means come and talk to me and the key message I want to get across basically is the subtitle in the slide, Random Walks and Little Bets because it's all very well to talk about empowerment and yeah you just need to go for it, you know, you go girl you can do anything. The reality is there are so many different forces in the world your job, your boss, your family, your degree whether you did computer science or marketing so going into a new career like analytics which is not actually very well defined actually requires you to manage all these factors which is the biggest lesson that I've taken away hence the title of my talk which is Random Walks and Little Bets so what do I mean by a random walk is basically you sort of meander around right and then little bets means you don't know what the right answer is so if you're like a poker player you don't put all your eggs in one basket you diversify and you experiment and then you see what pays off so those are the two themes I want to get across basically so what do I mean by a random walk this brings me on to my first story so I studied in the UK and my degree is actually in plant and animal biology and I don't even mean genetics and working in a lab although I did work in a lab what I also did a lot of was go on field trips in Wales and go to Spain in the cloud forest to look at flowers and we filmed badgers and beavers at night and then we would go and count the bats as they came out and some unfortunate freshman got caught doing it on camera that was the camera was set up to film the badgers ostensibly but then you know freshers and getting it on in a field trip and they accidentally got filmed as well so ouch don't do that but that was my degree basically and then I realized out of my degree that I was very interested in systems and code hence genetics because genetics itself is a code as well but then okay what are you going to do the obvious path is to then do a PhD or work in a lab right but why you're severely underpaid that's the reality of it and then many people end up burning out so what do you do so then you I started experimenting a lot again little bit so random walk I was in the UK doing something like plant and animal biology that was a bit random so what do I then do with it okay let's experiment so I did a stint with the UK government looking at experimental trials in education so experiments you know science all that maybe that would work and then I also did a lot of work with charities and trying to quantify their impact so again data and how that works and then I worked with a social media company looking at their data so I didn't know what would pay off so I just did lots of different things and I found that I really love data it wasn't even about oh data scientists get paid a lot maybe this will be better than working in a lab I just completely fell in love with it oh my god an Excel file really geeky but that's how I got started and so I discovered through making these little bets that okay authors and badgers are really cute but maybe data is just as good so okay step one then what happens after that I didn't want to I didn't want to do further studies and I didn't want to get a job straight out of uni so I went on an exchange program to Japan to teach Japanese and learn English so I mean sorry not teach Japanese and learn English teach English and learn Japanese so a completely different about turn again a random walk and I taught everybody from young kindergartners to the military like I talked to you about an 80 year old grandfather whose English was actually better than mine and so it was a really cool experience I also worked on an organic farm so actually planting in the hot sun not very, it's actually quite tough and then baby sat the little boy we made our own jam ran a cafe on the weekends and where does that all bring you you know it's a totally random walk how does that relate to biology which relates to data and so my year in Japan was up I had a lot of fun and so then I went back to the UK because a return flight was included in my scholarship and okay now you get down to it let's start applying for a job so you spend all your days tailoring your resume and sending it out and talking to people because I mean we've all been there right we've all been there and you wait and you wait and you wait and it's like crickets no response and again so lots of random walks and then little pets okay no one's taking the bait why is that what am I missing here you know some key misconceptions or actually shallow perceptions oh I want to get into data or I will be a data scientist oh to be a data scientist they're asking for a PhD but I don't have a PhD or you think oh people say that AI is the next thing to get into I'll do AI what exactly is AI what exactly are the skill sets needed you know and then I found there was this mismatch between what I had to offer and what people were asking for and so again lots of little bets so I got in touch with an education technology startup from a university connections and we did a project together and that was really cool because A it taught me that you can always blend what you already have and then sort of work from there so I'd worked in data and I've worked in education so education technology startup was the next logical choice but then they couldn't sponsor my visa so then that fell through you know real life intrudes and okay that little bet didn't work out and then randomly I got on the train to go to Reading to go to this Google developers conference even though I wasn't even a developer I'm not a software engineer I'm more into the data side of things and then I got talking with the organizer because as a girl you really stand out and I was like this sore thumb in a sea of men but then thankfully the organizer herself was a woman and so she came up to me and I'm like oh my god another one