Funny, is it just me, or are all these inventions that you mention ones that came about long before Europe become the socialist paradise it is today. Innovations that came about 60+ years ago really doesn't seem to suggest that Europe is thriving in innovation. Maybe I'm wrong, it's known to happen, but your post did nothing to back up your assertions.
ARM has a couple processors already that are pretty high on the performance measurement. For instance the Arm Cortex A9 has a dual issue pipeline, and limited support for out of order processing (similar to the original Pentium processor in that regard). This chip also can contain up to 4 cores, and have up to a 2MB L2 cache. I think they can run up to about 1GHz. They also have full support for floating point and all that good stuff. I'm pretty sure ARM is also working on developing an true OoO processor that will likely be running in the GHz range which would likely be ideal for a netbook.
Remember, with a netbook, you don't gain much by lowering the CPUs power consumption to less than 5 watts or so. The reason for this is simple, the display, ram, hard drives and everything else consume enough power that it won't really help battery life very much. I can imagine though that a quad core ARM A9 at 1GHz would make for a really nice netbook. Having multiple cores is nice on those for web browsing (playing flash in the background of your tabs, etc), and also for many media tasks. It would also be great if they included a graphics chip (or gpu as part of a SoC system) that could handle h.264 decoding for the netbook.
Phil
Re:Used car salesmen use the same thing
on
Cellular Repo Man
·
· Score: 1
I'm pretty sure most car loans don't let you put up the car for sale without first contacting the bank. Generally they're going to have to require payment in full, or some other form of security before they'd okay the sale of the car. Basically you don't have the right to sell the car. By the same rules, you have to have certain levels of insurance on a car you buy with a loan. The bank won't allow you to secure your loan with an asset that could become worthless overnight (for instance if you total the car, or get it stolen).
Just because the ISA supports reading and writing memory in one byte increments doesn't mean the chip physically reads or writes a single byte to the DRAMs. Data is generally read and written to the caches, which will later get written back to memory. Most caches don't bother to contain the state needed to know which bytes in a cache line were modified. Additionally, most memories themselves aren't really good at single bye reads/writes. They operate much faster when reading and writing larger data chunks at a time.
Wait.... I'm not really sure if you are agreeing with me or disagreeing with me here. You first state that you have an interest in research, and have done some. You then state that you don't think the major research universities are that research specific etc. Believe me, you are getting special attention because of your work, and what you have done. Professors appreciate the bright students, and will go out of their way to help them. The fact that you have lunch with them is wonderful, and happens often enough, but you don't get a normal coursework student doing that often. Most wouldn't want to anyhow. In this regard, you've basically proven my point.
I wasn't trying to say in my post that professors don't care at all about teaching, and that undergrads aren't their concern, but in general that is not their primary concern. There time first and foremost tends to go toward research.
As far as other comments saying that many top schools have coursework only options. This is also true. However, at most major research-one schools, even the coursework option for masters will likely have a strong research angle. You will be expected to read many many research papers in graduate courses, and can expect to do research projects re-implementing other research papers, verifying their results, or sometimes extending upon it and adding to it. If you have no interest in research, and only care about a degree to get better jobs, this path is far from the way of least resistance, and will likely not work for you. You have to be willing to put in a huge sacrifice for it, and it will not be like an undergrad degree but with harder classes. I've seen that at places, but it's much harder to manage at top tier research universities.
Also, while the big name universities degree means more, what really means a lot is the recommendations and contacts you'll meet there. If you don't have the skills to hack it in one of these grad programs, you likely won't survive. Maybe I'm too hasty to judge, but most students who are going to graduate school solely because they want to earn more money aren't the best and the brightest students, and are going to be looked down upon by others. It's not so much that the academics are stuck in their ivory tower, so much as they tend to value knowledge for knowledge's sake. If your sole reason for learning is money, you have the wrong attitude. Obviously, making more money is a factor involved in most people's decision to go to graduate school. But if they are going to a major university for graduate school, chances are good they are interested in learning just because they want to learn. They may have decided it wouldn't be economically feasible to go back for more school if it didn't pay off in the end, but they truly wanted to go to school, and don't just want the money. Anyhow, if you fall into this category and don't want to learn for learnings sake, you will not, and should not be able to get through graduate school at a top research school. The degree is not merely about reciting facts, and memorizing ideas, but rather being able to analyze and come to new conclusions. It should show that you're capable of doing research, even if you haven't already done any real research. These are not skills that all top students have.
Part of it comes down to the fact that how much you get out of the degree is based on what you put into it. Many times there are easy and hard projects that you could do for a class. If you always go the easy route, and do the bare minimum work, you will be able to pass through the degree. However, when someone later interviews you and asks about your experience, you now lack the experience that you're expected to have coming out of one of these institutions (they can find this out pretty easily based on what you told them you have done), and you will likely not have great recommendations. Mostly because unless you really push yourself, you won't be able to stand out and get the attention of your professors. This attention is far mo
Seriously, if your concern for going to grad school is solely to have something on your resume that looks better and gets you paid more, don't go. As a grad student in computer engineering, I can't stand the people who want to get a masters just because it makes them look better. And, if you do get a masters, don't bother getting it at a big name university, because that likely won't mean anything once you get it. The big name universities have the name because of the research they do. The research determines the ranking of their graduate program. If you plan on going just to get a degree, and not do any research, you'll end up shorting yourself of a better education elsewhere, and you'll waste the time of professors and other students who are actually interested in doing research. After graduating from one of these schools it won't really make you look much better either. You'll talk to companies and get in the door for having a big research school's name on your degree, and they'll ask you about what research you did, or ask for recommendations from faculty etc. You likely either won't know any faculty very well (as they're concerned with doing research, and not some masters student who only cares about making more money), or they'll have a low opinion of you for wasting space in their program (that space could have instead been used by someone interested in pursuing research).
