Page Faults

Biology is on the verge of rediscovering the calculus...

2011-02-04T12:52:00.000-08:00

Sometimes you love a paper, and sometimes you hate it. Today's paper is neither, but it will offer a glimpse as to why I sometimes deride less mathematically rigorous disciplines.

Here's a paper that pats itself on the back for using the trapezoidal method of integral calculation. And gets published. And cited like it's nobody's business. Wow.

abstract and paper

Monkey Consumers

2011-01-10T15:25:00.000-08:00

In this paper, the researchers successfully introduce the concept of money to a group of monkeys, and run several experiments. Turns out that our concepts of supply/demand and loss aversion are not limited to humans...

PDF

Worm Hunters

2011-01-05T21:51:00.000-08:00

This first paper is quite possibly my favorite. If you're a math nerd, a computer networking enthusiast, or just a fan of general badasses, I promise you'll appreciate this one. From a handful of packets these guys analyzed an internet virus, and could construct the entire infection tree, identify which of the victims' disks were corrupted, who started the worm.

Exploiting Underlying Structure for Detailed Reconstruction of an Internet-scale Event

Kumar, Paxson, and Weaver

Abstract

In this work we apply such an analysis to the propagation of the Witty worm, a malicious and well-engineered worm that when released in March 2004 infected more than 12,000 hosts worldwide in 75 minutes. We show that by carefully exploiting the structure of the worm, especially its pseudo-random number generation, from limited and imperfect telescope data we can with high ﬁdelity: extract the individual rate at which each infectee injected packets into the network prior to loss; correct distortions in the telescope data due to the worm’s volume overwhelming the monitor; reveal the worm’s inability to fully reach all of its potential victims; determine the number of disks attached to each infected machine; compute when each infectee was last booted, to sub-second accuracy; explore the “who infected whom” infection tree; uncover that the worm speciﬁcally targeted hosts at a US military base; and pinpoint Patient Zero, the initial point of infection, i.e., the IP address of the system the attacker used to

unleash Witty.

PDF

Change in Focus

2011-01-05T21:43:00.001-08:00

Lately I've been reading a lot of articles and books that summarize cool research (think Freakonomics). This blog could use a new focus to recapture some of its author's passion, so I've decided to focus for the time being on the "greatest hits" of all the articles I've read. I'll give a short summary of why it's cool and provide a link to the original research (yay for Google Scholar) for each white paper that I choose. Enjoy!

College -> Real World

2010-07-07T18:12:00.000-07:00

Well, long story short, I've completed 18 years of school and am now off to the real world. I graduated from Stanford three weeks ago, and I start work on Monday.

So what have I been up to in the interim you ask?

Apartment

I've moved into an apartment near downtown Palo Alto. The location is pretty nice, and the rent is not too bad. Here are some pictures:

Finances

I've also started paying off some students loans. I have roughly $20,000 that needs to be returned at some point. Even though no payments are due until December, the juice is running (yikes!), so I'm trying to pay down the principle as much as I can. My goal is to get it all done by Jan 1, 2012.

I've also begun putting rungs down for a "cd ladder." A cd ladder works as follows (I'll use 12-month cds as an example): every month for 12 months, you put $x.xx in a 12 month cd. When each matures, you simply open another 12 month cd with those funds.

A 12 month cd gets a better rate than shorter term certificates of deposit, but you have to promise not to touch the money for a longer period of time (like, you know, 12 months). Anyways, once you've done this for 12 months you have a cd maturing every month but at a higher interest rate than if you had a 1 month cd. It's a nice way to set up an emergency fund that can still draw good interest. Ally.com gives some pretty solid rates (imo), plus they have some bizarre vehicles where you can withdraw early without penalty, so I've decided they've earned the honor of becoming my bank of choice.

Random Web Stuff

Since I'm going to start work shortly for a company that's on the web, I figured I would brush up on some html/css/php/mysql stuff. So I've been working on some short little projects to try to keep my brain from atrophying.