we have to get you involved and so that was a little incident that really helped me so she said okay well you can you know are I started out with are and she said well we're a developers group so actually developers don't know are it would be good if you could come and talk about your experiences with are and what it's used for so okay where is this gonna go I don't know but it turns out that they were actually working with Google and they put together this whole one-day workshop with Google and then I was featured alongside them you know like this little hanger on at the side and then but then I got to do a two-hour workshop talking about are and because I was talking to developers I managed to stand out because as developers are was new to them and so from there what do I do I don't know well it seems like I've got a combination of education and a little bit of data and a little bit of are and so again I kept on yeah are as a programming language for statistics yeah so okay so I kept talking to people and in the end I got in touch with the Singapore company that was like oh you've done training and you know are so can you come and do training for us and so okay this seems to be paying off so bye bye London hello Singapore and then I got my first real job in data analytics and now we work with NUS to put together their data analytics courses and for some reason actually listed as an adjunct associate professor but it's a lot better than it sounds like but the key point is that now I get to blend education with training with programming and I get to do a lot of research plus consulting which is like a dream come true for me but the reason why I'm sharing my story with you is if you want to get into analytics you know random walks and little beds you have to keep that in mind in my experience that has been the biggest lesson for me because it's all very well again like I said in the beginning to say yeah you can do anything women empowerment go go go but what do you do when real life intrudes when the resume comes in and the job description comes in and says sorry we're only hiring you if you have a PhD in computer science in this one algorithm or the resume comes in and says we want all these skills with her dupe and big data and Kafka and Spark and all that and you're like what is Spark what is Kafka you don't even know so yeah just do lots of little experiments and always be learning that's how you manage randomness and how you manage real life to get into analytics so okay let's start with what I was talking about first this is an exercise in mapping what is data so a key even amongst the experts they come in and we talk to them and you say what is data and nobody actually has a clear answer so again like I said this is a very simple question but it's not a trivial question and it's worth thinking about when you want to get into data because there are just so many different types of data people's first reaction is to do Coursera course say you want to be a data scientist and then move into AI but I want to illustrate through this session that you have that it's good to have a proper think about the landscape and I'm just going to map out the landscape for you so let's start with where people got your post-its so I'm going to start with where people are right now and then we'll sort of move on from there and take a tour of the data landscape so with your partner could you please take a couple of minutes to write down what you think data is and what are some examples of data okay so everyone probably has had some time to list down their thoughts what you think data is so I'm going to me and my lovely assistant we are going to get you guys to put some ideas out there so miss what do you think data is I think data is like a set of numbers or information that you could actually put together to like tell a story or to reflect like a sentence okay can we give her a song okay just in case my name is Phish, I'm not some random person so I'm going to play some more from this side and you look familiar so yeah you look familiar too um what we have brainstormed with our information is this is a random data that we can analyze and see and sometimes maybe massage it according to where we want it to go okay that's how we see data so we have numbers and we have this idea of massaging data so who knows what data cleaning is about who doesn't know what data cleaning is okay data cleaning is when your numbers come in and it's all in the wrong format your dates are like month and year but you want it to be day month and year then you have to get into the right format so that's what you do when you massage the data next let's get a few more going I want about 10 people to speak so any volunteers? yeah okay so data is quantified information quantified information is what I would call data is like okay information can be qualitative as in research or anything but if you put that research into numbers that's data okay can I give her a strong clap okay that's a very interesting point that I want to bring up because data is quantitative who agrees data is quantitative and qualitative who agrees okay keep your hands up okay miss why do you think data is both qualitative and quantitative okay because we can see the facts and evidence through the qualitative thinking okay qualitative and quantitative so traditionally quantitative means numbers right and qualitative means words especially if you are a social scientist so you're from a business background you collect your survey data you collect your opinion data and all of that is words so what happens when you can encode words as numbers then words can also become quantitative right so if you take out your phone and you use google translate or you use say a voice recognition software and you use xiaomi or you know okay yeah so when you use a translation app or anything like that or when