sidneylafontblr-blog

[sidneylafontblr-blog]

Narratives

Stories are important. Stories draw attention and captivate audiences. Stories are hidden in everything, every mundane sentence, and passing reference. Narratives drive conversations. So why not take advantage of the power of stories every time you communicate anything to anyone. I am a strong believer in this principle, that is why when signing up for courses this semester when I was deciding between two classes, Public Speaking and The Art of Storytelling, I quickly choose The Art of Storytelling. Frankly, it was a no brainer. I gave me the opportunity to continue to work on my public speaking every week and my ability to craft a story, a skill which can aid me my entire life. So when I watched Cole Nussbaumer’s presentation on the art of crafting data visualizations I was intrigued, when Nussbaumer highlighted the importance of telling a story with your data. Within a few seconds, I realized the truth in that statement, a line graph and a table are just as able to tell a story as a spoken word presentation.

   Beyond the ability to tell a story with data Nussbaumer brought up some other really interesting points about the psychology of designing a data visualization. Particularly about the use of color and style, both of which can work to draw the viewers eyes to specific points on a graph. I have a few friends who are Graphic Design majors and they have been telling me just that ever since I started talking to them. And having looked at their projects and talking through their thought process for each element of the graph I have realize how deliberate each of their decisions is, and how impactful they can be. Nussbaumer strategy of determining where the eye is drawn to by looking away and then back at the visualization is simplistically smart, and definitely something I plan on utilizing. The process of crafting an effective data visualization is yet another example that highlights the importance of interdisciplinary thinking, due to the attention paid to the data, design, and psychology behind the process.

   Hackers suck, hard. Sorry about my bluntness, but malware was downloaded on my last computer and surprise surprise it broke. Truth be told it was 100% my fault. I have a bad habit of using less than legitimate soccer streams, and I watch a lot of soccer. In the end, I was lucky though, I didn’t lose any documents or downloads, it just destroyed my computer. But still, it was all the hacker's fault, 100% their fault. Although used to love jailbreaking my phone, so I guess I’ll give hackers a break.

   Although it would be interesting to learn about hacking, take a cybersecurity class or two. It seems like the last frontier, the wild west, of the tech world. It would give me a better idea of what that field entails before I entirely write it off. It's the closest a computer science major can get to busting into the front door of a saloon with a pistol in hand (I have always wanted to be a cowboy).

[sidneylafontblr-blog]

Trees

Decision trees are wonderful. I have been making versions of them for years, and they have helped me make sense of some of my hardest and easiest decisions. Although I do not always follow their output. It is incredible to think about how widely decision trees are used. Everyone from kids in college to governmental leaders to CEO’s use the same technique I use to make decisions.    What I find really interesting is how intuitive decision trees are. Whether or not you know what a decision tree you have used one. Even children use tree-based decision processes. Trees as a data structure, in general, are very intuitive. Everyone naturally uses trees in their decision-making process, we all start from one decision and follow all the paths from that decision all the way to each possible conclusion, applying probabilities along the way. Just like certain structures of graphs, like pie and xy graphs, and tables, trees are so intuitive that they were never really invented they were always there, ready to be discovered by every individual person. It is the question of the chicken and the egg, which came first, did someone discover trees or did trees always exist? The same paradox exists in so many facets of life. For example, the superman pose, arms on hips, legs spread apart, and your chest pushed out, has been proven to make inspire confidence and make a person more likely to be able to complete a task if they hold that pose for some extended period of time before the task. So did the inventors of Superman invent this pose and the fact that Superman does it give this pose power, or did this pose always exist and the creators of Superman used it because they knew it inspired power and confidence? My point being there are just some things that are just inherently human. They cannot be invented because they always exist, but each individual person can discover them. Flowcharts are sort of in the same category. When they are just processes connected by arrows with a clear beginning and end they are incredibly intuitive. But, when specific symbols are added flowcharts become a language that takes some explaining.    What is really interesting is why trees as a data structure are so inherently human and intuitive. Is it because that is how data is stored in our brain? Is it because that is how decisions are made by our body? Or are trees just really easy to draw and read? These are questions for a psychologist or biologist, but nevertheless, they are interesting.    The end result is that is not surprising that computer scientists, information scientists, and data scientists have translated these intuitive data structures onto computers to store data. It also means that teaching these structures, within the realm of computer science, is not ultra confusing and allows students to transfer their ability to think in non-computer related processes into computer-related processes with relative ease. Which I for one am thankful for!