Sorry if I sound really negative about this, but this is the truth of academia. The big name schools are concerned with research. That is why they have a big name, and that is what they will focus on to maintain their reputation. They often do not offer a better education, and in fact they are often less concerned with teaching than smaller lesser known schools. The professors just can't afford spending too much time teaching, because in the end (for getting tenure at least), research is what matters. In fact, at many of these schools, it is looked down upon if a junior faculty members wins a teaching award. The rest of the university assumes they're spending too much time on their teaching, and not enough on their research.
My recommendation is to talk to the faculty at your current university. See what they recommend, and be truthful about why you want to go to grad school. Slashdot is not the place to find out about this stuff, most people here have no clue. Also remember that as far as graduate programs at top schools go, it's not really that one school is better than another. In reality its that one school is better in one particular specialty area. The choice of which school is best for you depends much more heavily on what you plan on specializing in rather than the US News ranking. Employers know what schools specialize in, and base decisions on that. If you don't plan on specializing (as you don't seem to be concerned with research), the rankings immediately become relatively worthless. Talk to faculty that you know and trust. They can help you, but you have to show that you're worth spending time on. They likely have more important things to do, and don't want someone wasting their time.
I meant that a coal plant can't change its load in the timescale of hours. They will produce the amount of power they are producing for a while, and they will not be producing more during the day than during the night. They just don't operate that way.
Now, the power plants do establish trends, and forecast power demands so they know how to run their plants. This is a different problem though. The demand trends are to make sure they can meet peak demand, and are less concerned with offpeak hours. There is little that can be done to a coal power plant to allow it to run at say 80% load during the afternoon, and 50% load at night.
I'm not saying conserving power is bad, and that it's a waste to do, just that arguing that you are generating excess CO2 by using this power at night isn't necessarily true.
Selling power to other areas is fine and good, but requires a few things. First, you need a way to transfer power to the other areas. High voltage power lines however can manage this... the current infrastructure doesn't wasted too much energy to send power a couple hundred miles. However, at least where I live, when it's nighttime, it is also nighttime a couple 100 miles away. When there is peak demand, there is also peak demand a few hundred miles away, or close enough to it. Unless you have installed extremely high voltage DC lines to transfer power 1000s of miles (even then it's not that large of a distance when thinking on a "global" scale, which cost a TON of money to setup, you really aren't going to get far with selling your excess power. Really to make it economical, you'd need a way to transfer your excess capacity 10s of thousands of miles with little loss. The infrastructure to do this effectively would be both enormous, and wasteful, as much of the power generated would be lost in transit. Not to mention that doing this would be dangerous, and likely hurt many people. Networks that hold this much power are dangerous, also they can create a single fail point that becomes a prime target.
Actually, even the ones powered by coal are likely not wasting much CO2. Considering a machine is most likely to be sitting idle at night, and that the coal plants have to operate 24/7 (they can't dynamically lower their power output, that's provided by secondary sources during the afternoon). Power usage generally peaks in the afternoon, and so other power generation stations (those like natural gas that can be brought online quickly) handle the peak load, but, as coal power is cheaper, they try to get as much as possible from the coal. If the base load provided by the coal is greater than the power being consumed, than any additional power demanded isn't really "wasting" electricity. It's just using electricity that has already been generated. Of course, if this amount is great enough to change the power plants operating conditions, it does matter, and as far as the businesses are concerned, this power does cost money, and quite a bit of it.
However, saying the plant is releasing more CO2 for these computers is generally not true.
there might be some truth to what this guy is saying in that link, but i find parts of it to be wrong. For instance, he's talking about heap fragmentation, and not actual memory fragmentation. This should be an issue handled on the application level rather than the OS level. The article also mentioned sending a request for 256 bytes to the OS and failing..... Unless he's working on some machine I don't know of, you can't really just request 256 bytes of memory from the OS. The OS allocates pages of memory (4K is the standard size on the x86, although larger sizes exist), and can give applications pages of memory, but can't allocate space in smaller increments. The reason for this is memory protection. If byte 0 of a page is readable by an application, so is byte 4095. Same with write and execute permissions. Generally, an application should handle the heap in the user space. A request for 256 bytes will check available memory to the application and see if it can find it in a free list. If it can't the heap manager should request another page (or more depending on allocation policies) of memory, and then service the heap request. The link seems to confuse some of these points.
* Tag editing or even reading failures when approaching the 256MB limit because software will try to put each ID3v2 frame in a single memory block and allocating a single block of such size is likely to fail in 32-bit address space because of fragmentation issues.
What the hell are you talking about here? It might fail to allocate a 256 MB block if the machine doesn't have enough memory, or if the program decoding the module is running in the kernel and using kmalloc, but for the most part, applications do not have to worry about memory fragmentation. Virtual memory takes care of fragmentation for you, as only 4KB pages need to be contiguous.