you type a product review into amazon then you can actually model language by this is very mathematical but you can put your words into a vector everyone should know what a vector is you're very simple okay you put your words into a vector and then you can do cool mathematical things with that a vector or you can also apply statistics to your language so then I want the reason why I'm saying this is I want to push the boundaries of what you think data is because like I said it's not a trivial question if you want to get into data analytics you better know what you're getting into and so language is big nlp if you think about carousel or at sea or amazon they've got all those product reviews all of that is going into their natural language processing number crunching machine so if you're thinking about data analytics and you work in marketing or sales that may be something you want to get into okay so fish pick another person volunteer have a good discussion we just have some keywords so we thought data relates to like useful details information secrets numbers privacy parts complexity and yeah it's been like all those keywords relates to data so you said many keywords and maybe what if you pick three that you think is most relative to data what would you say what would you say kind of like point out I guess it's bits of information when you actually put them together I think those are very good points so we are now moving on to another definition of data you have issues like security and privacy and bits and bytes another interesting one that I want to expand so security and privacy does anyone work in cyber security here is anyone in an IT department working in IT admin yeah okay so if you want to work with what you have right as an IT admin maybe you manage user names and passwords does anyone do that or admin privileges if someone someone new comes into your company you have to give them their email address right and show them how to use outlook or whatever yeah so that also in itself is data because data and then you have how do we take care of it it's like a pet I see data as a pet to be honest it's my little pet I have to massage it I have to keep it clean so if the numbers are not right I have to get it all in nice neat rows I have to keep it clean and I also have to protect it so I have to protect data that's something that you might also want to get into because that is also in itself a whole different field and you have data on data you know how is the company performing in terms of cyber security so that's another definition of data to have in your little toolkit and also bits and bytes okay this is really geeky but who is familiar with somebody called von Neumann well it sounds like a really weird name to you or who in here knows about Alan Turing yeah okay so I see a few okay Alan Turing, von Neumann they all invented the computer like what we have right now they all invented the computer and this computer is based on the binary system like you know ones and zeros like if you were and yeah ones and zeros however you also have something called neuromorphic computing and so when people say get into AI and you say I want to get into AI well one thing to consider is that von Neumann architecture of computers right now which is based on ones and zeros doesn't actually correspond to the models that are being run by Google right now so when you talk about artificial intelligence it's not some cool thing that you have to get into artificial intelligence has a very specific definition and currently the state of art with artificial intelligence is neural networks everybody anyone heard of that before neural networks deep learning right deep mine alpha go all the buzz words yes so the algorithms that those machines are trained on have take up massive amounts of GPU and it's like this huge room full of computers because the von Neumann architecture I love saying that word the traditional computer architectures are not suited to these kinds of computations because the traditional computers are ones and zeros these algorithms mirror the brain and so they're more of a neural structure and so one thing that people are looking at is something called neuromorphic computing so if you want to get into data what is data is it something that you can process with traditional machines or do you have to look at familiarizing yourself with a new breed of machines and this is what I'm super excited about so food for thought again here's a different perspective to think about data if you want to get into it so one more person what is data I'm going to ask the gentleman to share his perspective I'm going to bring in information that has the potential to be used ok stop clap numbers and zeros so stop clap this is a theme isn't it information bytes, numbers, ones and zeros but we all know that that doesn't cover the full picture right so one more person from the side of the room any volunteers anyone else here I said I said data is the record of the real world in quantifiable formats quantifiable formats ok I think that's a good definition do you want to use a mic and repeat it to the room I said data is the record of the real world in quantifiable format ok very interesting information about the world in quantifiable format ok let's explore that for a moment ok quantifiable format but information about the world encoded in a quantifiable format so let's think about we've had language encoded as a vector like I said and what about images you can encode images in a quantifiable format you can can you encode music in a quantifiable format audio signals yes and what else is there can you encode feelings in a quantifiable format five stars four stars three stars yep in a way yes satisfaction maybe right yeah so I think that's a really good definition because if you think about anything you basically think how do we represent this