A blog for a class

[prostheticknowledge]

Mobile Robotic Fabrication System for Filament Structures

Project from the Institute of Computational Design uses wall-climbing robots to create web-like structures:

The project Mobile Robotic Fabrication System for Filament Structures, demonstrates a new production process for filament structures. It proposes multiple semi-autonomous wall climbing robots to distribute fiber filament, using any horizontal or vertical surface, or even existing architecture, to support the new structures. Compared to larger scale industrial robots that are limited by position and reach, these robots are enabled with movement systems and a collection of sensors that allow them to travel and interact accurately along typical ground, walls, roofs, and ceilings. One can imagine a fabrication process where an operator arrives to the scene with a suitcase housing all the necessary robots and materials to create a large structure. These agile mobile robotic systems move robotic fabrication processes beyond the constraints of the production hall, exposing vast urban and interior environments as potential fabrication sites.

More Here

[blurrypaths]

Mmm…

[thethingaboutprogramming]

The thing about programming is that just because something can be done in a given language, doesn’t mean it should be done. I’m looking at you, JavaScript.

(via thethingaboutprogramming)

[ryanandmath-deactivated20180118]

Report Cards for Famous Mathematicians by Math With Bad Drawings

I don’t think I’d usually post these, but something about them is cute in the right kind of way. Also, I think these really succinctly speak to certain controversies / flaws in the current math education system.

[mathprofessorquotes]

”I wanted to show you all a Halloween joke and I can’t stop laughing at this one.” - Probability professor

[bendoescomputers-blog]

The end.

This is it. The end. El Fin. The (probably) final original post on this blog. My computer science class has come to an end and my final blog prompt is "where do I go from here?" A fantastic question really, one whose answer has changed so much since this class started. When I came to this school to pursue a degree in software engineering my long term focus in computer science was cyber security. Now, eighteen weeks, thirteen blog assignments, two free textbooks, and a switch from software engineering to computer science and math later I can say with the utmost certainty that the following is the unequivical answer: I have no clue.

My focus is still probably going to be on cyber security, but there are a number of other things that hold an appeal to me now. Network administration, software design, AI development, cybernetics, and even robotics (a field I never would have considered after a partcularly bad highschool experience) are all things I would now consider studying.

Looking forward scholastically (since school is a large part of my forseeable future, yay for grad school), the classes I'm probably looking forward to most are the cyber security classes. The two lectures we had in that class demonstrated that I know next to nothing about cyber security beyond footprinting. At the very least I do know plenty about footprinting. Outside of school I'm hoping to get a government cybr security job or any job abroad. I'm not too sure how those would work together though.

My plans for the future are ultimately flexible. As I've said before, I'll be happy in any job with computers, and there are probably a good number of non computer jobs with which I would be happy. In the end what I learned from this class is that the field of computer science is a lot broader than I had imagined. It is also one that is more widely spread than I had believed. It turns out that computer skills can land you a job with nearly any company. It's a field that may seem volatile, as the ex-techie at my dad's company said "technology always changes, so I figured I may as well get a degree in sociology." There's a reason he's the ex-techie. The most important thing I learned is that, while Computer Science has changed a lot since Babbage's counting machine, it really isn't all that different.

[bendoescomputers-blog]

As much as I hate to admit it, and despite my most valiant efforts, I fell asleep during our industry speakers panel. Sadly, I missed about twenty minutes off information in addition to however much time I spent focusing on not falling asleep. This wasn't because I wasn't interested in the speakers themselves or what they had to say. I had been looking forward to that panel since it was announced. Sadly, my class schedule for the day has me getting up at six am and going to class until five fifteen with no time for naps in between. I'm slowly beginning to realise this is not a good thing.

During the parts I was in my right mind for, I did get some interesting information. For example, a lot of them mentioned it would be a good idea to work before going back for a master's degree. This wasn't my original plan, and isn't my exact plan at the moment, but it did remind me I should probably make sure I like the field I plan to study before I begin Graduate School. They also talked about different benefits their companies provide and, while not every benefits package is made equal, they all sounded pretty neat. As my readers (if any are out there) remember, I didn't have too much info on tech companies when I had to write about those. Now I know about four more companies that are located nearby.