[sidneylafontblr-blog]

Common Sense

Common sense is a real blessing. Especially in Assignment 7. I could not figure out how to write the table then run queries. I was completely stuck. Super glue stuck. But after a few hours of research and frantically guessing how to write the table then be able to use it, I took a step back and applied some common sense. In general, you have to write something before you read it. That one piece of common sense unlocked the whole assignment. Once I realized that I had to use dbwritetable before I used dbreadtable the rest came pretty easily.

This assignment also provided a great opportunity to research. Since I had no computer science experience before college this course has helped me dip my toes into computer science research, and this assignment, in particular, emphasized it. Frankly, websites like StackOverflow and rdocumentation are really confusing and to makes sense of what I was reading I had to use a few websites at the same time to piece everything together. But this is valuable experience because a large part of my career is going to be spent reading those sort of websites and trying to make sense of them. That is just the reality of the work I guess. Furthermore, I am sure that as I gain more experience understanding documentation will be easy so might as well start now so that later it is that little bit easier.

Also now that I know how to actually combine data from an excel file and a SQLite database with R, I have begun my exploration of R. By that I mean, now that I am able to move data between a few different programs, including R, I am able to do a ton of different things with that data in R. It feels vaguely like I am starting my education in data science. It is exciting. Hopefully, if I can motivate myself I can start some projects I have been thinking about for a while. Specifically, I have wanted to check the accuracy of sports predicting websites, specifically in terms of soccer. It has been something I am curious about for a while and now I am beginning they have the skills to attack this problem.

Recently we have been learning about Bayesian analysis. Although I have learned about this before, twice in both a statistics class in high school and discrete structures at Northeastern, I am excited where it is heading. In this class, we have started by learning some algorithm or principle then returning the realm of computer science by applying that new information to some program. So the reason I am excited about learning about something I have already learned about is because I am excited to learn how to implement it in a program that I can use really easily. Who know maybe I will start using it in my life to avoid making dumb, bone-headed, and idiotic decisions because god knows I have made enough of those.

[sidneylafontblr-blog]

Excel and Roy Batty

Wow. Excel is really something.

   And on this weekly installment of a technologically challenged kid being a computer science major (just joking… sort of), I did not realize how powerful Excel is. From my understanding of some of the uses of Excel is can essentially be used as a database for pretty complex data tables.  Not to mention that is much like eclipse or atom it can support programming languages. Mind blown. Beyond that Excel has its own language, even if it is limited. I guess when everyone told me that I should learn Excel they were not joking. I remember my mom used to strongly recommend that I learned Excel really well because certain people she worked with only used Excel, people she deemed Excel gurus. Honestly, I assumed that was a bit of a bullshit job because I did not think there was much to Excel. I assumed it was about as complex as word or powerpoint. But you know what they say about assumptions. So this Excel is another thing added to the ongoing and very large list of things I should invest some serious time into learning this summer.

Recently I watched this documentary called Alpha go, and it was incredible. It is about this company, DeepMind, that created an artificial intelligence program that beat a top professional Go player. If you do not know what Go is it is an incredibly old Chinese board game, that also happens to be the most statistically complex board game ever created. For example, it is significantly more statistically complex than chess. And in certain parts of Asia, the professional Go competitions and players is a huge deal, like chess in Russia big deal. So obviously this is a huge deal. It confirms that humans can create an artificial intelligence program that can surpass a human in a singular task. And there were a few interesting revelations from how artificial intelligence program viewed the professional Go players moves. The one game was the human won, it was due to a move that experts would call creative, but according to the artificial intelligence program, it was just the best statistic move to make. It is just intriguing that what humans view as creative, an artificial intelligence programs view as effective. I guess that just speaks to humans’ need to distinguish between levels of intelligence qualitatively because we do not know how to quantitatively, while an artificial intelligence program can do that quantitatively, so they do.

But let's not run away with ourselves, we are not going to be able to create a conscious artificial intelligence program. There is still the chess problem. Which is the problem that this Alpha go program does not know that it is playing go. It is just doing all that it knows how to do. There is no ambitious or ability to reach beyond its given domain, which is obviously not a characteristic of consciousness. So although we will not going to be able to create Roy Batty, the replicant from Blade Runner, any time soon this is a huge step and marks huge progress.