mathematically or with numbers or even in a structured format that you can then manipulate yes ok so we have had a lot of ideas about data and to sum up I think we've got this idea of information and of quantifiability so I'm going to get you to open your laptops now and let's have a look at this dear data wifi here you call and okay for those anyone have their phones or no so while we're waiting for wifi to get sorted we are going to again talk to your partner about how does data apply to you so we've had this we've explored this idea of what data is remember from my sort of like story at the beginning if you want to get into data analytics you don't want to reinvent the wheel you start with what you have and then think about how you can move it into the analytics space so where are you coming from I'd like you to share with your partner a few thoughts about how data applies to you and what you want to do with it go crazy okay so these are the connection instructions and this is the website that it would be good to visit so who has not connected to the wifi yet okay any problems who has connected to the wifi okay for our next exercise I'd just like you to browse this website and talk to your partner about your favorite week so these these projects are organized by week so I want you to talk to your partner about your favorite week and why it was your favorite week okay so if you go to this if you go to this page then you can see a week of clocks, a week of thank yous a week of to-do lists, a week of emotions so pick a favorite one and talk to your partner about it anyone with no wifi anyone who hasn't accessed the is not working so for those with wifi problems it's probably because so many of us are trying to access the service at the same time so if you can share with somebody or browse when it goes forward that is probably the best solution right presentations so what was your favorite liberalization? week 52 a week of goodbye what stood out for you? it constitizes all the nice messages in you like one readable chart and it's really pleasant to see okay thank you stop okay so we've been at this for an hour and I think it's time we had a little physical break so if everyone can stand up it's a little cramped but sitting is the new smoking right okay have a seat again what person do I have a volunteer no volunteers okay okay could I ask you to share with us your favorite week so you see that she represented the complaints as musical notes and on the other hand you also have data as color right so could you if you understand how the data is represented can you explain to us what the musical notes mean oh okay okay so if we look at the description if you scroll down you can see how they come up with it so the key the key message I wanted to get across with this exercise is that data is not only all about computers actually pen and paper itself is a technology if you think about it for many many centuries it was the most advanced technology the cutting edge so let's not forget pen and paper and personally I always sketch before I approach a problem and so yes the takeaway is that data is everywhere and if you want to get started often art is another way of doing it data visualization and data journalism and storytelling is something really important right now so if you are into that by all means go for it so we've looked at sort of like very analog old school data now I want you to visit this website called Hint FM and honestly I'm a huge fan of Fernanda Viegas and Martin Wadenberg because they are sort of like a they're a team and they always turn up to conferences together and she speaks for half of it and he speaks for half of it they're sort of like this dynamic duo and it's really really fun to watch so Fernanda Viegas and Martin Wadenberg work with Google and you can actually see an embedding projector is high dimensional data is used when you use your machine translation app and like I said can you convert language into numbers and you can and the way people do that is to construct an embedding but what is an embedding you know honest question not many people know so this is a visualization to understand embeddings and what I actually want you to explore is this one TensorFlow Playground so has anyone here heard of TensorFlow yeah I see some raised hands right so TensorFlow is Google's machine learning platform and Microsoft also has a platform that similar is called CNTK and there are others as well I can't really remember off the top of my head but they're all open source so Microsoft or Google doesn't matter which one you use but to give you an idea of what is also going on right now we're moving to the other end of the spectrum play around with this is a neural network which is brain inspired computing that's all the rage right now so just have a play around first play so what it's actually trying to do is that if you have two you have a blue column and then you have an orange column and what you want the algorithm to do is to separate to recognize that this is blue and this is orange so you can imagine if you're in the fashion industry and you have two dresses and you don't want someone to sort it you want a machine to sort it then you would put this algorithm into the machine and then the machine will learn how to differentiate the two categories and this is how you're actually training the algorithm right here in your browser so have a play around with it and add more features or make it squiggly and then watch the network being trained and here you say what is a neural network so this tells you about it and then this is cool what library are you using so take about five minutes to see what it's actually doing read the descriptions so I think everyone has had a chance to watch all the little squiggles come together so the point of this exercise is to show you the other end of the spectrum so data