I can never claim I'm a master of sarcasm. I tend to accept things at face value unless someone says something that's completely out of character or nonsense. I can't help it that's just the way I am. Over time as I learn a person's personality I tend to anticipate sarcastic remarks. It takes me a while to pick up on tone of voice, and body language associated with sarcasm because each person's presentation is unique. Frankly I don't know how most people read sarcasm. I'm not demented and I tend to think of myself as being reasonably smart, so I tend to believe mastering sarcasm is not universal human trait so measuring dementia based on lack of perception of sarcasm cannot be that accurate.
I will say that my inability to read sarcasm, among other things, causes many of my friends and colleagues to claim that I seem to have Asperger's syndrome but I haven't been diagnosed with it by a physician.
Actually I am a math major (near top of my class if that even matters), and no my math is not wrong...I pay tuition per semester, not per year, and each semester is four months long.
but I'd rather have universities host a music service for their students. I live in Canada so I'm immune to the RIAA (for now, the equivalent Canadian entity make come to bite us soon enough), but I wouldn't mind an extra $100 fee on top of tuition for unlimited access to legal 256kbps+ DRM free music. That comes out to be $25/month, which is less than what I pay for music per month anyway.
This is a general reply to this thread rather than to the specific parent. I used look at things the same way, "When am I ever going to use this?" How can I apply this?
By 2nd year in computer science, I realized, there's a lot of cool stuff in computer science that I had no clue how to do, and none of these CS courses were going to teach me. Frankly, it's easy to be ninja programmer by age 20, at least 10% of my class started coding at age 7 or 8 (a couple started programming at age 6). What is hard is developing new and innovative solutions. By end of second year, I was convinced I don't want to be a code monkey but instead I wanted to innovate, I want to apply Computer Science in ways that haven't been done before. I decided CS courses weren't going to cut it for that. At my school, CS courses tend to be more hands on, and coding oriented whereas math courses tend to be more theory oriented. Thus, I decided to take the majority of my courses in math: combinatorics, optimization, group & ring theory (abstract algebra), computational linear algebra, calculus even.
Now, two years later, I find I've developed a sizable toolkit of techniques to solve problems that are not well defined. I can look at a problem (constructing directions on a map based on traffic conditions say), and realize, ohh, that's a network flow problem, I can model it as such, and the solution is easy. Or more subtle, how to find the optimal seating arrangement of guests at a wedding given a set of constraints (family members can't sit together say).
Basically, rather than thinking, ohh, I want to be a web developer or I want to be a game developer, I thought, I want to solve problems with computers. And in that scenario, math is vital and you can never learn enough math.
I'm pretty sure one of my profs got his PhD in CS at age 20 so those ninjas do exist. I've been developing as a hobby since I was 7 years old myself as were at least 20% of my class. Never got into extremely theoretical stuff but yes, a guy who just has a couple of years experience and hadn't touched a computer until college is at a significant disadvantage because companies would prefer to hire those "whiz kids" for the same salary but get 10x the work done.
I'm 22 years old now, entering grad school in computer science and I knew I was in love with computers since I was 8 years old. I started programming at age 7 or 8, in BASIC. My mom picked up a BASIC manual, learned the language herself and taught me what she learned (out of family rivalry really because at the time I was the only kid in the family who wasn't a self-proclaimed computer expert). Since then I've been enthralled with computers and algorithms and I'm doing very well in school as a result. About a dozen people in my class also started programming well before the age 10 as well, so it's a relatively common thing. If you can do basic arithmetic and reason about complex or abstract things at that age, you can program to some degree.
Waterloo rocks for undergrad no doubt, especially computer science. Bill Gates explicitly said Microsoft hires more new grads from the University of Waterloo than any other university in the world. It's sure going in that direction for other major corporations as well such as Google, and NVIDIA which are beginning to hire more people from here than other schools including top U.S. schools.