The only time this wouldn't work is if the application that you're running doesn't have 256MB of its address space free. Unless the application is using close to the 2GB or 4GB address space the application is given this shouldn't be an issue.
unfortunately, the overhead of pipe operations render many chains of operations slow enough that the parallelism gained by using multiple cpu's can be eaten up. Especially when some of the pipe operations perform extremely simple operations on data. These don't tend to scale too well, and the overhead will often eat up all the benefit of multithreading. Additionally pipes have their place, but can fail when sub-channel data is needed, or when multiple pipes are needed (multiple input or output streams, pipes work well for problems with a single input source and a single output destination). Also there are some operations in a long pipe chain that are likely to be the most expensive by far of any of the operations. Often these operations could be parallelized, but the program performing the operations doesn't realize it. This leads to inefficiencies, because if 95% of your time is spent in one operation, Amdahl's law says that you could throw an infinite number of processors at the problem, and only speed up your application by 5%.
What you should look into if the ideas behind pipes seem logical for multiprogramming is the use of stream-based languages. I know StreamIt was one of the early ones developed. These languages tend to take the pipe paradigm to a whole new level, allowing splits in a pipe, joins, etc. Also, the idea of futures (I think they were introduced in Scheme, but I could be wrong) is something that really needs to be enabled in more programming languages. A future is a subroutine or function that is "called" at one point, but the system knows it doesn't have to be evaluated until any of its outputs are requested by the application. With futures you could trivially launch 10 "parallel" jobs at once, and the system decides how to schedule them. Of course, a big problem with these is that they tend to fail pretty bad in a language like C where a compiler doesn't really know where outputs of functions can be later used.
yeah... because all code from '01 will compile with gcc 4.0... hell, I bet a lot would have issues with gcc 3.3. Not to mention include files have changed, installed libs on the system are different etc. I'm still working on (actively developed) code now that doesn't compile with g++ 4.0 (granted the codebase is part of a research project so their priorities are different). I've also used multiple other research programs developed in the late 90s or early '00s, and many of them require quite a bit of tweaking to get working. Nothing like fixing 100 errors brought about by changes to how gcc implements the standards, and where headers are included, what is included in what headers etc. But if you consider that stuff straightforward, be my guest.
Hell, both are pretty nasty at times. Have you ever tried installing software that's 5-10 years old from source on a linux machine. Oh wait, the source relies on ancient versions of gcc, and doesn't work with libraries you currently have installed, etc. It's definitely not a trivial problem. If you download all the old compilers and libs first, you might manage to get it working, but this is far from a simple solution.
Isn't one of the points of using DLLs to reduce the size of programs on the machine. If an application is using 3rd party libraries, it often just makes sense to install those libraries into the windows directory because.... THATS WHERE THEY BELONG! If you were installing a linux application that relied on libraries, you would not want to install the libraries into a folder where only that single application could use it. You'd install it somewhere global so all applications can use them, and you don't need multiple copies of them.
Of course, one could say it's not an application's job to install its support libraries, but that's not really an option, as users want to be able to install an application and have things work without first having to install 10 other libraries (the same reason on Debian systems people use apt). Unfortunately as there's no standardized windows install setup, things are a bit messier.
Just because wifi works fine does not mean the CPU is not the bottleneck. I do not own an iphone, and don't know how the implementation of the radio stack on it, but it is entirely possible that stages of the data reception/transmission are handled by the CPU (or possibly a separate DSP) on the iPhone. In this case, if this CPU is not powerful enough, the iPhone can't max out the 3G bandwidth. Just because a different protocol that uses an entirely different radio stack that requires different processing power works fine does not mean all radio stacks do.
Am I the only one who found this quote from the article ironic:
"Although it now stretches to fewer than 100 sites there is still fun to be had in the text-only web, providing your web browser still supports Gopher."
I think the author of this article either doesn't know what he's talking about or got confused. Gopher sites are not part of the web at all, and definitely aren't part of some "text-only" web. Maybe text-only gopher, but definitely not the web.
Speeding is not a victimless crime, at least not if you're going fast enough to warrant a ticket. 5-10 over isn't bad, but I've not seen anyone ticketed for 5-10 unless it was a super busy day.
I wish I could say that I haven't seen cops pull you over for 5-10 over the limit, but I have seen it first hand. I was once driving across North Dakota, and going 80 in a 75 mph zone. When the cop pulled me over my first thought was "huh, what did I do?" An hour later and a search of my car later the cops let me go with a $25 ticket for speeding, and a $45 ticket for having a box in the back seat of my car (i was obviously moving and my car was loaded up with stuff) that had an opened bottle of booze in it (unless I spent 10 minutes pulled off to the side of the road trying to get to it there was no way I could without first running my car off the road). What I learned that day was an unshaven, wired on coffee 20-something person of hispanic descent driving with out of state plates is going to get pulled over for whatever reason a cop likes. They will then justify searching your car with stuff they found while searching your car (in my case this was the truth, the report cited that the cop had seen an open container and wanted to search the car when in fact I told the officer that yes there was a box with booze in it in my back seat when he asked if he could search my car and asked if I had anything to tell him. Of course at the end of the day the fines were relatively light, and it really made an amusing story. Especially when they put me in the back seat of the cop car and one cop started demanding I tell him where the drugs are. I'm not sure if I could keep a straight face at that point, it was just so surreal.
Of course, I don't see speeding as a victimless crime. Sometimes it can be (like the few times I've been on a completely deserted highway and just wanted to see how fast I could push it), but often it isn't. It's the asshats who fly down neighborhoods and through town that tend to endanger people. I just wish I could say that they're the ones who get punished the most for it.
The president of the united states also said he wasn't going to appoint any lobbyists on his cabinet. We saw how long that lasted (until he started announcing who was on his cabinet). People have a false sense of belief that just because a politician has been spouting on about hope and change for years that he will actually do it. Believing that ignores years of political precedents. Obama will likely change things, but for the most part, his administration has largely been a continuation of Bush's policies. The only major exceptions I can think of are funding to abortion groups, and funding for stem cells. People will point to Guantanamo, but it was likely going to get closed down within a year or two as people came up with a good solution for how to deal with terrorists (in reality we'll likely return to pre-9/11 conditions where we simply hired other countries to tor^k^k^Kinterrogate them for us).