[sidneylafontblr-blog]

The Labyrinth of XPath

Ok, it turns out XPath is a bit more than just syntax. Turns out you need to actually think a little. But it was not bad at all, in fact, it was kinda fun playing around with the queries until I figured it out. It was not too difficult, but it was just hard enough to be satisfying when I got it. Furthermore, the log behind Xpath is kind of interesting. Way more interesting than actually building the XML document. It is like a labyrinth and you need to find your way to just the right spot then find your way out again. And much like a labyrinth, you navigate it was done with a metaphorical blindfold on. I had the tutorial, so I had a lot of information on how the basic functions of XPath, but with those, I had to figure out the right order of functions and the right combination. Although I did not being reminded that Jeff Sessions is a member of the American government.

Also through learning about Xpath before I learn about databases, excluding XML, it gives me a unique and I think a positive base of understanding for how data retrieval works. Although XPath works for the tree-based data storage of XML it is illuminating how one actually goes about retrieving information period. There are built in calls for specific sets of data, to begin with, that the company/person knows will be used. It makes a lot of sense. Furthermore, it underlines the importance of having queries that will not change if new data is added to the database.

Beyond that, I am excited to start learning R and SQL. As a data science major, I am going to have to use the programs, or similar ones, for my entire career. I know a little about each but not nearly enough, so this will be an extremely helpful section of this class.

Also, I just applied to an REU at Georgia Technical University for civic data science. I know I stand a very small chance, but it is exciting that I am in a position where I can even apply to something like that. Especially because it is about civics. I have always volunteered and it a real point of pride for me and having the opportunity to learn so much from my civic duty related to my career is exciting. And the stuff we are about to learn will be really useful if I do end up getting it.

Just because I can I want to talk a little more about data analytics in sports. It is a travesty how unequal statistics are still compared to each other. A basketball player’s field goal percentage who shoots mostly close to the basket should never be compared to another player’s field goal percentage who shoots mostly three-pointers. It is ridiculous, yet it happens all the time with no consideration of risk versus reward. Of course, the player who shoots more three-point shots will have a lower field goal percentage, it is harder, but a three-point shot is worth a whole one and a half times that of a two point shot. And we have the metrics to measure true field goal percentage, yet it does not appear on a box score and the total field goal percentage does.

[sidneylafontblr-blog]

Stat padding

This week we learned about XPath, which was pretty self-explanatory. It just involves some basic syntax and some knowledge of discrete. So this week I am going off the reservation in this blog post and talking about statistics in practice. Specifically, some of the problems with statistics if those who are being measured know that and how they are being measured.

I am going to use basketball and soccer as an example here but anything will do. Frankly, I just like basketball and soccer, and I will try to make it understandable by keeping the buzzwords to a minimum.

If someone or something knows they are being measured, and how they are being measured they will take advantage of it and make the data useless. Using Russell Westbrook’s, a point guard for the Oklahoma City Thunder and a former league most valuable player, rebounding as an example, he knows how important rebounding is and how important it is on a statistics sheet, so he has figured out a way to use this to his advantage. He DEMANDS that all his teammate let him get defensive rebounds no matter what so he can record triple-doubles. In fact, he has averaged a triple-double for the last three seasons, an incredible feat. So by looking at his season statistics, you might think he is one of the best rebounders in the league, when in fact he is not. And this problem goes beyond just rebounds, Westbrook does the same thing for assists by passing when he should shoot or forcing teammates to shoot.

Basically, the point is statistics are important when in reason. Statistics are a powerful tool when their flaws are understood and accounted for.

Another example of this is in soccer with shots. Shots often valued over goals, and although this may seem counter-intuitive, shots often are an indication of how many goals a player should score and if a player has a bad shot to goals ratio it can even indicate potential, especially shots within a certain distance of the goal. The idea is that accuracy can be taught but positioning is harder to teach and if a player has a lot of shots per 90 minutes they have good positional awareness. Yet shot numbers are drastically inflated by a number of factors, most importantly the quality of players on the team. If the team is good a player will have the opportunity to take more shots and vice versa. This leads to players who play for good teams being drastically overrated and players on bad teams being underrated. Someone looking at these statistics must go farther and look at how the team plays and actually watch a player to judge their positioning. Also, a player can take bad shots, ones that most likely will not go in, to pad their stats.