As a Computer Science major at the University of Waterloo, I'm proud to hear that the Waterloo region and it's research and higher education institutes are attracting top minds to the area. Arguably, the university is already the top undergrad tech school in Canada and competitive with the top in the U.S. as well, but currently only on the undergraduate level. However, huge amounts of funding have been recently pouring into UW's graduate programs, and associated research institutes. Some Waterloo folks like to label the university "MIT of the north" but unfortunately, I'd argue that's currently only true on the undergraduate level. Its co-op program makes Waterloo grads among the most employable new grads in the world. With this kind of exposure I'm sure Waterloo will reach its goal of being competitive with the top U.S. tech schools in all areas within the next 5 to 10 years. Maybe they'll start making movies with references to Waterloo as well.:P Then again, most movies are made/produced in the U.S...
No one seems to be commenting about the new features of this release but simply on TrueCrypt in general. Am I the only one excited about the multi-core/processor support? Finally a piece of systems level software that scales with the number of cores! Makes getting a multi-core processor all the more worthwhile.
In Canada, Rogers has set the cap at 60GB per month for their standard "Express" service and 95GB per month for their "Extreme" service. 250GB per month is very generous.
People don't seem to understand why parallelism is the next big thing right now. It's simple, processors are NOT going to increase significantly in frequency and in instruction-level parallelism in the upcoming years. Why? Well, we've hit the "power wall". These CPUs are producing too much heat at such high frequencies. Thus, until a breakthrough in manufacturing technologies comes along, we need to keep clock speeds low. So how do we improve performance then? Multiple cores! Every PC in the world will now have multiple cores.
People argue that servers have been doing this for ages but they neglect the granularity of parallelism. It's usually at the per-user level. As web technologies advance, servers will be getting extremely complex queries from users, and we need to find ways to split the processing of that one query onto multiple machines.
So, the truth is, not enough people are learning about parallelism, and many of those who claim to understand parallelism by saying they're familiar with concepts such as semaphores and mutexes aren't necessarily up to the task of this next generation of processing. We need to redevelop our most basic algorithms to run on an arbitrary number of processors. We need to develop programming languages that let us easily express parallelism. We need to understand how to build scalable systems. These are very, very difficult problems.
Parallelism isn't new, yes, but now we urgently need it or else we won't see any progression in computer performance for many years to come.
I went into "software engineering" at a top Canadian university. Two years later, I realized I hated it...the math courses all sucked, because they weren't math. Profs de-emphasized proofs, and emphasized application...cool, that's nice, except I found myself memorizing patterns in problem solving rather than understanding the underlying theory. This hurt my performance, as memorization was a weakness of mine, my strength is REAL math, logical reasoning. I was getting grades in the A- to A range in my first two years of undergrad and working quite hard. I decided to switch to mathematics (combinatorics in particular) and computer science and now I'm getting A+s because we discuss proofs and derivations in math courses rather than jumping straight to application without a strong foundation. That makes things easier to remember and it makes it easier to solve problems. I don't know if this is an issue in the U.S., but I can with some degree of confidence say, if you aren't doing that great, but you think you're good at math, maybe you should've been a math major rather than an engineering major. Engineering does NOT test whether you're good at math or science, it tests how good you are at blindly following patterns.
Itanium? The processor? A processor can only do so much on its own. Ultimately I think we need to reinvent sorting, searching, encryption, compression, FFT, etc and teach undergrads how to design these algorithms to run on n processors where n is some arbitrary number. We need libraries that do this stuff for us as well. We also need programming languages and platforms like RapidMind to keep us in the "data-parallel" mindset. I've attended two talks by Professor David Patterson from UC Berkeley and in each case this roughly the argument he's been giving major corporations and universities in those research areas.