After hearing about these companies I would certainly consider working for them, (especially if they give their employees oppurtunities to travel). At this point in my life I'd like most to work in cyber security (a nearly universal field), but I feel like as long as I'm working in computers I'll be as happy as I'd be in Cyber Security.

I'm very greatful for having the industry speakers despite what impressions my state of consciousness may have given to them. The information I got from them was helpful and will continue to aid me as I go forward in computer science and possibly cyber security.

[bendoescomputers-blog]

In a shocking choice I decided, when asked to write about a prominent technology company, to talk about Google. Of course, Google the company isn't exactly the same company as it used to be now that they've restructurred themselves. A number of their projects are now being managed by their holding company Alphabet. Nevertheless, Google is still a company I have a lot of interest in. The numerous services Google provides, including maps, music, drive, and hangouts have all become important tools for me since switching from an iPhone to an Android. Granted, most of their Apple equivalents were important to me when I have an iPhone, but I've found a number of Google's services to be improvements. Using Google Maps has made navigation a lot simpler since it's a lot more precise. I recall an instance where my Iphone got me to a crossroad with two buildings that could've been my destination on either side, but my iPhone told me to take a left when my destination was right. Returning to the same place once in the next year my Android told me the correct directions (although at that point I remembered to go right anyway). Granted it had been a year or two, and Google maps had been around for awhile longer than apple maps (which has improved a lot since its initial launch). Small differences like that are the reason I decided to continue using android phones after my first one.

Alongside using their products, I wouldn't mind working for them either. Every story I've heard about working at Google (or any other tech company in their area) makes me wonder why anyone would choose another field. Of course, their hiring system probably isn't nearly as entertaining as it was presented in The Internship, but I'm not so sure that's a bad thing.

Google has come a long way since they started out as a search engine for the internet. Since then they have expanded into doing maps, distributing music, supplying phone software, having a mostly decent translation system, supplying cloud storage for documents and slideshows. They even bought out the best known video sharing website. Frankly, I'd be happy at any tech company but Google is the one I know anything about, however limited that knowledge may be.

[bendoescomputers-blog]

Tubes by Andrew Blum

This book did not hold my interest. The topic could have been interesting if I had no backround information on the subject. In addition, the subject of what the internet is and it's history are, like many other things that involve history, entirely uninteresting to me. The author talks relatively early about the development of the internet as a military pursuit. This is something I already knew about it from lectures in various classes, including the one for which I am writing this blog, and science museums. Then there's the subject of how the internet works today, as a series of data centres located in various buildings around the world. This seems to me to be a bunch of trivia that would only be useful to Jeopardy afficionados such as Ken Jennings, Watson, and Sean Connery. Don't get me wrong, There could be some useful instances where knowing that the internet is more than just some abstract thing connecting people to each other, but a collection of tubes connecting everyone, but I feel like a number of those situations involve challenges to one's true status as a computer scientist from a computer scientist looking to justify his knowledge of how the internet works. Or maybe when you're trying to get a job maintaining the internet. Beyond a general uninterest in the topic of the internet I'm still a little annoyed about a poorly made documentary about the internet that was only hosted on a website that decided every few seconds it needed to load the next few seconds, and only the next few seconds of video.

I wouldn't necessarily say I don't recommend this book to anyone, it's not exactly a bad book. The problem is that reading a book about the internet when you already know everything you want to know about the internet is just as interesting as reading a book immediately after listening to the radio show adapted straight from the book with almost no changes (which may or may not be something i've done in the past). All in all, the Tubes is a well written personal account of a man's journey to find out more about the internet. It's an especially interesting take, starting from an outlet on his wall and journeying further and further into the depths of the internet. It's a more interesting perspective than simply presenting the information he found with no embelishment whatsoever.

[bendoescomputers-blog]

Humans vs AI

Typically when I refer to AI, I use Marvel's Vision as an example. This time, I think I will refer to Futurama's Bender since I believe he represents a more perfect human AI. This may or may not have to do with the fact that we were binge watching Futurama in the lobby yesterday. In Futurama, Bender is a robot that, although his sole purpose is to bend things, displays all the traits of typical human intelligence. Over the course of the show, not only does Bender demonstrate a level of computational ability that far surpasses that of a human (even god-like at one point), but also a full range of emotions.