I see no reason to suspect that the Obama administration will act any differently than the Bush administration concerning net neutrality.
Mod this poster up. While companies pay credit card handling fees, handling cash is often more expensive. While banks generally don't charge them for depositing, etc, you have to have places to put the cash, deliver it to a bank, etc. All this costs money. More importantly is the concern over employee theft. If most everyone pays in credit there isn't much cash in the drawer to hide the fact that someone stole a $20. Most stores do checks at the end of the night, but don't care too much if its off by a few dollars as employees make mistakes. The more money in cash they collect the more mistakes are allowed. Credit means that the exact amount is charged, and the employee has no easy way to take the money.
Plus of course there's the fact that credit cards tend to move lines faster. If this means a store can have only 4 lines open instead of 5, they're saving money right there.
Actually, in this instance it can be considered theft. Copying a cd for personal use isn't necessarily theft, but copying a cd, and then selling burned copies of it for profit is definitely very close, if not indistinguishable from theft. It's one thing to illegally use someone else's content. It's another altogether to use someone's content as your own while they pay the cost to distribute it. Companies do have legal rights to prevent others from doing this, and they can and should use them. If there was no such thing as copyright, information might exist, but it would be in a wide range of disarray and it would be hard, if not impossible to determine which sources you can trust. If some other website could republish all the content on the new york times for instance, what incentive would the new york times have to hire writers and reporters in the first place? The same goes for most content providers. They would cease to exist, and all content would be from questionable sources without any real professionalism put behind it. Of course it's arguable how much professionalism goes into much of the content produced today, but that's a completely separate issue.
honestly, how many complex libraries are there that have multiple implementations that could work as a drop in replacement for the other? In general this would be considered a waste. Why reinvent the wheel? Most applications today build across countless platforms and frameworks that they're going to require to run. It makes development far cheaper if you use modular reusable code. If Microsoft didn't use trident for rendering help documents, and everything else, people would be pointing out how Microsoft only hires incompetent programmers who can't reuse any code. They should really know better and build modular applications like the smart people developing cool open source project XYZ. The other blast people make is how other applications now rely on MS' trident library to run and how horrible it is.
You do realize that MS was being good to their development community by allowing access to trident and their other interfaces. It's just part of how good software development works. Large complex dependencies end up existing, but you have complex applications that can be written in much fewer lines of code. You know, MS has also integrated their widget set into their apis.... this forces people to use MS widgets by default and it isn't easy to replace those with another set that behave differently. Maybe if MS only allowed developers to have access to a raw window that the programmers completely control you'd be happy. Hell, it would spell the end of MS dominance of the SW industry overnight. MS Windows is so successful because of their many partnerships with developers more than anything. Without applications that can run under windows, they have nothing.
You apparently do not understand the economics behind a brand name. A brand name that someone knows inherently gives something value, even if the owners of the brand did absolutely NOTHING to help make the product. From the article, they mention that the NFL gets 30% of profits from the Madden series. Do you really think the NFL does 30% of the work? Do you really think the NFL does much of anything besides let them use their branding and the statistics from their league? Another football game could come along, and be better than Madden in every conceivable way, yet Madden would likely still sell more because it has the brand names. It has names people recognize in the game, and that is worth a lot. Do you think MIchael Jordan spent time designing Air Jordans? Yet, his name entitled him to a hefty amount of money.
In the same regards, the big labels have brands that the guitar hero games want (songs and artists). If you want to use those brands, you must play by their rules. It's just how things work. I can't tell you what a fair price is for these things, I don't really know. It is possible that the labels are simply asking for too much. It's also possible that their brand is generating enough extra interest in the games that they're worthy of getting extra money for the tracks. It's all a matter of supply and demand, and I'm sure the rockbands will try to pit the labels against each other in this fight. Each of the big labels has hit songs that would work in the game, and I'm sure they'd go heavier from labels offering better prices. However, there are certain songs that could be huge for the game (think the Beatles catalog as someone has mentioned before). These would generate a LOT of extra interest that just wouldn't be there for indie band songs.
Fraid to break the news to you, but this just isn't the case. It is true that the diehard guitar hero players will buy these games no matter what is released on them. And, yes there is a sizeable number of them. However, what sells these games in the massive volumes they've had is name recognition. Guitar hero is one of those party games that people play in groups. Sure there's some fun in sitting around and playing songs you don't know, but when you have a group of 4 or 5 people (where only 2 can play at a time), the others want something they know on. Every time I've played those games, I've looked for songs I recognized, it just makes it easier to play. If only one person knows the song it gives them an unfair advantage when playing the game. Plus, it's just more fun to be "rocking out" to songs that you recognize. Sure the other songs might be just as good, but if you don't know them, you just can't get as "into" the game.
A compromise will likely end up being made between the music industry and the game companies. If the music is the reason these games are selling, then they do have legitimate reasons to want more of the profits made from these games. If you read the article, they even mention that the NFL makes 30% of the profits from the Madden series. I imagine where the big discrepancy lies is in aftermarket content for the game. For instance if the game publishers are charging $3 to download a new song (I have no idea what it really is), and the music industry only gets $1 of it, I can see why they're complaining. This content is almost entirely their own, and they legitimately want some of the profit.
uh.. . I don't know, maybe do what I did back when I was in high school and make plans ahead of time? It's not that difficult to do. Believe it or not but society has been around far longer than cell phones have. Sometimes I think it might be better off without them.
