Okay, this might sound like a troll, but hear me out. I'm not interested in getting into a discussion about abortion, but it makes a pertinent example, regardless of whether you are opposed to it or not.
I have a hard time believing that chimps would be granted any rights in today's society, especially considering that roughly half of the population argues in favor of a woman's "right" to have her unborn child killed. If the rights of an unborn human child are so small that they may be outweighed by the convenience of the mother, I fail to see how a chimp's right to life would ever take precedence over the possible value of the medical research obtained.
Abortion doesn't cure disease - in fact, it is, more or less, last-resort birth control. If you can't convince society to respect human life, I doubt you'll be able to convince them that medical research should be halted so that chimps can be spared. After all, at least the medical research has the potential of providing cures for disease someday.
I'm not trying to troll here - you can believe what you want with respect to the merits of abortion. That's not the issue. The issue is that in order to convince people to give animals the same rights as humans, you are going to have to offer a compelling case for doing so. People (sadly) aren't interested in the moral arguments, and the arguments against giving animals rights are strong:
Be prepared to be called an enemy of science. Much scientific and medical research depends on using animal subjects.
Be prepared to be called uncompassionate toward humans - after all, without animal subjects, you delay the cures for things like cancer.
Be prepared to be called a corporate shill or anti-environmentalist. Because many chemicals are discovered toxic by testing on animals first, the lack of testing would allow corporations even greater freedom to dump environmentally damaging chemicals into the environment.
Be prepared to be accused of attempting to force your private morality on the public.
Lastly, why would we grant rights to animals when we are taking them away from humans? Things like the elimination of habeas corpus, government sponsored torture, indefinite detention, mandatory abortion, and summary execution, etc... are all on the horizon and are far more pressing issues than that of animal rights.
It isn't an easy subject to take on. Granted, we shouldn't ever intentionally inflict pain on living things, but then, how would we eat? There are vitamins and minerals our bodies need which are only present in living things. So without a binding set of moral principles, the debate is going to remain centered around the pragmatic aspects, and I doubt this will result in any action being taken.
After all, the Democrats successfully convinced Americans that it is wrong to "imposing your private view of morality on the general public". Given this is considered evil, how could one convince the general public that your particular moral imperative applies to the public at large? Isn't religion supposed to be a private thing now? (I suppose we could get involved in the related discussion about private versus public morality, and how law reflects the morality of the public at large, for better or for worse.)
You forget that according to the DMCA, someone who issues a bogus complaint (aka takedown notice) is subject to civil liability on behalf of the affected party. In this case and, IIRC, Cringely (or PBS, whoever owns the copyright) could sue the issuer of the takedown notice for $25,000.
You know what really gets me about the whole thing?
I'm not opposed to copyright. And I think artists and composers should be able to make a living by their craft.
But the problem is that by having copyright as an ironclad right, we only reward the greediness of the media conglomerates, who, while not actually producing the art or music, make the most money from it.
After the digital revolution, how are we supposed to get back to the former model of paying per performance? On one hand, it isn't fair to pay the artist every time you listen to their work, because they aren't performing it again. On the other hand, it isn't really fair to the artist that those who never paid for the performance can enjoy it as if they had. And then there are the rights of the listeners, who would like to enjoy the music they buy in the manner they like, without being restricted unnecessarily by DRM. And then there are the students, who cannot learn about current music because it is still copyrighted, and the teachers cannot circumvent the DMCA for any purpose.
I don't have the answer to the problem, but I know that the current copyright system is broken - the only ones who benefit are the corporations. Instead of actually enlarging the public domain and enriching culture and providing a living for the artist, today's copyright only impoverishes artists and, thanks to the DMCA, will never functionally enter the public domain. Our culture is simply lost to our fading memories. Copyright as it exists today doesn't benefit either the artist or the fan.
The impression I get from YouTube is that it is more about sharing home videos than infringing content. Everything about the site seems to indicate their intention was to encourage amateur filmmaking:
The movies are transcoded into a lossy, low resolution format. Someone wanting a high-quality pirated copy of Star Wars knows to look elsewhere.
The plethora of bad home videos seems to suggest that it is more about information sharing than copyright infringement.
They have taken down videos upon request from the studios, and in some cases, (think: Viacom) even took down content that wasn't infringing.
I really don't want to live in a society where public debate can be effectively stifled by the mere accusation of copyright infringement. I understand that media companies want to maintain their profits, but there comes a point at which the restriction of distribution has no marginal commercial value. YouTube is that point - their format is low res, low framerate, and generally unsuitable for anything but public discussion.
