timuralp

Really? Harvard? You need to pick carefully what field you're interested in. Most schools are good in some subfields and bad at others. I know that first hand about UIUC - great school rankings wise, but not very good when it comes to the field I'm in. This is true about every school - no school is great at everything. That is true. In fact, it has so little impact that a number of top schools don't even ask for it. The verbal score in particular has a pretty low bar. The quant part is easy and you really should get something above 700. No one cares if you get 800 on it or not. 1. Shouldn't. There are people coming back to school in their 30s and 40s. I wouldn't worry about it. 2. You should. It is true that many (most?) grad students in CS apply without having any inkling about which area they're interested in, but it is useful to have some idea. At the very least you should identify a couple of fields you're interested and pick schools who have active research in the area with multiple professors. 3. Not very beneficial from what I heard from other professors. May be useful if you're pursuing Theory or Econ related things.

Don't fall in that trap. GRE is not that important. Also, Stanford is notorious for admitting students with no funding. If you're looking for a PhD, you shouldn't take that. Finally, the statement of purpose advice I like is this: http://prisms.cs.umass.edu/mcorner/cs-admissions. It's also targeted for PhD applications and international students, but is useful in general. Finally, a publication is great, but just doing research work may be enough, assuming you work on a good project with a good professor. If you want to do research, you should be doing research Also, I have no idea what your area of interest is and the schools you list are all over the place. A few of those I wouldn't apply to, but that's for my field.

That is very true. I actually always wore shorts to everything (weather permitting). I recommend a polo shirt for the spiffier look, but a button down shirt is a little too far. At least make sure it's not tucked in

This is what I had to learn. According to my advisor B+ is the optimal grade, unless it's one of the courses in my area, so if you're doing research, this may be useful advice.

Good for you. Now if only GREs counted for much In all honesty, it sounds like bragging more than anything else.

Exactly. Also, I forgot to give a link to this before and maybe you've seen it already, but grad school is more like a kindergaten. And phd comics become actually depressing and not funny once you realize they're true.

Yea... I pretty much skipped my orientation - both of them. They also turned out to be kinda unhelpful, but that could just be a function of my department/grad school. Anyway, don't worry about it - grad school is not a competition, no matter how much some want to make it into one. As for talking to people, I found that asking questions always worked for me when I was in an awkward setting. (Most) People love to talk about themselves and you can learn a lot about who you want to be friends with in the process.

No: http://en.wikipedia.org/wiki/List_of_United_States_mobile_phone_companies With the sim card, I would then just look at T-Mobile, because I used their prepaid card with the G1 dev phone, a windows mobile phone, and an old nokia phone. All of those worked fine.

But... when you setup pay as you go, you just buy a sim card and throw it in. I'm not sure how/why they would know/care you're using an iPhone vs any other phone.

That doesn't really matter. Anywho, here's the map of some Austin hotspots: http://hotspotr.com/wifi/map/14-austin-tx Keep in mind this is not a full map and there's probably a bunch of open APs setup by random people. Also, Starbucks now has free wi-fi as well, which should help. Personally, I find that in a city, like Austin, I'd be able to find some sort of open wifi within a few blocks of where I am. With the providers, I'd ask other people around you what their experience is with coverage. Some are pretty bad in a certain locations. The prepaid card idea is good. I bought one from T-Mobile before and it was pretty cheap and convenient.

That's the most common way of doing it that I've dealt with. My girlfriend recently moved to another city and that's exactly how it was handled because she couldn't come until after the lease started.

I sort of do? I don't really care, I guess. The team I cheer for is in the same division, so I definitely don't cheer for the grad school team when the two meet.

Sure. I was just commenting on what you said, that's all

Well, you could get core 2 at the same clock frequencies (2.4 and 2.66, as the two comparison points), so that statement was just awfully misleading. This comes from the "GHz" confusion above and the point I was harping on was this: "The 15 inch uses newer technology which will allow you to use more programs without any noticeable slow downs." Since at that point it sounded like you're talking about higher frequency. Anyway, my point was that even with more threads of execution certain I/O intensive workloads will not benefit since the bottleneck is elsewhere. I have no idea what the person is planning on doing. I was only saying there is a difference and then I did say that you're write in that it probably doesn't matter. Not sure how that got misinterpreted. And that's why it was a suggestion. If this person figures out how to do it from numerous how-tos online or asks a friend, more power to her/him. Especially, parts like memory and hard drive, since they are easy to install. If not, that advice can go on ignored - not twisting anyone's arm to do it. Did not expect this to turn into a big debate of some sort.