Slashdot Mirror
Funny, is it just me, or are all these inventions that you mention ones that came about long before Europe become the socialist paradise it is today. Innovations that came about 60+ years ago really doesn't seem to suggest that Europe is thriving in innovation. Maybe I'm wrong, it's known to happen, but your post did nothing to back up your assertions.
Phil
ARM has a couple processors already that are pretty high on the performance measurement. For instance the Arm Cortex A9 has a dual issue pipeline, and limited support for out of order processing (similar to the original Pentium processor in that regard). This chip also can contain up to 4 cores, and have up to a 2MB L2 cache. I think they can run up to about 1GHz. They also have full support for floating point and all that good stuff. I'm pretty sure ARM is also working on developing an true OoO processor that will likely be running in the GHz range which would likely be ideal for a netbook.
Remember, with a netbook, you don't gain much by lowering the CPUs power consumption to less than 5 watts or so. The reason for this is simple, the display, ram, hard drives and everything else consume enough power that it won't really help battery life very much. I can imagine though that a quad core ARM A9 at 1GHz would make for a really nice netbook. Having multiple cores is nice on those for web browsing (playing flash in the background of your tabs, etc), and also for many media tasks. It would also be great if they included a graphics chip (or gpu as part of a SoC system) that could handle h.264 decoding for the netbook.
Phil
I'm pretty sure most car loans don't let you put up the car for sale without first contacting the bank. Generally they're going to have to require payment in full, or some other form of security before they'd okay the sale of the car. Basically you don't have the right to sell the car. By the same rules, you have to have certain levels of insurance on a car you buy with a loan. The bank won't allow you to secure your loan with an asset that could become worthless overnight (for instance if you total the car, or get it stolen).
Phil
Just because the ISA supports reading and writing memory in one byte increments doesn't mean the chip physically reads or writes a single byte to the DRAMs. Data is generally read and written to the caches, which will later get written back to memory. Most caches don't bother to contain the state needed to know which bytes in a cache line were modified. Additionally, most memories themselves aren't really good at single bye reads/writes. They operate much faster when reading and writing larger data chunks at a time.
Phil
Wait.... I'm not really sure if you are agreeing with me or disagreeing with me here. You first state that you have an interest in research, and have done some. You then state that you don't think the major research universities are that research specific etc. Believe me, you are getting special attention because of your work, and what you have done. Professors appreciate the bright students, and will go out of their way to help them. The fact that you have lunch with them is wonderful, and happens often enough, but you don't get a normal coursework student doing that often. Most wouldn't want to anyhow. In this regard, you've basically proven my point.
I wasn't trying to say in my post that professors don't care at all about teaching, and that undergrads aren't their concern, but in general that is not their primary concern. There time first and foremost tends to go toward research.
As far as other comments saying that many top schools have coursework only options. This is also true. However, at most major research-one schools, even the coursework option for masters will likely have a strong research angle. You will be expected to read many many research papers in graduate courses, and can expect to do research projects re-implementing other research papers, verifying their results, or sometimes extending upon it and adding to it. If you have no interest in research, and only care about a degree to get better jobs, this path is far from the way of least resistance, and will likely not work for you. You have to be willing to put in a huge sacrifice for it, and it will not be like an undergrad degree but with harder classes. I've seen that at places, but it's much harder to manage at top tier research universities.
Also, while the big name universities degree means more, what really means a lot is the recommendations and contacts you'll meet there. If you don't have the skills to hack it in one of these grad programs, you likely won't survive. Maybe I'm too hasty to judge, but most students who are going to graduate school solely because they want to earn more money aren't the best and the brightest students, and are going to be looked down upon by others. It's not so much that the academics are stuck in their ivory tower, so much as they tend to value knowledge for knowledge's sake. If your sole reason for learning is money, you have the wrong attitude. Obviously, making more money is a factor involved in most people's decision to go to graduate school. But if they are going to a major university for graduate school, chances are good they are interested in learning just because they want to learn. They may have decided it wouldn't be economically feasible to go back for more school if it didn't pay off in the end, but they truly wanted to go to school, and don't just want the money. Anyhow, if you fall into this category and don't want to learn for learnings sake, you will not, and should not be able to get through graduate school at a top research school. The degree is not merely about reciting facts, and memorizing ideas, but rather being able to analyze and come to new conclusions. It should show that you're capable of doing research, even if you haven't already done any real research. These are not skills that all top students have.
Part of it comes down to the fact that how much you get out of the degree is based on what you put into it. Many times there are easy and hard projects that you could do for a class. If you always go the easy route, and do the bare minimum work, you will be able to pass through the degree. However, when someone later interviews you and asks about your experience, you now lack the experience that you're expected to have coming out of one of these institutions (they can find this out pretty easily based on what you told them you have done), and you will likely not have great recommendations. Mostly because unless you really push yourself, you won't be able to stand out and get the attention of your professors. This attention is far mo
Seriously, if your concern for going to grad school is solely to have something on your resume that looks better and gets you paid more, don't go. As a grad student in computer engineering, I can't stand the people who want to get a masters just because it makes them look better. And, if you do get a masters, don't bother getting it at a big name university, because that likely won't mean anything once you get it. The big name universities have the name because of the research they do. The research determines the ranking of their graduate program. If you plan on going just to get a degree, and not do any research, you'll end up shorting yourself of a better education elsewhere, and you'll waste the time of professors and other students who are actually interested in doing research. After graduating from one of these schools it won't really make you look much better either. You'll talk to companies and get in the door for having a big research school's name on your degree, and they'll ask you about what research you did, or ask for recommendations from faculty etc. You likely either won't know any faculty very well (as they're concerned with doing research, and not some masters student who only cares about making more money), or they'll have a low opinion of you for wasting space in their program (that space could have instead been used by someone interested in pursuing research).
