Brian's World

Cornmaze

I can’t seem to end some problems without creating more. A very strange vicious cycle of sorts. I recognize that I must keep opening the doors, even if they all look unpleasent. I guess its a poor algorithm to apply… imagine yourself in a room with three doors, and which ever one you go through leads to a room with another three doors. Yes, the doors automatically close after you open them. No, you do not have any objects to use as markers. Without prior knowledge of the system, ones fastest strategy might be just to run as fast as possible, constantly choosing a random door to open hoping it holds a slightly more favorable outcome. Of course statiscally, with 3 choices, you’re not likely to get the right door very frequently. Tranversing tangential paths are never a good thing. I dunno, this strategy has worked out well for me in corn mazes. I find running, and choosing turns very quickly is a fast way to go, its never too long to reverse my route. In life I’m not sure how well this works though, I probably need to apply more thinking and less running. Yikes, even more thinking… I’ll be moving at a sloths path through this maze.

I am 85% sure what my replacement for the shoebox computer will be. Tomorrows task is to finalize this, and route the appropriate funding to wherever I need to route it. I officially deemed my Shuttle pc dead today. Its probably the motherboard, but I’m not in the mood to verify its not the processor. I’m considering selling it as-is on ebay, identifying that it was performing strangely in the past, but I haven’t decided to do this yet. If anyone out there wants it, let me know! I’d be willing to part with it for a very nominal fee (potentially free)… I could even throw in some RAM if you wanted. Why the sudden move to decide? Well the dell I’m writing you from starting throwing one of its fits again. It has a bad hard drive, and when it hits a certain part a crash becomes rapidly eminent. I am faced with a challenge to temporarily store ~500gb of data while I reformat my raid array… or de-raid it for that matter. I am unsure where I will store that data.

I know nobody cares, but I pulled together this quick table of IPs that have been automatically identified as hacking into nsb1.info. If you happen to own any of those IPs and are unsure why you are on the list (i.e you’re not a hacker, and your wifi borrowing neighbor isn’t), then I suggest you scan for spyware and stuff. Maybe you’re computer is hanging out with the wrong crowd and you don’t even know it! I would call that a botnet.

Sometimes I get slightly frustrated at circular logic. People can state point A, and have reasoning B to back that up. I might inquire about the basis of B only to find its built on the identification of A as true. I don’t know which came first, the chicken or the egg, but it sounds a lot like A spawned B and then became both dependent on and a dependent of it.

As they say in Belgium, Bon Soir

Applicational

I’ve never liked it when I read stuff and get depressed. Today I was reading accounts of people’s experiences applying for internships at Google/Microsoft/Yahoo. I was bored at work, and felt like doing some research. I might apply to intern at a company like those three at some point like next summer, but I’m not sure yet… and today’s account sure wasn’t a booster. I read descriptions of interviews and stuff where people are asked to develop and algoritm for x or some code for problem y and I sit here trying to figure out what exactly the problem statement is. Now maybe my sources (generally blogs), weren’t as clear as the interviewer was. I know I don’t have the strongest grasp on the science part of computer science, but I think I have a fairly strong understanding of the computer part. If I’m given a problem that might exist, in the context of a scenario, I could probably generate a fairly effecient solution. If you tell me to design a program to compute <some term I don’t know> then I will probably do poorly at it. If you tell me you need me to design a program to track some statistics for words in a file, I might be able to do that. I’ve always placed heavy value on the applicational side of computers. Yes, theory is very important, but its what you can do with it thats most important. Maybe this relates to why I didn’t excel in DSA as much. Sure, a slow algorithm can be a problem; but it can be improved from several ends.. not just the source code. Hardware, operating conditions, input, etc all play factors.

Katie got her Macbook today. It was pretty neat to set it up with her. The worst part of the entire process was the hours of updates. I guess Windows machines are pretty much the same in that respect. Only in Linux do you get the ability to install the freshest copy of files during initial installation. I can’t say its had a huge impact on what machine I’ll be getting next. I’m going to try Openfiler on the machine with a bad motherboard. I can’t isolate specifically whats wrong, but Windows XP triggers it more frequently than Linux. Once I get my 2 external hard drives in a multiplatform filesystem, I should be ok if the machine completely dies on me. That conversion will be tricky though, right now they are Windows Dynamic Disks, configured for software RAID 0. Of course Linux understands none of this, so there will be some trick involved in copying >500gb of data into a temporary storage location that doesn’t exist so I can format and start again. I’ll try and update as the process goes.