I think we'd all be better off if copyright infringement was restricted to only to cases where the plaintiff could prove substantial monetary damages. After all, (to quote Mark Cuban) everybody knows that nobody is losing money over clips on YouTube, so it shouldn't matter what gets posted.
** Yes, I know some people watch tv episodes via YouTube, but quite frankly, these folks wouldn't buy cable anyway. I mean, why would someone pay $50 a month for broadband internet so they can watch YouTube at 1/4 the resolution and 1/2 the framerate of regular tv? They'd be better off buying cable because at least they'd get to watch their shows without squinting.
Rather than argue over what is and what isn't pornography, why not just setup a.kids domain which is explicitly for children?
That way, those seeking to register a.kids domain would have the onus of proving their material was appropriate for kids. (Not that this is difficult). With the.xxx domain, every.com.net.org, etc... site has the burden of proving they don't belong in the.xxx domain. But, if the opposite approach is taken, only those sites specifically meant for children will have the burden of proving their content appropriate for children, and we can leave the rest of the internet as is.
I'm not sure why politicians keep on beating a dead horse when neither liberals nor conservatives want a.xxx domain. And "protecting the children" is as simple as giving them their own TLD, rather than trying to disrupt the internet as we know it.
1/10000 performance difference to Core 2 Duo is actually quite true.
Not remotely. Even the 8086 didn't spend more than a few dozen clock cycles executing the most complicated instructions. IIRC, the worst instruction could take 28 cycles. But the typical instruction took between 2 and 3 cycles. So an 8086 clocked at Core 2 speeds would only be slower by perhaps a factor of 10 or 20 in the average case.
But I think my point was more along the lines that the silicon on the die would be better used building CPU logic than making up for the shortcomings of the memory subsystem. As someone else has posted, the majority of the transistors on the die are used for the cache and memory controller, and glue logic. Why not make the CPU simpler, and use a better memory architecture?
The goal is to establish an environment of fear, such that most users are afraid to anything other than what the media conglomerates say is okay.
And that, I believe, is the problem. I can no longer buy RIAA-parent-company DVDs and CDs in good conscience, because I know a portion of the proceeds will be used for suing women and children. And I think the impressions being formed are overwhelmingly negative. Where are the college students protesting, "Save the RIAA!"? Teenagers and college students are starting to believe that buying CDs will only fund lawsuits against the defenseless and poor.
I understand their position regarding copyright infringement. But it is infringement, not theft, and certainly not murder on the high seas. Their strong-arm tactics make the entire industry look bad, and call into question the legitimacy of their cause. When such an entity chooses to pursue lawsuits on such frivolous evidence, one can only conclude that the real goal is not justice but merely the acquisition of additional wealth. First, they steal from the artist through oppressive and one-sided contracts, and now they are trying to extort money from those whom they believe will not have the resources to resist. When was the last time the RIAA filed a lawsuit against a millionaire?
It just makes me sick. And the artists, of all people, stand to lose the most. Instead of buying from big labels, I've begun looking at the smaller, independent artists precisely because of the RIAA tactics.
I wonder if they even considered the fact that dropping CD sales might be related to people unhappy with the fact that they are suing their customers. Unhappy customers tend not to be repeat buyers.
Do you have any idea how large a 1GB dual-port stick of ram would be? It'd probably cost 2 to 3x as much to make
I realize that many people share your opinion, and that it is a valid point. However, I would add that it is people like you who hold back the desktop PC. When everything is about low cost, does it surprise anyone that the standards and hardware produced are optimized not for performance, but for low price?
One thing which impressed me about the designs of IBM mainframes and of the Apple computers (prior to Intel) was that they were absolutely beautiful from an architectural standpoint. When you have clients willing to pay a decent price for your computers, you can build them the way they should be built. (Of course, this is just my opinion).
Instead, we all buy PCs because they're cheaper, and then gripe about the low performance.
Granted, their opinions do not affect their legal standing. I'm not sure why you interpreted my post as stating so. My point was that there's a cognitive dissonance in claiming the wrongfulness of copyright infringement of your own work while simultaneaously infringing the copyright of others. IOW, are these the same students who illegally share music? And if so, why are they being hippocrits?
So I wonder if these same students consider the "cataloging" and sharing of copyrighted music just as infringing?