Sorry if I sound really negative about this, but this is the truth of academia. The big name schools are concerned with research. That is why they have a big name, and that is what they will focus on to maintain their reputation. They often do not offer a better education, and in fact they are often less concerned with teaching than smaller lesser known schools. The professors just can't afford spending too much time teaching, because in the end (for getting tenure at least), research is what matters. In fact, at many of these schools, it is looked down upon if a junior faculty members wins a teaching award. The rest of the university assumes they're spending too much time on their teaching, and not enough on their research.
My recommendation is to talk to the faculty at your current university. See what they recommend, and be truthful about why you want to go to grad school. Slashdot is not the place to find out about this stuff, most people here have no clue. Also remember that as far as graduate programs at top schools go, it's not really that one school is better than another. In reality its that one school is better in one particular specialty area. The choice of which school is best for you depends much more heavily on what you plan on specializing in rather than the US News ranking. Employers know what schools specialize in, and base decisions on that. If you don't plan on specializing (as you don't seem to be concerned with research), the rankings immediately become relatively worthless. Talk to faculty that you know and trust. They can help you, but you have to show that you're worth spending time on. They likely have more important things to do, and don't want someone wasting their time.
phil
I meant that a coal plant can't change its load in the timescale of hours. They will produce the amount of power they are producing for a while, and they will not be producing more during the day than during the night. They just don't operate that way.
Now, the power plants do establish trends, and forecast power demands so they know how to run their plants. This is a different problem though. The demand trends are to make sure they can meet peak demand, and are less concerned with offpeak hours. There is little that can be done to a coal power plant to allow it to run at say 80% load during the afternoon, and 50% load at night.
I'm not saying conserving power is bad, and that it's a waste to do, just that arguing that you are generating excess CO2 by using this power at night isn't necessarily true.
Phil
Selling power to other areas is fine and good, but requires a few things. First, you need a way to transfer power to the other areas. High voltage power lines however can manage this... the current infrastructure doesn't wasted too much energy to send power a couple hundred miles. However, at least where I live, when it's nighttime, it is also nighttime a couple 100 miles away. When there is peak demand, there is also peak demand a few hundred miles away, or close enough to it. Unless you have installed extremely high voltage DC lines to transfer power 1000s of miles (even then it's not that large of a distance when thinking on a "global" scale, which cost a TON of money to setup, you really aren't going to get far with selling your excess power. Really to make it economical, you'd need a way to transfer your excess capacity 10s of thousands of miles with little loss. The infrastructure to do this effectively would be both enormous, and wasteful, as much of the power generated would be lost in transit. Not to mention that doing this would be dangerous, and likely hurt many people. Networks that hold this much power are dangerous, also they can create a single fail point that becomes a prime target.
Phil
Actually, even the ones powered by coal are likely not wasting much CO2. Considering a machine is most likely to be sitting idle at night, and that the coal plants have to operate 24/7 (they can't dynamically lower their power output, that's provided by secondary sources during the afternoon). Power usage generally peaks in the afternoon, and so other power generation stations (those like natural gas that can be brought online quickly) handle the peak load, but, as coal power is cheaper, they try to get as much as possible from the coal. If the base load provided by the coal is greater than the power being consumed, than any additional power demanded isn't really "wasting" electricity. It's just using electricity that has already been generated. Of course, if this amount is great enough to change the power plants operating conditions, it does matter, and as far as the businesses are concerned, this power does cost money, and quite a bit of it.
However, saying the plant is releasing more CO2 for these computers is generally not true.
Phil
there might be some truth to what this guy is saying in that link, but i find parts of it to be wrong. For instance, he's talking about heap fragmentation, and not actual memory fragmentation. This should be an issue handled on the application level rather than the OS level. The article also mentioned sending a request for 256 bytes to the OS and failing..... Unless he's working on some machine I don't know of, you can't really just request 256 bytes of memory from the OS. The OS allocates pages of memory (4K is the standard size on the x86, although larger sizes exist), and can give applications pages of memory, but can't allocate space in smaller increments. The reason for this is memory protection. If byte 0 of a page is readable by an application, so is byte 4095. Same with write and execute permissions. Generally, an application should handle the heap in the user space. A request for 256 bytes will check available memory to the application and see if it can find it in a free list. If it can't the heap manager should request another page (or more depending on allocation policies) of memory, and then service the heap request. The link seems to confuse some of these points.
Phil
What the hell are you talking about here? It might fail to allocate a 256 MB block if the machine doesn't have enough memory, or if the program decoding the module is running in the kernel and using kmalloc, but for the most part, applications do not have to worry about memory fragmentation. Virtual memory takes care of fragmentation for you, as only 4KB pages need to be contiguous.
The only time this wouldn't work is if the application that you're running doesn't have 256MB of its address space free. Unless the application is using close to the 2GB or 4GB address space the application is given this shouldn't be an issue.