The real question, of course, is this: will we ever build an AI capable of the emotional and computational abilities of which we ourselves are capable of? I don't think so. For that to happen we would first have to understand how the human mind works, then understand how those processes could be applied to commputer programs. We don't entirely understand the human brain (or at least I don't). For example, how does our brain percieve consciousness and think about itself? We don't quite know yet, and even once we figure it out we have to transfer that from the signals in our brains into signals in a computer. Another example is emotion. We understand which areas control different emotions, but we don't really know what causes them. Human emotions often seem complex and even irrational to each other. If our emotions aren't "calculated" by our brains, then they can't be calculated by a computer very easily.

At this point, it looks like AI will be very interesting in the near, and far, future. Even if it will never reach the singularity point where AI is as capable as human intelligence, It will still be very capable. It is possible that computers can begin to take over a number of important jobs, but these will likely be jobs that most people tell their children to go to college so they can avoid them. Already, cars are capable of driving themselves safely as can be seen from Tesla's self driving software. Perhaps sometime in the next fifty to one hundred years entire bus networks and subway systems will be autonomous and perfectly safe. Already, some restaurants have computers that take and relay orders to the cooks. Maybe some day robots will bend pipes and metal beams for us.

[prostheticknowledge]

Physics Forests

Computer graphics research from ETH Zurich produces efficient and realistic fluid dydnamics simulations using machine learning (and kinda unintentionally creates Net Art in the process a la @fruitsoftheweb :

Traditional fluid simulations require large computational resources even for an average sized scene with the main bottleneck being a very small time step size, required to guarantee the stability of the solution. Despite a large progress in parallel computing and efficient algorithms for pressure computation in the recent years, realtime fluid simulations have been possible only under very restricted conditions. In this paper we propose a novel machine learning based approach, that formulates physics-based fluid simulation as a regression problem, estimating the acceleration of every particle for each frame. We designed a feature vector, directly modelling individual forces and constraints from the Navier-Stokes equations, giving the method strong generalization properties to reliably predict positions and velocities of particles in a large time step setting on yet unseen test videos. We used a regression forest to approximate the behaviour of particles observed in the large training set of simulations obtained using a traditional solver. Our GPU implementation led to a speed-up of one to three orders of magnitude compared to the state-of-the-art position-based fluid solver and runs in real-time for systems with up to 2 million particles.

More Here

[blurrypaths]

Amazing application of machine learning to fluid solvers.

[bendoescomputers-blog]

Privacy or Security?

As with many other issues, in the question of whether personal security or privacy I tend towards the center of the issue. I feel like the government should be able to collect a certain amount of data in order to protect it's citizens. At the same time I think citizens should be able to maintain their own sense of security and be able to do as they will without having to watch their backs in constant fear of Big Brother.

If a government can protect its citizens from threats that come from within its own borders, then they have an obligation to do so. Within reason this includes a duty to moniter their citizen's interactions for signs of a threat. Obviously this could easily be extrapolated into Orwell's vision of a society ruled entirely by 'thought police' seeking to weed out all citizens with even a hint of resentment in their minds, but this is not what I mean. In Orwell's 1984, the thought police seek out citizens who have so much as considered for a half of a second the thought of rebelling, then they lure them into swearing to do all they can to overthrow the government. The rebellious citizen is then reeducated and executed for his 'thought crime'. This is not my vision.

My vision is one in which a government collects data that is filtered through various computer programs looking for specific keywords. If the correspondences are found to have an abnormally high number of buzzwords then it would be sent to a human analyst, who knows nothing about the parties in correspondence, who may determine whether the messages represent an actual threat to the government. At this point if it is a threat it would necessitate further investigation and action. Otherwise, it would be scrapped along with other useless data. All data collected should be collected lawfully by the government rather than by covert and illicit deals with computer and software manufacturers, as is feared by a number of proponents of privacy.

To an extent, the loss of privacy for the sake of security is a natural extent of the 'Social Contract Theory' set forth by Jean-Jacques Rousseau to explain that in order for a government to provide all the benefits to its citizens as is possible (security) those same citizens must be willing to sacrifice certain luxaries (privacy).

