timuralp

To do that at shutdown, the file system has to be scanned to find all the fragmented files -- that is an expensive operation. Then it's not just moving files around - it's finding an appropriate space for the file and copying all the fragments. It doesn't seem like that actually happens, but there is a nifty trick in HFS+. I managed to dig up the reference that seems to be what you're referring to: So, HFS+ can indeed ensure small files are not fragmented, but not at shut down. Seems like large files, especially ones that are frequently appended to, can get fragmented anyway: http://reviews.cnet.com/8301-13727_7-10328197-263.html I think in the end how much fragmentation occurs on the disk and how high the benefit will be from placing the files in a contiguous space will highly vary by user.

Err, I made a significant mental typo in the last post. I meant the charge level in storage and not capacity. Li-Ion batteries do have circuitry to prevent overcharging. However, this only ensures the battery does not swell up due to overcharging or catches fire. For storing the battery, however, the charge level is important. The difference in degradation over time between keeping a battery at 100% and 40% is discussed on the same site: http://batteryuniversity.com/learn/article/how_to_store_batteries (that's the source in the wiki article). This may vary depending on what people do with laptops, but I find that most of the time mine is plugged into an AC power source, so it makes sense to discharge the battery to about 40% and remove it and use it when I plan on not being around AC power for a while (flights, experiments in the city, etc).

From what I can find (see prior link), the two factors affecting the capacity change over time is the average capacity level and temperature. It's really about keeping it close to 40%, when not in use, it appears. Taking it out and putting it back in is definitely more of a pain with some models than others. I guess one could also plug in and unplug the cable every 20 minutes but that sounds even more irritating, and I'm not sure if it would be as beneficial as there is periodic load on the battery. So... how/why does it do that? Apple certainly claims it does not happen: http://support.apple.com/kb/ht1375 Are you referring to caching writes? That's a bit different from defragmentation. Moving files around before shutdown also seems like it could be a high-intensity, long process. I'm very skeptical that it actually takes place. Where did you learn about it?

Are you sure the age is what impacts the "swelling"? Can you find any sources confirming this? One explanation I've found mentioned, which seems plausible, is overcharging the battery and I've found one article in the Journal of Power Sources that seems to corroborate that claim. I don't think it's just the number of charge/discharge cycles the battery has gone though. By the same token, if all the user does is writing papers, making presentations, and playing music, Win 7 is also more than adequate...

I know MITRE and BBN (Raetheon) require the employees to be citizens. I don't know of any private companies (non-defense contractors) that offer this option, but they might exist.

Well, WiFi is a collection of standards that describe wireless interaction. Wireless router is a device, which allows multiple machines to connect to the Internet connection (wired and wirelessly), while providing some additional functionality, such as a firewall. The most common setup I've seen is to get a cable modem (on one end the wire from the cable company plugs in and on the other an Ethernet port) and then connect it to a wireless router. This may be changing and some of the modems, like the one you've found, have integrated wireless access points. Bottom line is that it will probably work fine. I would be cautious about "nobody can break the encryption", though. Someone with enough time on their hands and dedication could probably find a way to get the password, but the bar is raised high enough that most people would give up. P.S. Sadly, I made a typo in the Schneier link. If anyone cares to read it, remove the period (.) at the end.

While you could get a modem with WiFi, I think the most common solution is to get a wireless router (2 seconds of googling alleges that it's also a cheaper option, but then again... 2 seconds). The open vs secure wireless network debate has been going on for a while. I think the comments on Schneier's blog illustrate it pretty well - http://www.schneier.com/blog/archives/2011/04/security_risks_7.html. The argument for open wireless is that if someone breaks the encryption that the owner setup and gains access to the network (and is somewhat intelligent about it), there could be a load of circumstantial evidence pointing finger at the owner, while the signs that a third party accessed the information would be scarce or non-existent. In that case, it'd be even harder to prove that the owner did not break the law, even if the files are not found on his/her device.

Why are you talking about Psychology PhD programs? Anyway, the best bet is finding employers that would benefit from employees earning a PhD. One example is MIT Lincoln Labs. Allegedly, one can get paid most of the regular salary, while working on a PhD from MIT or Tufts. It's definitely hard to find something like that and even harder to get, but seems like the best shot.