Well, that's what I've been up to. Someday I'll get the hang of this blogging thing and post more than just updates.

May Projects

2010-05-02T23:24:00.000-07:00

My past lamentations over boredom are about to be resolved... with a vengeance. This month I have not one, but two, major projects to complete for my spring quarter classes. They're pretty cool, though, so I thought I'd give a brief overview.

CS315a - Parallel Computer Architecture and Programming

For this class, I'll be working with two others to code and parallelize an Item Response Theory based program. Props go out to Andy for introducing me to the idea, and for pointing me towards several white papers to get us started.

CS343 - Advanced Topics in Compilers

For this class, I'll be working with one partner to create a test-case generator. I'm pretty excited about this project, as it allows me to get my hands dirty inside a compiler and could prove to be pretty useful (hopefully). The idea is pretty simple. Let's say you have the following block of code:

foo(int a, int b) {

if( a < 10 )

//do something

} else {

if( b != 0)

//meant to do something, but instead causes a runtime crash

else

//do something

}

My program would create three input test cases: foo(1,0); foo(11,1); foo(11,0); These can then be run by the user, knowing that they likely get X% path coverage on their code. It should be pretty sweet.

Computer Documentaries

2009-08-23T13:38:00.001-07:00

Here are a couple of really cool computer documentaries. I personally find the history of the computer really interesting, so naturally I found these two documentaries to be fun to watch.

1. The Machine that Changed the World. This is one of my favorite documentaries of all time. It's really interesting, and they find a lot of interesting people to interview. Similar to the second of these two documentaries, they made this at a time when the Internet was very infantile, so their predictions and insights are hilariously way off.

2. Nerds 2.0.1: A Brief History of the Internet. This documentary provides some interesting tidbits about the ARPAnet, various protocols, and other technical features, but overall I think context of when it was filmed is what gives it real value. This was made in 1998, when the Internet as we know it was barely three years old. Everyone still seems very unsure of where the Internet will go and the documentary is enfused with the detail-bereft optimism that the Internet was an untapped moneywell. With over ten years passed since the filming, it's also interesting to look at which companies they profiled. The internet was still an electronic wild, wild west, and the .com bubble was still a few years away. These companies are the ones that made millions in their IPO, and of course became nothing after the crash. Guess they should've profiled the brand new company known as Google :).

Pwn Beast Tom Scholz

2009-08-20T21:33:00.000-07:00

After a discussion with my carpool partner, I've become fascinated with the life of Tom Scholz. I figured it's worth a cool, yet short, blog post.

Scholz earned a bachelor's and master's degree from MIT in mechanical engineering. After that, he worked as a product design engineer for Polaroid and played some guitar in his free time. In addition to playing, he also tinkered with the recording equipment. One such piece was an amplifier that he would later sell called the Rockman Guitar Amplifier (it would sell big).

Using these pieces of equipment, he recorded a couple of random demos, and signed with a record label to produce it. You probably know the album as 'Boston.' You know, the one that was the fastest selling debut album in history?

How cool is that? This is like the inventor of the typewriter deciding to show off his new contraption by penning The DaVinci Code. Absolutely incredible.

Populism vs. Baseball

2009-08-20T18:11:00.000-07:00

The usual populist rage is against corporations and people who get paid too much money by them. AIG's parties are too lavish; Merrill Lynch's bonuses are too big; insurance companies earn too much profit. Whether justified or not, these are very common arguments nowadays. Any company, person, or industry who makes a lot of money should expect at least some outcry over their finances.

Baseball presents a paradox, though. Come this time of year, instead of chewing out the greedy teams, or over-paid players, we love to hate on the people that represent the players. Bringers of all things unholy, the agents fight to make sure the best players get the most money. Well, if agents are evil satanists, then that makes Scott Boras look like this guy:

Hey, at least he doesn't eat monkey brains. Boras grabs more money for his players than any agent in the game, and his work has fundamentally changed the economics of baseball. If you're wondering what kind of magic he can work, just try picturing whether you can trick a team into paying Barry Zito $126 million to be a 5th starter. Uh yeah, that's what I thought.