[bendoescomputers-blog]

It's hard to name every way the internet has changed my life because I've had the internet my whole life. I do know that my life would be very different without it, and I can name a number of ways that it has changed my life though. When I left for a college two hours away from home, only one of my close friends came along and nearly all the others went to a school in my home town. In the past that would have meant I would only be able to talk to them and see them on those rare occasions when I can visit home and make time to see them, which hasn't even been once so far. Thankfully, thanks to the wonders of the internet and video messaging, though, I've been able to keep in touch very well and even video call a few of them. I won't say it's as if I never left, but It does a lot to ease the pain of seperation.

Additionally, The internet is an incredible and very accesible aid for school and learning. In addition to the wealth of basic information on wikipedia, I can also access professional resources like google scholar to find research articles. There are also free knowledge resources like Kahn Academy (the reason I passed calculus), and codeacademy (the reason I know how to code). Without the last two resources alone I probably wouldn't be in computer Science right now. Had I not had Kahn Academy I would not have passed calculus and thought, like many who got C's in that class, that I was bad at math. Without Codeacademy, I wouldn't have been able to learn about coding and wouldn't have maintained my interest in programming without having access to a place where I could learn the fundamentals, and a few advanced concepts, quickly. It's kind of incredible that the internet can influence anyone so much. If I didn't have the internet to teach me so many things I could be pursuing my other fascination, music, but even then it's unlikely. I first heard the song that made me interested in classical music and music theory on YouTube. In all honesty I have absolutely no idea where I'd be without the internet, probably in a van down by the river. Not that I'd get the joke without YouTube videos of Chris Farley on SNL.

[bendoescomputers-blog]

My computer science class recently had our group presentations, and today's blog post is dedicated to giving a review of one group's presentation. Fortunately, the group I was assigned to blog about was giving a presentation on a topic that interests me a lot: The Internet of Things. For those who don't know, the internet of things (TIOT) refers to a concept that may be implemented in the near future of having interconnected devices. This prbably isn't an entirely new idea to many people in Computer Science, but it's always fun to hear updates on it. The group presenting on TIOT did a great job presenting and holding my attention. A lot of their examples, such as the oven that asks if you want to leave it on when you are away or the fridge that reminds you to pick up some milk, were things I had heard of before, but there were some other implementations I either had not heard of or that I hadn't considered as part of TIOT. These things included a baby moniter that broadcasted via bluetooth to the owner's phone, or my smartwatch. These things were interesting and thought provoking in their own right even before they mentioned the potential for someone to implement something similiar in their house without waiting for the consumer market to catch up. It was a thought I had had before: teach a raspberry pi to tell a lot of arduino robots to open and close doors, or to turn lights on and off, but I had forgotten it until then. Perhaps when I graduate and have money again I can build a system like that.

The speech wasn't perfect, but that was mostly because TIOT is mostly hypothetical and a narrow field, there's not a lot of real things to dwell on. Fortunately that didn't stop the speech from reaching the minimum time, in fact, like a lot of groups before them, they were cut off by the clock before finishing their presentation. Up until that point they had only talked about the good things TIOT could bring us, but had only mentioned one con, and even that in passing. It would have been fun to learn more about the downs to TIOT beyond what I had heard, "Anything connected to the internet is hackable" and the typical "someone's storing that information and selling it to the government". As I said before though, the gaps in their information were more due to time constraints than lack of research as far as I could tell.

[bendoescomputers-blog]

After many years and many jokes about my identity as an android, the time has come for me to admit that I am not, in fact, a computer. Computers and humans are different in more than a few respects, the most obvious of which is the way humans and computers "think" (if that's what we call what humans do most of the time). While humans are capable of the same logic computers, humans are far more capable of emotion. Not only can a person feel for itself, but a person can also think original thoughts. Unlike humans, a person can think thoughts it wasn't told to think. At this point someone may begin to wonder what makes computers better than humans. it's simple, computers have an advantage in speed with regards to the ability to, well, compute. Sure, if you give a smart human a pencil, a piece of paper, and a lot of free time he could tell you the sin of every number between one and ninety. But it'll go a lot quicker if you ask a computer for those numbers. Similiarly, according to this XKCD it is possible for a bored human with a few rocks to simulate the universe, it is much easier for a computer. Another reason computers are still useful to people is that while computers used to be people who compute, I highly doubt people would want to do that job anymore. Even as much as I love math I feel like there are a plethora of jobs I would rather have than to sit at a table writing down a lot of calculations.