Signing up for the promotional rate with comcast is definitely the way to go and then keep calling back every time it runs out. Currently, I'm paying $25/month for the Internet connection and $30/month for cable supposedly with no expiration although I don't see that plan on their site anymore One thing to keep in mind when sharing the connection over wifi without physical access to the modem and router is that sometimes one or both might need to be restarted and that could possibly be inconvenient if it happens when you're working at 2 am and your neighbor's asleep.

matplotlib (python graphing library), gnuplot, dia. Sadly, for presentations I have to usually replot the graphs in Excel or they won't look nice.

If you're still taking classes, you're probably in your 2nd-3rd year at most, right? While 3rd might be a bit late, but on the other hand it could be important to try to meet with other faculty members with whom your interests may align better. Sometimes there is a personality clash and there is not much you can do. Consider also talking to the other students in the group and see if they've had to figure out how to deal with a similar issue. That's a good way to try to deal with it, but if it doesn't work either there is only so much you can do. It may help to try to meet weekly and provide updates on the important decisions you're taking and directions you're considering. The low-level details are not going to be helpful and most likely you won't get help with debugging anything, but having a week to week continuity is important. I would still keep sending weekly emails as they are as much for your benefit as they are for his.

Good point. Also, bear in mind that overall rankings are not as helpful either. What matters is how good of a fit and how good of research is being done in the particular sub-area the applicant is interested in. Personally, I didn't end up paying much attention to rankings, since I've found they were quite deceptive in my case: the two "better" schools had either 1 faculty member interested in my sub-field or none. Going into the "best" school, I believe, would have been a worse choice in my case.

On the topic of batteries, keeping it constantly charged will kill it pretty quickly: http://en.wikipedia.org/wiki/Lithium-ion_battery#Cell_life So, this is somewhat inconvenient, but I've been removing mine when not requiring a battery to work (most of the time) and keeping it charged around 40%. The battery now is 5 years old. After a full charge, it still lasts 2-2.5 hours depending on the brightness of the screen and whether the 802.11 card is on or not and workload. Something to consider if you want to prolong its life.

Also, consider the availability of ATMs. For example, I use KeyBank because it's close to my house, but they only have 3 ATMs in town. One's on Main St, which is nice, but there is nothing close to central campus. Some banks have terms where they waive some number of ATM fees/month and/or reimburse fees imposed by other banks. Finally, I've found that now I charge everything to my credit card, since I get Delta miles for it and then pay it off each month. Just another model to consider. In terms of interest rates, sadly I think interest on checking is virtually non-existent and savings account interest is well below inflation. I don't think they change significantly across the board. I might be wrong though

And the wiki supports all the theories so far, including "pleases and thank-yous". Feel free to actually follow up on the references.

To be fair, specifically for condos, the association does quite a bit of maintenance (depending on the particular association). For example, in my case the hot water, laundry, exterior, parking lot, cutting grass and some other small things are all covered by the association fee. The only big things I need to worry about in terms of maintenance are the appliances, the plumbing inside the condo, and the furnace, since each unit has its own furnace. In my case, the mortgage fee also includes the escrow payment, which covers the property taxes. Since I bought mine before the rates plummeted, right now I'm paying on par with renting a 1 bedroom apartment (including the association fee), but a friend of mine in the same complex who bought a year later pays about half to 2/3 of that. It is definitely a bigger commitment, but I also like the fact that I have definite housing for the foreseeable future and do not have to move apartments or deal with rising prices. Also, the property taxes + interest payments on the mortgage are tax deductible and often exceed the standard deduction. So, in a way they help to offset the fact that you're paying them.

Ok, let me qualify this: since the summer 2008, aside from foreclosures, the condos in my complex have not dropped significantly (about $5,000 at the worst and since then climbed a little bit upward). As far as I could tell, condos in the similar price range (<75k) did not experience a tremendous decline.

Fair enough. I guess I should qualify I was speaking about an MS/PhD program and the first 2 years are coursework. The proposal is due after 3.5 years.

The one thing I've found very true for myself is that no matter how busy you are in the first year, it is by far the year with the most free time -- you're not fully involved in research yet, not thinking of prelims/quals, proposal, making the next paper deadline. Make the first out of the first year in terms of exploring the town/city/surroundings and meeting people! It's a lot harder to make time for that later.