If it isn't infringing to share music via P2P, then why would it be infringing for school staff to share papers? Especially when in the latter case, they are doing it with the explicit purpose of preventing copying of the students work? Presumably, a student who finds his hard work copied by another student would have a copyright infringement case against the other student, provided that it was indeed copied without permission. After all, high school students would never stoop so low as to allow others to copy their work.
I'm thinking that more than a few of them have downloaded and shared music under the justification of "Maybe I'll buy it later... if I like it." But for some reason, it's only considered copyright infringement when it is their work being copied.
After all, if it isn't your work being copied, it's sharing, right?
And your 5000 transistor processor would have 1/10000th the MIPS of that core 2 duo.
BTW the core 2 duo comes with 4MB of L2 and 128KB of L1. That's 207,618,048 transistors for the SRAM cells
OMG go take a course on processor architectures
I'm stunned, really, I am. I'm not quite sure if you are deliberately trolling or just didn't bother to think before posting.
In the first place, as someone who has presumably had a course on computer architecture, you'd know that even a non-pipelined processor can reasonably execute one half to one instruction per clock cycle. With each core on the Core 2 duo issuing up to 4 instructions per clock cycle, the MIPS of those little cores would be closer to 1/8 to 1/4 that of Core 2 Duo core. The 1/10000th figure you suggest is off by at least 3 orders of magnitude.
Next, there are only 151 million transistors on the die, so I doubt Intel is using 6 transistors per bit for SRAM, as your numbers indicate. Functional SRAM cells can be made with 2 transistors per cell, which is probably closer to the design used by Intel.
And the OMG part just wasn't warranted. Please do explain how having a memory controller on die reduces latency between the processor and memory. Either way, the processor has to go through the memory controller, so I'm not sure how having it on the die, as opposed to on the motherboard, changes inherent latency. (OTOH, if having it on die allows the processor to schedule bus access more appropriately, then it would reduce latency, albeit through maintaining better control of the bus - the reason I thought I made clear in the original post).
And I agree with you that main memory is slow, and perhaps we should be looking into solving that problem, rather than figuring out how to add yet more complexity to the processor.
I'm well aware of the cost issue, but if I had to choose between 256 MB of dual ported RAM and 1 GB of single ported, I'd pick the dual ported RAM every time.
Do you have any idea how large a 1GB dual-port stick of ram would be? It'd probably cost 2 to 3x as much to make.
Versus, say, putting another core on the die? The Core 2 Duo is 151 million transistors. According to Tanenbaum, a rudimentary (non-pipelined, non-cached) microprocessor can be put together for about 5000 transistors. You can do the math: with the transistor budget of the Core 2 Duo, you could have 3000 processor cores.
What it comes down to is that a large part of the die is being used to manage shortcomings elsewhere in the system. For example, the SRAM cache is used to compensate for the fact the DRAM families of main memory require frequent refreshes by the memory controller, and are frequently unavailable during DMA transfers from elsewhere in the system. The latency of DRAM and SDRAM type memories can be in the dozens of clock cycles, even under ideal circumstances (the design trades latency for throughput). And placing the memory controller on the die was done to address precisely the problem I mentioned: something, somewhere else in the system commandeers the bus during a critical time, and the processor ends up stalling, waiting on access to main memory. The problem isn't with the processor; it is doing its job - it's with the design of the rest of the system, in which the CPU is literally, just another peripheral.
But we shouldn't fix a problem in the memory system by adding a kludge to the processor. And that is exactly what this is. Consumers buy the cheapest RAM they can find, without regard for how fast, or how well designed, the rest of the system is.
And yes, I understand the cost issue. But who really uses 1 GB of memory? Maybe 5 percent of users out there. The rest of them just buy the 1 GB box because, (Joe Sixpack voice:) "Hey, it's got more than that other box..." I doubt if most users could even give you a ballpark figure for the memory used by their applications.
In fact, if most users knew how much memory they actually used, we wouldn't be having this discussion. Instead, users would demand faster memory, rather than more of it.
It is interesting to note that Intel has now decided to put the memory controller on the die, after AMD showed the advantages of doing so.
However, I'm a little dismayed that Intel hasn't yet addressed the number one bottleneck for system throughput: the (shared) memory bus itself.
In the 90's, researchers at MIT were putting memory on the same die as the processor. These processors had unrestricted access to its own, internal RAM. There was no waiting on a relatively slow IDE drive or Ethernet card to complete a DMA transaction; no stalls during memory access, etc...