I'm about to start a graduate degree in this area so I'm a little biased. However, I think a lot of problems can be solved in parallel. For example, maybe, LZW compression as it's implemented in the "zip" format might not be easily parallelizable but that doesn't prevent us from developing a compression algorithm with parallelism in mind. I did some undergraduate research in parallel search algorithms and I know for a fact that there are many, many ways you can parallelize search. Frankly, saying that you can't parallelize algorithms is a bit closed minded. Many problems don't inherently require serial solutions, it's just current algorithms handle them that way. Rather than trying to implement existing algorithms on a massively parallel processor, you want to re-tackle the problem under a new model, a model of an arbitrary number of processors. You build around the idea of data-parallelism rather than task-parallelism. Many, many things are possible under this model and I think it's naive to think otherwise. You don't need to think, how do I juggle 1000 threads around, you think, how do I take a problem, break it up into arbitrarily many chunks and distribute those chunks to an arbitrary number of processors and how do I do all that scheduling efficiently? This model doesn't work for interactive tasks mind you (where you're waiting for user input), but I'm very confident a model can be developed that can.
I have plenty of friends who've interned at Google (though I haven't myself yet), and so I have a good idea of what working at Google is like. One of my friends is a very academically passionate guy. He's at work on average 12-15hrs per day (He's at work until 5am on many occasions). He doesn't always "work" per se of course. He goes to the gym or plays one of the other recreational activities there. He goes to research seminars and does independent (but related) research to his job. I guess it's a part of his 20% independent time. I've visited the Google campus (Googleplex I guess they call it), and it's fabulous. It has a unique university feel to it. Plenty of lounges and open seating areas. People just sit around doing work wherever with their feet up and some Naked juice (ohh how I love the Naked juice). Yes, Google cares about where your degree is from, but Stanford is not the only school they hire from by any means. The University of Waterloo (where I'm from) in Canada has the #1 co-op program in the world and so Google hires plenty of interns from there (many of whom become full time employees presumably). I can't make any claims about the Microsoft life style though because I only have one close friend working there, and I haven't been to the Redmond campus yet (though I may be going up for a visit in a few weeks). So yes, the young eager ones do not have a social life, but you're free to be as academic and research oriented as you want to be. If you think you have a brilliant idea, the company gives you the flexibility to explore the idea. There are plenty of young inventors out there with ideas that never make it commercially because they weren't willing to take the chance to start up a company and their employer didn't give them the time to explore it. At Google, you have that time, and I've heard if you come up with an idea that ultimately is released, they'll reward you handsomely making starting up a company less appealing.
Slashdot Mirror
I can never claim I'm a master of sarcasm. I tend to accept things at face value unless someone says something that's completely out of character or nonsense. I can't help it that's just the way I am. Over time as I learn a person's personality I tend to anticipate sarcastic remarks. It takes me a while to pick up on tone of voice, and body language associated with sarcasm because each person's presentation is unique. Frankly I don't know how most people read sarcasm. I'm not demented and I tend to think of myself as being reasonably smart, so I tend to believe mastering sarcasm is not universal human trait so measuring dementia based on lack of perception of sarcasm cannot be that accurate. I will say that my inability to read sarcasm, among other things, causes many of my friends and colleagues to claim that I seem to have Asperger's syndrome but I haven't been diagnosed with it by a physician.
Actually I am a math major (near top of my class if that even matters), and no my math is not wrong...I pay tuition per semester, not per year, and each semester is four months long.
but I'd rather have universities host a music service for their students. I live in Canada so I'm immune to the RIAA (for now, the equivalent Canadian entity make come to bite us soon enough), but I wouldn't mind an extra $100 fee on top of tuition for unlimited access to legal 256kbps+ DRM free music. That comes out to be $25/month, which is less than what I pay for music per month anyway.
This is a general reply to this thread rather than to the specific parent. I used look at things the same way, "When am I ever going to use this?" How can I apply this? By 2nd year in computer science, I realized, there's a lot of cool stuff in computer science that I had no clue how to do, and none of these CS courses were going to teach me. Frankly, it's easy to be ninja programmer by age 20, at least 10% of my class started coding at age 7 or 8 (a couple started programming at age 6). What is hard is developing new and innovative solutions. By end of second year, I was convinced I don't want to be a code monkey but instead I wanted to innovate, I want to apply Computer Science in ways that haven't been done before. I decided CS courses weren't going to cut it for that. At my school, CS courses tend to be more hands on, and coding oriented whereas math courses tend to be more theory oriented. Thus, I decided to take the majority of my courses in math: combinatorics, optimization, group & ring theory (abstract algebra), computational linear algebra, calculus even. Now, two years later, I find I've developed a sizable toolkit of techniques to solve problems that are not well defined. I can look at a problem (constructing directions on a map based on traffic conditions say), and realize, ohh, that's a network flow problem, I can model it as such, and the solution is easy. Or more subtle, how to find the optimal seating arrangement of guests at a wedding given a set of constraints (family members can't sit together say). Basically, rather than thinking, ohh, I want to be a web developer or I want to be a game developer, I thought, I want to solve problems with computers. And in that scenario, math is vital and you can never learn enough math.