Frankly, who can blame these players? If their statistics are better they will get paid better and it will help their reputation. So, of course, they will pad their statistics, but the consumer of these statistics must go further and judge them beyond statistics.

[sidneylafontblr-blog]

XML in Practice

Although watching a video and reading an article about a new language might have made me feel as though learning a new language would not be THAT difficult, that was misleading. After watching the skills sidebar and reading the tutorial I felt ready to attack the assignment, but half a week later when I actually sat down and started writing my XML document it was a completely different story. I ended up having to re-read the article, re-watching part of the skills sidebar, researching some syntax issues and examples, and going to office hours. Those are hardly the mark of quickly learning a new language.

But that is not a bad thing. Struggling is good, it is a sign of improvement. Or at the very least it is a sign of attempting to improve, which is nothing to scoff at. Many, too many, people would not even go that far. So I am being positive, positive about my what my work will lead to. Furthermore, I am still confident in my ability to learn what I need to even if it does not coincide with class material, for any of my classes.

It is two days after I wrote that first bit and I just finished assignment 3, XML document and all. I have to say that after spending the time to relearn and focus on the basics of XML, I feel confident about my work, although I could be entirely wrong. I hope not. It was not nearly as difficult as my complaining above from a few days ago made it seem like it was going to be. In fact, once I had finalized my UML chart it was not that difficult to translate it to XML. So I guess this is just a really good learning experience that I can use to motivate myself later.

Since we have been learning so much about XML a did some research on it, really I just skimmed the Wikipedia article on it. Just to get some background. And it was really interesting how much it is used, and how much I have used it without realizing it. Specifically, Microsoft Office uses XML. I have been using Microsoft Office since fifth grade, it is baffling that I never heard of the programming language behind it. I guess the brilliance of good software/programmes is making the whole thing seem easy and seamless, and not advertise what happens in the background.

Although I have learned so much in the last week there is still something that is bugging me. I do not know how XML works. I know it is a data storage visualizer, but I do not know how to visualize the data in an XML document. I know later we will learn about parsers and more about XML’s utility in general. But my complete lack of knowledge about the why and how in terms of XML’s utilization is extensive. And I am the sort of person where I am haunted by lacking knowledge so hopefully, I will be able to understand the upcoming sections in class so that I am just a little less confused.

[sidneylafontblr-blog]

A Learning Week

Firstly, before I delve into any revelations that I had this week related to this week’s material I need to give some context. Context is always key. I thought I was going to be a math major before I got to Northeastern University, and maybe pick up a computer science or data science minor. I had never actually taken any computer science courses before or tried to learn any programming languages by myself. In fact, I considered myself pretty technologically inept. I was the kid that it took an hour to figure out how to right orient text in word and ignored the complexities behind the technology I used daily. So it was a surprise to me that I actually enjoyed fundamentals of computer science 1 and decided to switch into a straight data science major. Nevertheless compared to many people around me I am clueless about computer science-related issues. This week’s material in principles of information science has been eye-opening about the reality of coding for websites. I knew that javascript and HTML coded the exterior of many websites, but I also thought they coded the internal stuff too. Frankly, I did not realize that there were different languages for the appearance of websites the internal stuff. Now that I type that it seems incredibly obvious. It makes a ton of sense that multiple languages go into programming a website, not just javascript or HTML or some other language.

Furthermore, this week’s material has started to open my mind about the incredible amount of programs and languages that have specific and unique uses. Having just watched the skills sidebar that proves to be especially true. The number of languages and programs dedicated to testing and complimenting XML is staggering. Frankly, I would have thought there would be just a few, like maybe three, complimentary languages and/or programs to XML. Also, the skills sidebar illustrated how much XML is actually used. I had no idea it existed just a week ago and I am supposed to be a data science major.

I mentioned the skills sidebar before, but I want to highlight how illuminating it was, not necessarily to teach me about XML, because I had read the weekly reading and watched the lecture beforehand, which had focused on the basics of XML. It was the skills sidebars’ use of HTML and javascript that was illuminating. I have never studied or used either of these languages, bar using java for the last few weeks in fundamentals of computer science 2. But I found myself generally understanding both throughout the videos, which was incredibly surprising and definitely a positive revelation. When I say generally I mean very generally, but that is still something. It shows that I have at the very least gained the ability to vaguely understand languages I do not know because of the similarities to languages I do know. It definitely gave me confidence because this summer I am planning on learning Python. Also, the people who teach those skills sidebars seem to be required to be very monotone.