What is really needed is a redesign of the basic PC memory architecture. We really need dual ported RAM, so that a memory transfer to or from a peripheral doesn't take over the memory bus used by the processor. Having an onboard memory controller helps, but it doesn't address the fundamental issue that a 10 ms IDE DMA transfer effectively stalls the CPU for those 10 milliseconds. In this regard, the PC of today is no more efficient than the PC of 20 years ago.
It has long been known that certain racial and ethnic groups have a certain gait, as a matter of simple conformance with their culture. It shouldn't surprise anyone that a child grows up imitating their parents, including their manner of walking.
So, by teaching computers to recognize a person's gait, and single out people based on their gait, we have, in essence, taught computers to give greater scrutiny to individuals of a particular ethnic or racial group.
In other words, we've programmed computers to be racist.
While it is arguable whether this was the intent or not, the fact is the effect is the same. It was bad enough when a person was racist, but at least there was the hope of a defense in exposing the racism of an officer. How do you expose racism in a computer program, when computers are presumably incapable of racial bias or personal feelings?
Pthreads has been out for a while. It is open source, and runs on Linux, Windows, and Mac(?).
Whether or not you believe concurrency should be an explicit library or a matter of compiler extension is a bit of a religious argument. But pthreads does offer the functionality, and works fairly well.
First of all, I, and many others before me, have been writing multithreaded applications for years in the likes of Linux and UNIX. I have had to maintain multithreaded applications created by others. My collective experience tells me:
It is not trivial.
Let me repeat: It is not a trivial task. Even if you have libraries and an API which abstracts out the ugly stuff, you still have the problem of concurrency, proper locking, deadlocks, etc...
The majority of problems with using multithreaded programming come not from "ugly" parts of the OS/API layer, but from a misunderstanding of the problem. A few problems in computer science - particularly in the physical sciences - do benefit from multithreading. And it is easier to use threads when writing a game than just to execute all of the IO in one big loop (Hello DOS!). But for most applications, using threads is not only unnecessary, but overkill, and introduces the possibility of yet another class of bugs for which the application must be tested. Furthermore, as deadlock and race conditions are often timing related, they are the most difficult type of bug to find and fix. Finding and fixing this class of bugs is still somewhat of a black art in the industry, and is highly dependent on the skill and experience of the programmer.
In short, unless your system/application design cannot do without multithreaded programming, it is best not to use it. Even with a glossy API, you still cannot escape the fact that debugging a multithreaded application is an order of magnitude more difficult than a single threaded one. In any case, you shouldn't be using threads just because you can.
If you aren't happy with where you're at, at least figure out where you'd be most happy working. Talk to people - new people. Give yourself a year to figure out where you want to be in 10 or 20 years from now, and work out a plan to get there.
But just dropping employment probably won't be a good idea. It may very well limit your employment options later on. Besides, not having money will depress you anyway. And even if you aren't happy with your current job, it might be alright if you know you're working toward something better. The trick is finding that better thing.
I can tell you the reason why the industry needs a subsidy:
No one wants to pay for DRM.
The market hasn't failed. Rather, the content companies have begun to realize that people don't want to pay more to get less.
I mean, why would I buy a tv with fewer features?
The content companies have begun to realize that they need to provide some kind of short-term incentive to get the customer to give up the rights to which they have become accustomed. Once the first generation grows up without the ability to record tv, they'll think it's normal. And the worst of it is that it isn't the content companies paying the bill, but the American taxpayer!
With DTV, the public domain goes away. DRM isn't there to prevent some content from being rebroadcast; it is there to keep all content away from the net. Even things legally in the public domain.
Call your Senator and tell him to oppose this bill. Tell him you don't want Congress wasting money...
There are a few people who are focused on "how do I make this better?
That's because a computer programmer's time is a lot more expensive than the machine on which his code runs. In some of my computer science classes, I was told flatly to ignore efficiency because it was cheaper for business to buy a faster computer than it was for me to optimize the algorithm in assembly.
This is just the classic conflict between engineering and business - the business wants a shed, and the engineer wants to build a cathedral.
The problem with his, analysis, and yours, is the mistaken assumption that the majority of the opinions expressed constitute the majority of opinion.
There are a lot folks like me who have found Linux useful, use it for what it is good at, and refrain from getting involved in the religious wars in the mailing lists.