I'm pretty sure one of my profs got his PhD in CS at age 20 so those ninjas do exist. I've been developing as a hobby since I was 7 years old myself as were at least 20% of my class. Never got into extremely theoretical stuff but yes, a guy who just has a couple of years experience and hadn't touched a computer until college is at a significant disadvantage because companies would prefer to hire those "whiz kids" for the same salary but get 10x the work done.
I'm 22 years old now, entering grad school in computer science and I knew I was in love with computers since I was 8 years old. I started programming at age 7 or 8, in BASIC. My mom picked up a BASIC manual, learned the language herself and taught me what she learned (out of family rivalry really because at the time I was the only kid in the family who wasn't a self-proclaimed computer expert). Since then I've been enthralled with computers and algorithms and I'm doing very well in school as a result. About a dozen people in my class also started programming well before the age 10 as well, so it's a relatively common thing. If you can do basic arithmetic and reason about complex or abstract things at that age, you can program to some degree.
Waterloo rocks for undergrad no doubt, especially computer science. Bill Gates explicitly said Microsoft hires more new grads from the University of Waterloo than any other university in the world. It's sure going in that direction for other major corporations as well such as Google, and NVIDIA which are beginning to hire more people from here than other schools including top U.S. schools.
As a Computer Science major at the University of Waterloo, I'm proud to hear that the Waterloo region and it's research and higher education institutes are attracting top minds to the area. Arguably, the university is already the top undergrad tech school in Canada and competitive with the top in the U.S. as well, but currently only on the undergraduate level. However, huge amounts of funding have been recently pouring into UW's graduate programs, and associated research institutes. Some Waterloo folks like to label the university "MIT of the north" but unfortunately, I'd argue that's currently only true on the undergraduate level. Its co-op program makes Waterloo grads among the most employable new grads in the world. With this kind of exposure I'm sure Waterloo will reach its goal of being competitive with the top U.S. tech schools in all areas within the next 5 to 10 years. Maybe they'll start making movies with references to Waterloo as well.:P Then again, most movies are made/produced in the U.S...
No one seems to be commenting about the new features of this release but simply on TrueCrypt in general. Am I the only one excited about the multi-core/processor support? Finally a piece of systems level software that scales with the number of cores! Makes getting a multi-core processor all the more worthwhile.
In Canada, Rogers has set the cap at 60GB per month for their standard "Express" service and 95GB per month for their "Extreme" service. 250GB per month is very generous.
People don't seem to understand why parallelism is the next big thing right now. It's simple, processors are NOT going to increase significantly in frequency and in instruction-level parallelism in the upcoming years. Why? Well, we've hit the "power wall". These CPUs are producing too much heat at such high frequencies. Thus, until a breakthrough in manufacturing technologies comes along, we need to keep clock speeds low. So how do we improve performance then? Multiple cores! Every PC in the world will now have multiple cores. People argue that servers have been doing this for ages but they neglect the granularity of parallelism. It's usually at the per-user level. As web technologies advance, servers will be getting extremely complex queries from users, and we need to find ways to split the processing of that one query onto multiple machines. So, the truth is, not enough people are learning about parallelism, and many of those who claim to understand parallelism by saying they're familiar with concepts such as semaphores and mutexes aren't necessarily up to the task of this next generation of processing. We need to redevelop our most basic algorithms to run on an arbitrary number of processors. We need to develop programming languages that let us easily express parallelism. We need to understand how to build scalable systems. These are very, very difficult problems. Parallelism isn't new, yes, but now we urgently need it or else we won't see any progression in computer performance for many years to come.