My cell phone frustration has been growing at an exponential race. My Verizon Motorola v325 which is 2 years old is pretty much junk, whereas the battery is the most dense location of junk. I think the battery is ~95% scrap metal, and 5% actual charge holding stuff. I was barely able to send/recieve 20 text messages, and place a 2 minute phone call today. Then the “low battery” beep that is impossible to shut off. That constant beep really worked well with my desk in the library at work. I ended up shutting it down. My family is up for replacements, but I’m unsure where to go. I’m interested in getting a phone that can do more, I think. I see 3 different phone options really. Get a Windows Mobile Smartphone, get an iPhone, or get something cheap to hold me over the 6-9 months it will take for Andriod to hit production. If I was alone in this I would be much more apt to move quickly, but I’ve got this family plan attached to me. Its a saver on the monthly bills, but makes switching phones and everything a more challenging process (since 1 user doesn’t learn new phones quickly). I’m going to try and hold off 11 days until iPhone 2 ships and then I’ll make a move. I’m not sure what effect it will have on Verizon Wireless, or the phone industry as a whole. [No Verizon, I do not want you to try and push a Voyager on me.]

My boss is out from work until Thursday, or so the employee status program indicates. I planned on him being out part of today, and I had a few hours of work to do. I got done with that task in 2-3 hours, so I’m not sure what I’ll do to occupy my time tomorrow and Wednesday. There are a few neat ideas I have, but I’m hesitant to go too far with them knowing that a deployment in the timeframe I’d be there might not be possible.

Mom needs to buy mayonaise so I can make better sandwiches for lunch. She keeps reminding me she has bough luncheon meat (ham and turkey), but she hasn’t bought the mayo to hold the sandwich together. Looks like another PB & J is in the future.

Good night moon.

Pre-Alpha Feedback Team Group

This week has been pretty standard for me. I can’t say I’ve done anything uber exciting but I have managed to keep productivity relatively high. I just figured out to VPN off my Ubuntu box. Well, re figured out how, set it up so long ago I forgot how to connect. THe command is ‘vpnc rpiex’. I’ve found that if I configure my proxy settings correctly at work, I can SSH and FTP to various machines, so I keep busy during my lunch break working on what little code I can. All pretty boring stuff.

The team over at BriSpace.net has been working on a pretty neat project, and is looking for some testers to test stuff out. The project is kind of like a multiuser powerpoint/digital signage kinda thing (No, not Concerto). Essentially you need to be capable of working a computer, surfing the internet, and willing to provide feedback about your experience. The official signup process is pretty easy, comment below including your email address or shoot me an email (bmichalski@gmail.com). No, this isn’t a paid position, and you shouldn’t feel obligated to sign up. I could just use some help from a user interface standpoint. I have a pretty strong idea what the data is, and the basic components that need to exist, but the positioning and stuff are not my forte. I am not going to find the special e for forte.

I did print out the YUI earlier in the week. Its my favorite “javascript library” if thats what you call it. I enjoy the strong documentation and examples they provide, as well as how easy it is to use. Skinning it may be another challenge, but I’ll cross that bridge [likely burning it in the process] when I get there.

Oh yes, now I recall. Last night I saw the “Love Guru” in theaters. It was 85% terrible. There were lots of inappropriate jokes throughout the movie. It seemed the love guru wasn’t actually interested in helping people form a loving relationship, he was simply interested in helping people partake in an action verb form of love. I dunno, I don’t think I’ve ever been a huge fan of Mike Myers. I’ve avoided the Austin Powers series pretty effectively during my Middle and High School career… I just don’t usually enjoy that kind of humour I guess. I do look forward to seeing “Get Smart”. I hope it is half as good at the TV show its based on. Maybe when my cell phone dies, I will work to embed it in my shoe. Not quite sure what that would achieve except a spot on the terrorism watchlist, but I guess is a problem to be addressed later.

I don’t like it when my mind gets stuck doing crosstab queries, cubes are really the way to go. They are harder to digest, but you literally gain a whole new dimension to the data. Even then, I do tend to come to the same result.. usually just requiring a few more computations. The main problem isn’t really the dataset forming whatever model I want, but really the comparison between them. Lets say I have a datacube about object A, but all I have about object B is barely one dimensional, rarely two. Yes, I do my best to fill in the gaps but I don’t recieve the necessary feedback to determine how accurate my assumptions are, and an analysis conducted on assumptions that haven’t been tested (nevermind verified) isn’t worth the power to compute it. Energy is getting expense I guess, but that is most certainly not what I was trying to illude to.

Back to vim.