can be really simple and it can be beautiful and it can be art and can be used to communicate but you can also teach you can use data to teach machines essentially what you're doing with massive amounts of data which is what people are really excited about is to basically teach them how to do this so all the new fangled alpha ghost stuff that you are learning about or that you read about in the newspapers comes down to something similar to this however as we've explored before data and analytics encompasses a lot more than just what you read in the news keep an open mind and this is not the only thing out there right now it's sort of like flavor of the month but personally I'm not too excited because firstly you have all that history behind you and secondly there's so much more out there so this is the cutting edge right in your browser how cool is that so actually this is the cutting edge but there's also the bleeding edge so if you look at this one Reinforce.js so it's a reinforcement learning library that implements common rl algorithms supported with fun web demos whatever that means okay let's break it down so everyone knows what data science is right and now hopefully who here hasn't heard of data science before who has who hasn't heard of artificial intelligence before right it's the next big thing right again it's flavor of the month but what exactly is artificial intelligence people are saying get into it it's the next big trend as if you were at the panel just now then everyone's really excited about it but again you know data what does data actually mean and artificial intelligence what does artificial intelligence actually mean so artificial intelligence has a very specific definition it means how do you how do you program an agent to function in an environment and recall that our definition of data before this was how do you get information into a quantitative format right previously we were working with the definition how do you get information into a quantitative format and then that's what people are excited about right because data brings information brings profit in many situations and now people are getting excited about AI but what exactly is AI AI is programming an agent to function intelligently in an environment so is that the same as data and does that definition gel with what you think AI is so this is reinforcement learning is the algorithm behind AlphaGo and AlphaGo sounds scary because it feels like machines are going to take over the world right but again what I want to dispel is the fact that hey you can actually train an AlphaGo like algorithm right in your web browser it's all actually just a massive matrix matrix multiplication so there's a lot of hype but what I want to get across is that it's important to look past the hype and have a think about people are saying AI is the next big thing but is AI second generation data science or is it a completely different field altogether so let's have a look at this library what are you actually doing so remember previously we had our playground and you were trying to classify blue from orange right right now what you're trying to do is if you look at parkworld dqn so this is your monster parkman and then something is coming along and trying to eat you so this is your little monster and then this is the agent that you're trying to train here so you're trying to train it to avoid this monster of doom right and you see it's avoiding it it goes green and it's trying to learn the boundaries this is the environment and this is the agent here so as you're training you should see it get smarter and smarter because it's learning from its past experiences so you see now the monster is coming to eat it and then it's turned red and now it's away from the monster so it turns green so what you're doing is that you're training it by rewarding it when it's going away and punishing it so to speak by when it gets to the monster the red part so this is also why I said data is like my pet because you also train it like a kid so I encourage I encourage you to explore more about it UnforcedJS is all in your browser and this guy Andrej Kapathy also has a course in computer vision at Stanford so if you're technical I also encourage you to check him out well check out his courses and his blog he's very good and now he's at Tesla so this is state of the art but again to emphasize you also have this this is also data so data is everywhere and hopefully I've gotten you to have a think about what exactly is data and how do you get into it and what it really means so another thing to look at is this one so I had hello I had a lot of fun is it okay if I speak right? I had a lot of fun doing stuff like chicken rice compared to and one thing you see about this is over the past five years but if you look over say the past seven days and you restrict it to Singapore and you look at so what I found really interesting about this is that if you look at it the spikes the highest amount of time that you see people searching for McDonald's is actually at four a.m. in the morning and here as well four a.m. in the morning and then it dips and then you get a rise again at seven okay people want their breakfast maybe but actually throughout the day it's pretty low and then suddenly at night you get oh no I I'm suddenly hungry where's the nearest McDonald's and you see the spike at four a.m. so this is an example of how this is why data gives you insights because you get an insight into somebody's behavior aggregated over time but what's also really nice is if you compare it to say chicken rice and you see that chicken rice seems to have seems to be pretty normal like what you would expect so you see okay people having dinner and then it goes down as people try to sleep and then somehow again five a.m. maybe people are getting ready for breakfast and then again another spike at 