I don't know exactly which subfield of computer science you are in, but I do understand that it's not as easy as taking something out of a conference paper and implementing it. For one, it's not exactly "research". There is the option of going after the "future work" (also known as, "stuff we didn't get to and don't intend to do"), but in certain subfields the amount of effort required to get to the point where you can actually implement/test that future work is significant, especially as most papers don't come with any source code. My suggestion would be to go through the top conferences in your subfield for the past 2-3 years. Read the titles of all papers; next, read the abstracts of ones that seem interesting; finally, read the actual papers that seem promising. Make a list of any ideas you come up with from reading those (preferably, 5-7 ideas or more). Try to talk to someone else in the same subfield (other grad students are great, especially more senior, as well as any faculty) and get a sense for whether the idea is 1. research-y enough, 2. not impossible, 3. has not been beaten to death by others. Ideally, at the end you have some idea of how to proceed with the implementation and where it could be submitted. A single paper should be enough for an MS thesis. Hope that helps in terms of coming up with something.

Well, I did end up buying a condo. It's not a house, but it's been working out pretty well. There is the additional association fee on top of the mortgage payment, but I manage to get it all at around $900 -- with the extra principal payment each month. A one bedroom in this town would run around $800, so, it's not too bad. The property taxes are included in my mortgage payment (the mortgage company pays the tax). The only big maintenance item so far has been a new furnace, as the original one was from 1972. That ended up running about $3,000 -- ouch. Other than that, it's been great. Luckily, University towns also don't experience significant drops in housing prices, so I'm hoping to sell it for around what I paid for it or even a little more. Good luck!

A good place to start would be to look at what are the research groups at each school, what work the professors are doing, and if you're interested in it. Ideally, during your MS you want to be involved in research and either apply for PhD at the same school or apply elsewhere, but have at least one publication to show for the MS. I would urge you not to put too much faith into the rankings.

I'll start by saying all are good schools so you won't make a terrible choice; hope that relieves the tension a bit. As far as U-W TA requirement, personally I would not worry about that too much for a few reasons: 1. Coming in, it is unlikely you're going to be able to do great research; there is a learning curve as with anything and, personally, I've found myself looking for a topic for a while after coming to the university, even though I was working with a professor 2. TA is a good experience, especially if you decide to stay in academia. Further, doing a semester of TA later on in grad school will not be easier, but only harder as you will want to be pursuing research ideas 24/7, probably (I thought it would get easier, but no) 3. It is ok that you are not assigned a professor right away. A lot of students are not sure with whom they would like to work and often in what field. This gives you a chance to try working with a few professors and get a feel for how they interact with students, the problems they're interested in, etc. As long as you figure it out by the end of the first year, you're doing fine. As general advice, make sure you always go to a school that has options. You don't know the professors and their research/advising style at this point; it may turn out that you work better with someone else. It could also turn out you want to work in a slightly different area. If you don't have an opportunity to switch you're kinda stuck. Anyway, I won't tell you where to go, but hope this advice is helpful. All things being equal, you can also consider location, setting, proximity to places of interest or whatever other metric you can come up with Good luck with the decision! I know it's tough.

That's even more draconian than what's going on here! We have group packages (so you can sit with your friends) and most students who care enough buy the season packages. The package is then sent out over the summer and you get all the tickets for the season in one go, without having to go to the ticket office. So at least that part is pretty good. Mind you, this is all football because that's the only experience I've had so far. Also, craigslist becomes littered with people selling (and sometimes scalping) their tickets the week before the game, so it's definitely possibly to get tickets even without the season package. The other thing that sucks is that the group seating assignment is done by the lowest credit hours in the group, so we can't get better seats by including an undergrad in the group

Depends on the school. My particular institution treats grad students on par with freshmen, i.e. 0 credit hours logged at the school. As such, the seating is pretty bad. The football and basketball games cost money (football around $45/game or $250/season) and basketball is $25/game. Other games, due to low attendance, are frequently free, like women's basketball, volleyball, etc. It'll depend on the school and the particular sport. I'm sure you'll find out more from the school once you're there.