This is not entirely true. It can overclock the processor for periods of time to ~500 MHz higher, but that doesn't necessarily mean no slow downs when using multiple applications. For some workloads this could be true, but for others (think intensive I/O), it will not have much of an effect. Sorry, I just have a pet peeve about the "higher clock speed = better performance in general" statements It also doesn't have more cores or hyperthreading, so context switches will still be common with more than two processes going on (multiple applications) and just a bit less expensive. The shared memory (13") vs on-board memory (15") is a significant difference. However, it probably doesn't matter for most workloads, as you already said. By the way, a lot of the Apple accessories could also be purchased somewhere else cheaper, like a bigger hard drive, memory, etc.

I do wish it wasn't written in Java, however. Any thoughts of porting it to Python? Also, it's not free as in speech. Actually, why isn't open source?

The problem of extracting the right information out of PDFs is actually pretty hard because a lot of PDFs have nonsense set in metadata and parsing the thing is a pain. One way to do it is by looking up a DOI and finding the right citation (see the link I posted before and a number of articles have no DOI specified), but the only other way I could figure out was to write or use something to search for the text online and try to grab a citation there. I'd be curious if anyone knows of a piece of software that can actually that data out well. So far I've been tinkering with pyPdf and it looks like I might be able to pull out the names of some papers and authors but on some files it doesn't parse correctly

As far as I know, a number of zipcar locations are in garages. You could check that online... Also, with "maintenance" in the winter from corrosion, the only really important thing is to wash the car on regular basis (once in a couple of weeks). Other than that, the only thing that'll happen is if the car's already rusting, it may rust faster. My car is 12 years old and spent 6 of those in New England and 3 in Michigan. It's still in one piece, but it does have some rust problems around the doors, under the plastic.

If you guys want to organize a folder/look at how to deal with pdfs in a programmable way, here's a cool post for dealing with this in python: http://en.dogeno.us/2010/02/release-a-python-script-for-organizing-scientific-papers-pyrenamepdf-py/ The idea is to try to find the DOI and then look-up the information about the document. I'm trying to write something that would try to grab the title as the first line. At least that's the way a lot of CS articles are formatted.

Well, the two paths are really different (PhD vs professional). Either way, I would look at the BS requirements and if they're not that overwhelming, just do the BS and be on good footing for jobs, as well as schools.

What's the goal here though? If you're interested in jobs related to CS, then why not focus on a BS in CS first? I know of a number of companies who would happily hire you with a BS and pay for a part time (and in some cases, full time) MS degree in CS. Also, the CS GRE (unless was updated very recently) is: dated, does not cover all the topics you should know about in depth, focuses on things that do not help in the general proficiency in the field. As such, a number of schools don't care about it (unless you have no CS background, since then it's impossible to evaluate you). Personally, I would pursue a BS, then look for a job that pays for part time school expenses and do a part time MS if you really want to do it. If you want to pursue a PhD, then it's a whole different story and I'd go about it a whole different way, but it sounds like you're more interested in the professional degree.

At two of the schools I attended, access was done by IP, i.e. if you access through the university network, you don't even need to log in and just look at whatever you need. Another way, is libraries made journals accessible through their interface online too. Once again, that just required being on the university network. One way to access it off-campus, is to use a VPN. I'm sure your school supports that feature.

I did 3 years of research on 4 of different projects, which really didn't affect my admissions chances, as far as I can tell. I ended up getting into a top 5 school, a few top 10 schools, and had plenty of choice. It actually helped, as far as I can tell, since I could get good letters from a few professors. Anyway, about working on 2 projects simultaneously, one professor when asked said this: "You know what a futon is? It's neither a good couch nor a good a good bed. You know what an El Camino is? It's neither a good truck nor a good car. You know what a spork is?" That's where he stopped as everyone was laughing, but either way, it may not be a good idea and could be better to just focus on one thing.

It depends on the school - something you could easily check on the admission requirements page. At the time I was applying, only UC Berkeley required it.

I bank with KeyBank and had no problems. They have another promotion now, where if you open an account, setup direct deposit, and do some online payments, you can get an iPod touch, I think. A lot of people I know like Chase here, and others use BoA. In my opinion, all the options are pretty comparable. With buses, you can check http://mbus.pts.umich.edu/ for when it's going to get to your stop. Altogether, they also run often enough. And as far as winter clothes, there was a big thread somewhere about what to wear in the winter. People's suggestions range from packing heat packs (this was in NYC, but sounds like they were preparing for Antarctica) to just putting on a jacket and gloves. My attire in the winter only changes to flannel lined jeans, sneakers, and a jacket. If it's really windy, I'll wear gloves, and I prefer a headband to a hat. I don't think you need snow shoes or even boots, but I guess others play in the snow all day?