In an earlier post I alluded to a potential humanoid android such as Marvel's Vision. Androids like that do beg the question of whether the differences between computer and human can be blurred to non existence. Clearly the Vision is not a human because his skin is synthetic and he was programmed by his maker. But at the same time, he isn't necessarily a computer since he can think and feel outside of his programming. At what point do the capabilities of human and AI become so similiar that an AI can be considered human? at what point do humans get so detatched from emotion that they can be thought of as computers? Will we ever really need to answer these questions in our life times? I really have no clue.

[gardenvarietynerd]

me 50% of the time: programming is so fantastic look at all the neat and pretty things i can do like I DID A THING THAT WORKS

me the other 50% of the time: i fucking hate all my life decisions why did i choose this major and more importantly why is everything broken

[bendoescomputers-blog]

The Pattern on the Stones was a fascinating book that covered a great deal of the fundamentals of computer science. Going into my intro to comp sci class I knew virtually nothing about computers asside from how to program them to do difficult math problems, but this book taught me a number of things I had no clue how to do before. For example, I knew that boolean logic was the basis for if/else statements and I understood basic boolean logic from using academic search engines, Pattern on the Stones helped me understand how they're used for other computer operations and understand the logic of complex statements. I will confess that the writing style bothered me a bit, it seemed at times to hover between "you will understand this with no prior experience" and "google it if you don't get it". There were a few points I will admit I needed to google towards the beginning of the book, but I didn't actually look them up and ended up guessing meanings from context clues (my third grade teacher would be proud).

The most fascinating thing in the book was the explanation of how computers can be built out of anything with jerry rigged 'and' and 'or' gates, then made an example out of a tic tac toe playing computer made from tinker toys. As I said before I had known that boolean gates are used for If/ else statements, but I hadn't realised that they were used for everything in computer logic. This also helped me understand, in an abstract way, how information is stored in binary. This abstract understanding was, of course, brought into a complete understanding by the explanation in our class. I found the book interesting enough that, rather than give up when I fell behind, as in high school, I instead resolved to catch up as I read the next chapter.

I would certainly recomend this book to anyone looking for a better way to understand computer science than browsing wikipedia or looking for free resources online, as I did before this class. It is very engaging and can be helpful to anyone willing to think about the concepts and occasionally revert to wikipedia for more in-depth explanations. This is a good book for the casual learner or the intermediate who is looking for more information about a broad range of topics in computer science.

[bendoescomputers-blog]

When I was assigned to write about a computer system I deal with in my daily life, my first thought was to talk about my laptop. But that would be boring, so instead I'm going to talk about the crosswalk sign right outside my residence hall. Every day as I wait for it to tell me very aggresively that it's safe to cross, I contemplate the computer system running it. The cross sign is in the middle of the street, meaning that the only stoplight is triggered by the button press, but since there's a delay I know there must be some other factor that tells it when to turn on the sign. The most likely suspect is a timer that would allow a car going at the speed limit to cross if it were too close to stop safely. For awhile I believed this to be the case, but I soon realised that the wait wasn't always consistent. This leads me to believe there's a sensor hooked up to it some distance back that tells the light if there's a car coming. But there have been instances where I've crossed late at night and still had to wait a few seconds. So there's probably a timer to ensure that cars in front of the sensor can cross, then a timer that waits for the sensor to not be triggered, then if the timer reaches zero it overrides the sensor and just lets the now amassed cluster of students to finally get on their way to class or back to their hall.

This exact system is more or less new to me since arriving at college, but everywhere else I've been either doesn't require me to walk or hasn't updated their walkways. It is a system that became very clear to me on my way to school in the past few years since I had to cross at least one busy intersection no matter which route I took. It isn't an uncommon system system in terms of how widespread its use is, but it is something used almost exclusively for traffic. Some say that if there weren't traffic lights, the traffic would be intollerable and entirely unsafe, but I've also heard of areas with absolutely no signs or lights to regulate traffic that have far fewer accidents. Without it I think life may seem harder at first, but gradually at least get back to just as easy as it was.