Boras really draws the ire of the crowds, though. Why is that?

Back in the day, baseball was dominated by the owners. Just like in The Natural, or Major League for that matter, the man who owned the team owned the players, and he was not necessarily a nice guy.

Enter the Major League Baseball Players' Association. At the time of their formation (1965), the minimum wage for baseball players was $6,000 a season. If only paid for being at the ballpark on gamedays (excluding weight training and practice time), and assuming a 3 hour game with an hour for warmup and an hour for shower/interviews, the total time spent in a season is 162 games * 5 hrs/game = 810 hours. That comes out to $7.40 an hour, which even back then is not a lot, given that you "work" under 40 hours a week. Since the formation of the players' union, though, everything about baseball has gotten bigger. Revenue, attendance, merchandise, and salaries have all sky-rocketed. And the players have been able to reap the most benefit out of this growth.

Now, I'll try to come back to my original point. When an industry grows, be it from more sales or increased efficiency, there are several options as to what you can do with the money. Some pay off the top owners and investors. Others reinvest the money back in the business. Still others raise the salaries of employees. All are legal, yet only the last really catches the heart of protesters and union supporters. They love to fight for the little guy, and there's definitely some noble qualities there. This is not a criticism of those beliefs, just an errort to reconcile them with baseball's market. Shouldn't agents, who in effect are just trying to get as much money as possible for an industry's basic employees, be supported by the same protesters?

So that's why I find it so unusual that people flip out over Walmart squeezing stock boys, GM firing employees, or AIG throwing parties yet still lose their cool when agents fight for their players. One can argue that of any industry in the world, the one where the employees make the highest percentage of income is baseball. Baseball is an arena where the union won, and they won big. Yet we hate the people who fight for their right to 'fair' compensation.

I guess this provides some interesting psychological tidbits. For one, we love our team, not just the players on it. A player asking for more money is in effect weakening our team. Yet it's not the players' fault, somehow. We recognize that he's talented and deserves a raise. It's that damn agent of his! By having an agent, the player removes himself by one degree from the greed and the ire of the public.

In this way, we identify most with the owners or general managers. Perhaps if Merryl Lynch wants to get public opinion on their side, they should start a baseball team (and let allow their employees to get representation).

The Walmart Effect

2009-08-11T20:27:00.001-07:00

On my most recent plane trip, I finished a book by Charles Fishman called The Wal-mart Effect. The aftertaste of the book was very similar to watching Supersize Me. Every time I watch that movie, I get a huge craving for McDonalds, which is really strange because I seem to only eat there once or twice a year. I really felt like shopping at Wal-mart after reading about its operations, despite what I've learned about it. Overall, I really enjoyed reading it for a couple reasons:

1) It's very even handed. Lots of vitriol has been spewed over Wal-mart, but the author takes a very fair look at every controversial topic in the companies history. He paints thrift as a virtue and a curse; he equally covers inventors who hit it big with Wal-mart and others who got squeezed into bankruptcy trying to lower costs; he treads evenly between legal and ethical responsibility. All of it lends a lot of ethos to his arguments, and makes the reading more enjoyable than if he had preached from a pulpit instead.

2) It's interesting. The author traces all the problems of Wal-mart to its goal to deliver the lowest price possible, and handing that price to the consumer with only taking a small (fractions of a cent per item) margin for itself. Initially a virtue that we respect and appreciate, he shows it can lead to negative consequences for its suppliers, customers, and the community. He takes on a massive project (trying to describe the many relationships between all the people in the Wal-mart web) but he makes it engaging for the reader, which is of course a plus.

3) He's a good writer. He found a lot of employees, ex-employees of suppliers, CEO's, inventors, and other people to interview. This is a tough task, given how notoriously quiet Wal-mart is when it comes to talking to the media. He also writes great descriptions of the products he opens his chapters with.