12 okay people are having lunch right so chicken rice follows the more normal pattern breakfast lunch dinner and then you have Bakute okay let's remove what I like about this is that you see for Bakute you don't see a lot of activity in the morning and then after that okay some searches for breakfast for lunch you have this really big spike here at 6 a.m. so you can see a story starting to form right for McDonald's it's a desperate midnight oh no I'm hungry what am I going to do kind of thing and then chicken rice is that all time favorite all lunch dinner you know doesn't really matter maybe in the morning and then Bakute seems to be a dinner thing so you see this spike in the dinner time and then for the rest of the day okay maybe 3 a.m. I don't know but throughout the day not so much and then at dinner people want to have Bakute so you see this gives you an insight into the role that food plays in our lives and also the habits of Singaporeans and what's going on in people's lives so this is an idea of how data is accessible and how it's really cool as well so have a play around with trends with your partner and pick one that interests you and take about 5 minutes to do that and then we'll share what you've seen this is my question about Google or trend that you just introduced based on the search that we did right these are the search that Google has actually did it by the search engine yeah so this is every like you know you google stuff and this is the data that they collected so it's not like the actual for example if you type what you type into the browser thank you so much thank you do you want to facilitate yeah sure thank you I want to share what the friends that you've served for volunteer do you want to be on it oh you have one okay this is school holidays we don't get fidget spinners and we go over the past 12 months the interest in fidget spinners actually exceed and it goes at one point may around may have 27 27 20 days cool anyone else do you want to stand up okay can I have your attention please hi so we're kind of like looking at political figures all around the world so Trump versus I think like an interesting insight that we discovered was that there was a spike during election period so I think there's a lot of searches on full trend recently there was one about and his brother anyone search for that I'm curious sorry you did okay let's share okay I did it for the father and the children we have actually talking up talking about yeah what does the measure talk about what's the trend about how much I think it's more search it's more of a search and the I'm actually still looking at it I'm trying to correlate it to really certain events that might have happened announcements that might have happened which is actually I mean that's my current hypothesis right some of them one of them would have said something mentioned something what what thank you can we have a stop please okay so through this you have an idea of how data is actually reflective of the human condition you can say it contains insights about human behavior what's going on in people's minds and so this is why data is special so another question I get asked a lot is how does data apply to me how can I use it in a way that's beneficial not only for my business but for myself and so to come back to my theme of always starting from where you are rather than thinking about things about data in abstract terms like my job or a career I thought okay let's start with yourselves so I'm sure you've heard of the quantified self movement Fitbit everyone's heard of Fitbit right track your sleep track your steps track this track that so if you could track something what would you track and this is something again that sounds simple on the surface but actually I feel is not a trivial question so again like what is data sound simple but actually encompasses a lot so tracking what matters is again what matters is a sound simple but it's actually not a trivial question so to give you an example if you have a Fitbit and the quantified self and you want to track right the first thing you think how many steps did I eat my vegetables today how many hours of sleep who here is really interested in how many hours of sleep you got okay let's flip the question are you more interested in the number of hours that you've slept or the quality of your sleep right so if you buy a Fitbit and it tells you how many hours you slept is that data that you want then that's probably not data that you want so to drive home the point the point of this is to get you thinking about not only what is data we've gone through the whole gamut of what is data and I want you to think about what is useful data and a key thing that happens in work a lot is people say oh we have a lot of data what can we do with it and then you go into it and none of the data is useful and that's kind of a bummer because you could have spent all that time collecting real useful data so again starting with yourself what do you want to track so I had to think about this and it was like number of times I said thank you does that can I use that to show whether or not I'm being a nice person what about how many times I said please or how many times I opened the door for somebody things like that and that's how I if you think about the dear data project you can use that for inspiration they have a lot of metrics so for this exercise I'd like you to take about 5 minutes to think about if you're working with the quantified self what metric is meaningful to you how will you measure your life it's a meta deep question did my kids go to uni that is one metric did my kids turn out to be good people that's another metric and think of the metric we'll have, we'll get people to share and ideally try and be creative maybe with the teeth maybe with the teeth so I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I