Phil
unfortunately, the overhead of pipe operations render many chains of operations slow enough that the parallelism gained by using multiple cpu's can be eaten up. Especially when some of the pipe operations perform extremely simple operations on data. These don't tend to scale too well, and the overhead will often eat up all the benefit of multithreading. Additionally pipes have their place, but can fail when sub-channel data is needed, or when multiple pipes are needed (multiple input or output streams, pipes work well for problems with a single input source and a single output destination). Also there are some operations in a long pipe chain that are likely to be the most expensive by far of any of the operations. Often these operations could be parallelized, but the program performing the operations doesn't realize it. This leads to inefficiencies, because if 95% of your time is spent in one operation, Amdahl's law says that you could throw an infinite number of processors at the problem, and only speed up your application by 5%.
What you should look into if the ideas behind pipes seem logical for multiprogramming is the use of stream-based languages. I know StreamIt was one of the early ones developed. These languages tend to take the pipe paradigm to a whole new level, allowing splits in a pipe, joins, etc. Also, the idea of futures (I think they were introduced in Scheme, but I could be wrong) is something that really needs to be enabled in more programming languages. A future is a subroutine or function that is "called" at one point, but the system knows it doesn't have to be evaluated until any of its outputs are requested by the application. With futures you could trivially launch 10 "parallel" jobs at once, and the system decides how to schedule them. Of course, a big problem with these is that they tend to fail pretty bad in a language like C where a compiler doesn't really know where outputs of functions can be later used.
Phil
yeah... because all code from '01 will compile with gcc 4.0... hell, I bet a lot would have issues with gcc 3.3. Not to mention include files have changed, installed libs on the system are different etc. I'm still working on (actively developed) code now that doesn't compile with g++ 4.0 (granted the codebase is part of a research project so their priorities are different). I've also used multiple other research programs developed in the late 90s or early '00s, and many of them require quite a bit of tweaking to get working. Nothing like fixing 100 errors brought about by changes to how gcc implements the standards, and where headers are included, what is included in what headers etc. But if you consider that stuff straightforward, be my guest.
Phil
Hell, both are pretty nasty at times. Have you ever tried installing software that's 5-10 years old from source on a linux machine. Oh wait, the source relies on ancient versions of gcc, and doesn't work with libraries you currently have installed, etc. It's definitely not a trivial problem. If you download all the old compilers and libs first, you might manage to get it working, but this is far from a simple solution.
Phil
Isn't one of the points of using DLLs to reduce the size of programs on the machine. If an application is using 3rd party libraries, it often just makes sense to install those libraries into the windows directory because.... THATS WHERE THEY BELONG! If you were installing a linux application that relied on libraries, you would not want to install the libraries into a folder where only that single application could use it. You'd install it somewhere global so all applications can use them, and you don't need multiple copies of them.
Of course, one could say it's not an application's job to install its support libraries, but that's not really an option, as users want to be able to install an application and have things work without first having to install 10 other libraries (the same reason on Debian systems people use apt). Unfortunately as there's no standardized windows install setup, things are a bit messier.
Phil
Just because wifi works fine does not mean the CPU is not the bottleneck. I do not own an iphone, and don't know how the implementation of the radio stack on it, but it is entirely possible that stages of the data reception/transmission are handled by the CPU (or possibly a separate DSP) on the iPhone. In this case, if this CPU is not powerful enough, the iPhone can't max out the 3G bandwidth. Just because a different protocol that uses an entirely different radio stack that requires different processing power works fine does not mean all radio stacks do.
phil
Am I the only one who found this quote from the article ironic:
"Although it now stretches to fewer than 100 sites there is still fun to be had in the text-only web, providing your web browser still supports Gopher."
I think the author of this article either doesn't know what he's talking about or got confused. Gopher sites are not part of the web at all, and definitely aren't part of some "text-only" web. Maybe text-only gopher, but definitely not the web.
Phil
I wish I could say that I haven't seen cops pull you over for 5-10 over the limit, but I have seen it first hand. I was once driving across North Dakota, and going 80 in a 75 mph zone. When the cop pulled me over my first thought was "huh, what did I do?" An hour later and a search of my car later the cops let me go with a $25 ticket for speeding, and a $45 ticket for having a box in the back seat of my car (i was obviously moving and my car was loaded up with stuff) that had an opened bottle of booze in it (unless I spent 10 minutes pulled off to the side of the road trying to get to it there was no way I could without first running my car off the road). What I learned that day was an unshaven, wired on coffee 20-something person of hispanic descent driving with out of state plates is going to get pulled over for whatever reason a cop likes. They will then justify searching your car with stuff they found while searching your car (in my case this was the truth, the report cited that the cop had seen an open container and wanted to search the car when in fact I told the officer that yes there was a box with booze in it in my back seat when he asked if he could search my car and asked if I had anything to tell him. Of course at the end of the day the fines were relatively light, and it really made an amusing story. Especially when they put me in the back seat of the cop car and one cop started demanding I tell him where the drugs are. I'm not sure if I could keep a straight face at that point, it was just so surreal.
Of course, I don't see speeding as a victimless crime. Sometimes it can be (like the few times I've been on a completely deserted highway and just wanted to see how fast I could push it), but often it isn't. It's the asshats who fly down neighborhoods and through town that tend to endanger people. I just wish I could say that they're the ones who get punished the most for it.
Phil
The president of the united states also said he wasn't going to appoint any lobbyists on his cabinet. We saw how long that lasted (until he started announcing who was on his cabinet). People have a false sense of belief that just because a politician has been spouting on about hope and change for years that he will actually do it. Believing that ignores years of political precedents. Obama will likely change things, but for the most part, his administration has largely been a continuation of Bush's policies. The only major exceptions I can think of are funding to abortion groups, and funding for stem cells. People will point to Guantanamo, but it was likely going to get closed down within a year or two as people came up with a good solution for how to deal with terrorists (in reality we'll likely return to pre-9/11 conditions where we simply hired other countries to tor^k^k^Kinterrogate them for us).