[sidneylafontblr-blog]

Language

This week in class we focused on a language, UML, and all of its intricacies and ability to represent information. This got me thinking about languages in general how well they do their job, allowing people to exchange information. This exchange is key to our survival and always has been. In fact, the need to exchange information predates any language, especially the advanced modern ones we know today. That need has always been there and is tied deeply to survival. So due to this inevitable need to exchange information languages were developed. It took time and was done among many groups of people. At first, I would imagine it was incredibly non-verbal, pointing at something or demonstrating an action with your hands, but eventually, grunts were integrating. Then those grunts turned into discernable sounds and what we currently define as a spoken language. Those early humans adapted by developing mechanisms to transfer information so they could better survive.

Now there are millions of languages and not all of them are even spoken. The interesting thing about all of these modern languages is how they are all suited to who they are used to communicate to. For example, a deaf person cannot hear therefore they cannot communicate verbally, so sign language was invented so people could communicate to and with them. But it goes beyond just humans, we have created languages so we can communicate to and with computers too. In fact, I am spending a lot of money on tuition to learn how to do just that right now. These languages have developed massively and become so specialized. Much like there are tons of languages used to communicate with different groups of human beings for different reasons there are languages to communicate with computers in very specific ways to. In this way, our generation has adapted much like early cavemen. Furthermore, we have gotten to the point where we have outgrown languages. We have developed such effective new languages that others have been left in the past. We have used those dead languages to create new ones that can be used more readily and effectively.

We have created such efficient methods of transferring information that there are now people who dedicate their lives to studying the history of these methods. There is a certain irony in creating a language so advanced that it must be used by professionals to dictate its own history. It is almost autobiographical. It would be like me using DNA, as a language, to write about the history of human kind.

Now we also have the use of multiple languages at once to store information. Right now I am using English to communicate something to random people on Tumblr and my professor. But the program I am using to write this on probably took a few languages to create and the computer which the program runs on definitely uses a few other languages. Each set builds off of each other to allow me to type this out onto a google doc. It is really remarkable.

[sidneylafontblr-blog]

Information in Real Life and its Storage

For almost the entire first lecture we focused on the philosophy and definitions behind data, information, and knowledge, which at the time seemed like a perfectly legitimate exercise in contemplation. Yet after class, an interesting thing happened. I started to see those distinctions wherever I looked or listened, especially in other classes. Furthermore, it dawned upon me that classes provide an incredibly unique opportunity to study the distinction and passing on of data, information, and knowledge. That is the whole point of a class. But the definitions of these three concepts are very important in them passing on. Innately data is impossible to pass on in any sort of teachable way, you cannot hand a student a page with a bunch of numbers on it with no explanation and expect him to learn anything. But very few professors would ever do anything like that, but most will pass on information without explanation, which is data. Making the passing on of that data useless for students and a waste of time for the professor. But the real problem here, which I am beginning to understand is part of the knowledge that goes with being a professor, is whether the professor knows he is passing on data, which is useless, or information, which is the goal. And if I follow this problem to its true end, I find that the knowledge of being able to look out at your students and tell if you are passing on data or information is a piece of tacit knowledge. Which is very difficult to pass on, and in this case seems to come with experience.

Furthermore, this class has forced me to rethink how I store information, both in a physical and mental way. Physically my information is a mess, I do not tend to keep my notebooks or books neat and papers have a way of getting lost in the mayhem. While, before I would attribute this to a personality or a preference because for some masochistic reason I do not want to know where I put things, but I am coming to realize how unproductive it is. Frankly, its also a stubbornness thing too. But if I am going to take a course that focuses on information storage the least I could do is implement it to improve my life, also it is the start of a new term so it is a lot easier for me to start all my classes organized. Mentally this has meant consolidating lists. Over the last few years, I gave up on trying to remember everything I need to do, I just do not have the room in my head for all of it, so I started making lists. But I would make those lists wherever I thought of the stuff I needed to do, whether that was in a specific class’s notebook, on a sticky I stuck to the wall, or on my phone. So this term I am making a conceded effort to actually consolidate those list so I do not lose them, which ironically had become a problem. So now on my computer, I have a notes tab for both my personal and school to do lists and so far I have stuck to it.