I went into "software engineering" at a top Canadian university. Two years later, I realized I hated it...the math courses all sucked, because they weren't math. Profs de-emphasized proofs, and emphasized application...cool, that's nice, except I found myself memorizing patterns in problem solving rather than understanding the underlying theory. This hurt my performance, as memorization was a weakness of mine, my strength is REAL math, logical reasoning. I was getting grades in the A- to A range in my first two years of undergrad and working quite hard. I decided to switch to mathematics (combinatorics in particular) and computer science and now I'm getting A+s because we discuss proofs and derivations in math courses rather than jumping straight to application without a strong foundation. That makes things easier to remember and it makes it easier to solve problems. I don't know if this is an issue in the U.S., but I can with some degree of confidence say, if you aren't doing that great, but you think you're good at math, maybe you should've been a math major rather than an engineering major. Engineering does NOT test whether you're good at math or science, it tests how good you are at blindly following patterns.
Itanium? The processor? A processor can only do so much on its own. Ultimately I think we need to reinvent sorting, searching, encryption, compression, FFT, etc and teach undergrads how to design these algorithms to run on n processors where n is some arbitrary number. We need libraries that do this stuff for us as well. We also need programming languages and platforms like RapidMind to keep us in the "data-parallel" mindset. I've attended two talks by Professor David Patterson from UC Berkeley and in each case this roughly the argument he's been giving major corporations and universities in those research areas.
I'm about to start a graduate degree in this area so I'm a little biased. However, I think a lot of problems can be solved in parallel. For example, maybe, LZW compression as it's implemented in the "zip" format might not be easily parallelizable but that doesn't prevent us from developing a compression algorithm with parallelism in mind. I did some undergraduate research in parallel search algorithms and I know for a fact that there are many, many ways you can parallelize search. Frankly, saying that you can't parallelize algorithms is a bit closed minded. Many problems don't inherently require serial solutions, it's just current algorithms handle them that way. Rather than trying to implement existing algorithms on a massively parallel processor, you want to re-tackle the problem under a new model, a model of an arbitrary number of processors. You build around the idea of data-parallelism rather than task-parallelism. Many, many things are possible under this model and I think it's naive to think otherwise. You don't need to think, how do I juggle 1000 threads around, you think, how do I take a problem, break it up into arbitrarily many chunks and distribute those chunks to an arbitrary number of processors and how do I do all that scheduling efficiently? This model doesn't work for interactive tasks mind you (where you're waiting for user input), but I'm very confident a model can be developed that can.
I have plenty of friends who've interned at Google (though I haven't myself yet), and so I have a good idea of what working at Google is like. One of my friends is a very academically passionate guy. He's at work on average 12-15hrs per day (He's at work until 5am on many occasions). He doesn't always "work" per se of course. He goes to the gym or plays one of the other recreational activities there. He goes to research seminars and does independent (but related) research to his job. I guess it's a part of his 20% independent time. I've visited the Google campus (Googleplex I guess they call it), and it's fabulous. It has a unique university feel to it. Plenty of lounges and open seating areas. People just sit around doing work wherever with their feet up and some Naked juice (ohh how I love the Naked juice). Yes, Google cares about where your degree is from, but Stanford is not the only school they hire from by any means. The University of Waterloo (where I'm from) in Canada has the #1 co-op program in the world and so Google hires plenty of interns from there (many of whom become full time employees presumably). I can't make any claims about the Microsoft life style though because I only have one close friend working there, and I haven't been to the Redmond campus yet (though I may be going up for a visit in a few weeks). So yes, the young eager ones do not have a social life, but you're free to be as academic and research oriented as you want to be. If you think you have a brilliant idea, the company gives you the flexibility to explore the idea. There are plenty of young inventors out there with ideas that never make it commercially because they weren't willing to take the chance to start up a company and their employer didn't give them the time to explore it. At Google, you have that time, and I've heard if you come up with an idea that ultimately is released, they'll reward you handsomely making starting up a company less appealing.