I see no reason to suspect that the Obama administration will act any differently than the Bush administration concerning net neutrality.
Phil
Mod this poster up. While companies pay credit card handling fees, handling cash is often more expensive. While banks generally don't charge them for depositing, etc, you have to have places to put the cash, deliver it to a bank, etc. All this costs money. More importantly is the concern over employee theft. If most everyone pays in credit there isn't much cash in the drawer to hide the fact that someone stole a $20. Most stores do checks at the end of the night, but don't care too much if its off by a few dollars as employees make mistakes. The more money in cash they collect the more mistakes are allowed. Credit means that the exact amount is charged, and the employee has no easy way to take the money.
Plus of course there's the fact that credit cards tend to move lines faster. If this means a store can have only 4 lines open instead of 5, they're saving money right there.
Phil
Actually, in this instance it can be considered theft. Copying a cd for personal use isn't necessarily theft, but copying a cd, and then selling burned copies of it for profit is definitely very close, if not indistinguishable from theft. It's one thing to illegally use someone else's content. It's another altogether to use someone's content as your own while they pay the cost to distribute it. Companies do have legal rights to prevent others from doing this, and they can and should use them. If there was no such thing as copyright, information might exist, but it would be in a wide range of disarray and it would be hard, if not impossible to determine which sources you can trust. If some other website could republish all the content on the new york times for instance, what incentive would the new york times have to hire writers and reporters in the first place? The same goes for most content providers. They would cease to exist, and all content would be from questionable sources without any real professionalism put behind it. Of course it's arguable how much professionalism goes into much of the content produced today, but that's a completely separate issue.
Phil
honestly, how many complex libraries are there that have multiple implementations that could work as a drop in replacement for the other? In general this would be considered a waste. Why reinvent the wheel? Most applications today build across countless platforms and frameworks that they're going to require to run. It makes development far cheaper if you use modular reusable code. If Microsoft didn't use trident for rendering help documents, and everything else, people would be pointing out how Microsoft only hires incompetent programmers who can't reuse any code. They should really know better and build modular applications like the smart people developing cool open source project XYZ. The other blast people make is how other applications now rely on MS' trident library to run and how horrible it is.
You do realize that MS was being good to their development community by allowing access to trident and their other interfaces. It's just part of how good software development works. Large complex dependencies end up existing, but you have complex applications that can be written in much fewer lines of code. You know, MS has also integrated their widget set into their apis.... this forces people to use MS widgets by default and it isn't easy to replace those with another set that behave differently. Maybe if MS only allowed developers to have access to a raw window that the programmers completely control you'd be happy. Hell, it would spell the end of MS dominance of the SW industry overnight. MS Windows is so successful because of their many partnerships with developers more than anything. Without applications that can run under windows, they have nothing.
Phil
You apparently do not understand the economics behind a brand name. A brand name that someone knows inherently gives something value, even if the owners of the brand did absolutely NOTHING to help make the product. From the article, they mention that the NFL gets 30% of profits from the Madden series. Do you really think the NFL does 30% of the work? Do you really think the NFL does much of anything besides let them use their branding and the statistics from their league? Another football game could come along, and be better than Madden in every conceivable way, yet Madden would likely still sell more because it has the brand names. It has names people recognize in the game, and that is worth a lot. Do you think MIchael Jordan spent time designing Air Jordans? Yet, his name entitled him to a hefty amount of money.
In the same regards, the big labels have brands that the guitar hero games want (songs and artists). If you want to use those brands, you must play by their rules. It's just how things work. I can't tell you what a fair price is for these things, I don't really know. It is possible that the labels are simply asking for too much. It's also possible that their brand is generating enough extra interest in the games that they're worthy of getting extra money for the tracks. It's all a matter of supply and demand, and I'm sure the rockbands will try to pit the labels against each other in this fight. Each of the big labels has hit songs that would work in the game, and I'm sure they'd go heavier from labels offering better prices. However, there are certain songs that could be huge for the game (think the Beatles catalog as someone has mentioned before). These would generate a LOT of extra interest that just wouldn't be there for indie band songs.
Phil
Fraid to break the news to you, but this just isn't the case. It is true that the diehard guitar hero players will buy these games no matter what is released on them. And, yes there is a sizeable number of them. However, what sells these games in the massive volumes they've had is name recognition. Guitar hero is one of those party games that people play in groups. Sure there's some fun in sitting around and playing songs you don't know, but when you have a group of 4 or 5 people (where only 2 can play at a time), the others want something they know on. Every time I've played those games, I've looked for songs I recognized, it just makes it easier to play. If only one person knows the song it gives them an unfair advantage when playing the game. Plus, it's just more fun to be "rocking out" to songs that you recognize. Sure the other songs might be just as good, but if you don't know them, you just can't get as "into" the game.
A compromise will likely end up being made between the music industry and the game companies. If the music is the reason these games are selling, then they do have legitimate reasons to want more of the profits made from these games. If you read the article, they even mention that the NFL makes 30% of the profits from the Madden series. I imagine where the big discrepancy lies is in aftermarket content for the game. For instance if the game publishers are charging $3 to download a new song (I have no idea what it really is), and the music industry only gets $1 of it, I can see why they're complaining. This content is almost entirely their own, and they legitimately want some of the profit.
Phil
uh.. . I don't know, maybe do what I did back when I was in high school and make plans ahead of time? It's not that difficult to do. Believe it or not but society has been around far longer than cell phones have. Sometimes I think it might be better off without them.
Phil