Overall, I'm not sure how this book will effect me. I'm not a big shopper, but when I do shop I usually do so cheaply (as many of you know). But Wal-mart is kind of out of my way in terms of food and stuff, so I don't really go there that often. I'm unsure if I'll feel guilty or not for shopping there. They make everyone in their ecosystem more efficient, which is good, but I know that it comes at a price. I can handle the outsourcing of jobs, but I'm not sure how cool I am with taking advantage of other countries' lax labor laws. Anyways, it's something for me to think about the next time I enter the big box.

Predictably Irrational

2009-07-30T21:38:00.000-07:00

I just finished reading a really interesting book called Predictably Irrational. It's a nonfiction hardcover about behavioral economics. Behavioral economics as a field is kind of what you would expect if you took psychology and yelled at it until it became more legitimate. Sorry, but I've never really respected Psychology as a science. A big part of this is because one afternoon I was walking behind a pair of psych students as they spoke about color perception. One touted that there would be no way of knowing if one person really saw green as blue, but called it green, or something like that. The other nodded the same hippie nod seen in countless Comparative Lit. majors. I instantly thought of a simple experiment where you set up a picture with a known wavelength of light, and ask the people to identify it... but I digress.

Back to the book. The book's central thesis is that the field of economics is based on an incorrect assumption: that we are rational beings. Dan Ariely, the author, contends that we are anything but. And, as the title would suggest, this irrational behavior happens in a very predictable manner. What does he mean by this, you ask?

I think the book is best served through an example of his studies. Let's say you have to decide between two European trips. One is a trip to Paris, with free breakfast at the hotel, and the other is a trip to Rome, with complementary beverages. These two present a tough choice to most people, and in studies they are chosen at a very similar clip. The interesting thing is that when a third choice (Paris with no breakfast) is entered these percentages change. The rational person's choice (a 50/50 split between the two) should be unaffected by the introduction of a third option that is obviously inferior to the other two. The thing is, though, that when the lesser Paris option is present, Paris with the free breakfast becomes a very popular choice. Why is that? Well, the idea behind this one is that people tie value to an object by its ability to compare. In this case, Paris and Rome are both excellent cities, and can be very tough to compare, so it's a coin flip between the two. But when a lesser Paris is introduced, the Paris with eggs/bacon gains extra value because we can compare it against another option very easily (even if the alternative is obviously crappy).

The book is full of random, interesting dissections of mankind's decision making skills, such as the above example. If you found it interesting, then I highly recommend that you read the book.

Computer Summary

2009-07-19T20:26:00.000-07:00

I just finished purchasing a monitor on eBay, and with that bid, every component is now purchased and in my possession or making the scary trek through the postal system.

Here's my system, with all the specs:

Apevia X-Discovery-AL Silver Steel ATX Mid Tower Computer Case
AMD Athlon 64 X2 Dual-Core Processor 4400+

Pioneer Internal DVD DVD/CD Writer

Western Digital 300GB Hard Drive

Microsoft Basic Black Keyboard & Mouse

Gigabyte (company name) Motherboard

2 x 2GB Memory

IBM ThinkVision L170 17" TFT LCD Flat Computer Monitor

Ubuntu 9.04

And the prices:
Case/PSU= $70
CPU=$44.99
DVD/CD Drive=$34.99
Hard Drive=$31.00
Cables=$3.95
Keyboard/Mouse=$15.99
Motherboard=$79.99

Memory=$49.98
Monitor=$70
OS=Free
----------
$409.89

I'm pretty happy (now) with the cost for something I'm just going to screw around with, but we'll see how frustrated I get if I find out the computer sucks. It's looking like I'll spend next Sunday putting the pieces together, assuming I have all the pieces by then. Sunday is also crucial because I want to be able to return these things if they're dead; the sooner the better for returns (15 days from purchase).

Either way, I'm really excited all the pieces have come together.

Also, the keyboard/mouse combo came as a three-pack, so if I can sell the other two over eBay, perhaps I'll cut the cost under $400. Cool!