If you can't rationally discuss something in the "Linux community", you're in the wrong community. Consider security, for example. I happen to know that both the NSA and RedHat chose to offer alternative security models in their Linux versions. Yet I don't recall any religious arguments over the relative merits of the respective models. Instead, people seemed to comment on the pros and cons of each model when the issue came up - or maybe I just didn't read slashdot one day...
And I think the best part about Linux, that everyone seems to miss, is the honesty.
When someone tells me that something works in Linux, I can trust that statement to be true, because they usually speak from a position of experience and impartiality. Contrast this with the proprietary model of software, where someone's livelihood is at stake if they don't sell enough licenses. I simply can't trust any advice given about Windows compatibility because there is always exists a monetary incentive for Microsoft to lie about the quality of their software.
I can much more easily accept software that doesn't work quite right if I know what I'm getting from the outset. OTOH, it is particularly galling for a large company, like Microsoft, to claim their OS works, and experience otherwise after having spent a considerable amount of time and money in useless frustration. At least there wasn't any deception with Linux, and I don't bang my head against a wall trying to force it to do something for which it wasn't designed.
See how you can spend an extra 130 british pounds, and still have no working webcam!
I can show you how you can spend no money, install Linux, and not have a working webcam, WiFi, and be stuck at 1024 x 768 because X doesn't recognize your video card!
And for a bonus, I can show you how to spend half the night recompiling SCSI drivers into your kernel so that you can actually use that old SCSI drive which has been collecting dust.
I understand that the search engines are private entities, and can do as they please regarding what they choose to advertise. However, I'm not entirely comfortable with private ownership of the medium used for public discourse. There are opinions which are never heard in America simply because the media is corporate owned and simply refuses to publish said opinions.
I could care less what private entities do with their own networks, but our public discourse should not be limited to a privately-owned medium.
When religion is wrong, it's not possible to discover and fix it.
Tell that to Jesus. He fixed the misconceptions in the religion of his day.
It's not that "fixing" religion is impossible, it's just that some religions - like Christianity - have had a long time to hone their doctrine. Finding a mistake in Christian doctrine would mean that you have more insight than the 2,000 years of scholars who have come before you. Possible, but not likely.
And, from time to time, some religions with errors do arise. Remember Jim Jones? Well, when things like this happen, they usually correct themselves within a few generations.
Slashdot Mirror
Okay, this might sound like a troll, but hear me out. I'm not interested in getting into a discussion about abortion, but it makes a pertinent example, regardless of whether you are opposed to it or not.
I have a hard time believing that chimps would be granted any rights in today's society, especially considering that roughly half of the population argues in favor of a woman's "right" to have her unborn child killed. If the rights of an unborn human child are so small that they may be outweighed by the convenience of the mother, I fail to see how a chimp's right to life would ever take precedence over the possible value of the medical research obtained.
Abortion doesn't cure disease - in fact, it is, more or less, last-resort birth control. If you can't convince society to respect human life, I doubt you'll be able to convince them that medical research should be halted so that chimps can be spared. After all, at least the medical research has the potential of providing cures for disease someday.
I'm not trying to troll here - you can believe what you want with respect to the merits of abortion. That's not the issue. The issue is that in order to convince people to give animals the same rights as humans, you are going to have to offer a compelling case for doing so. People (sadly) aren't interested in the moral arguments, and the arguments against giving animals rights are strong:
It isn't an easy subject to take on. Granted, we shouldn't ever intentionally inflict pain on living things, but then, how would we eat? There are vitamins and minerals our bodies need which are only present in living things. So without a binding set of moral principles, the debate is going to remain centered around the pragmatic aspects, and I doubt this will result in any action being taken.
After all, the Democrats successfully convinced Americans that it is wrong to "imposing your private view of morality on the general public". Given this is considered evil, how could one convince the general public that your particular moral imperative applies to the public at large? Isn't religion supposed to be a private thing now? (I suppose we could get involved in the related discussion about private versus public morality, and how law reflects the morality of the public at large, for better or for worse.)
You forget that according to the DMCA, someone who issues a bogus complaint (aka takedown notice) is subject to civil liability on behalf of the affected party. In this case and, IIRC, Cringely (or PBS, whoever owns the copyright) could sue the issuer of the takedown notice for $25,000.
You know what really gets me about the whole thing?
I'm not opposed to copyright. And I think artists and composers should be able to make a living by their craft.
But the problem is that by having copyright as an ironclad right, we only reward the greediness of the media conglomerates, who, while not actually producing the art or music, make the most money from it.
After the digital revolution, how are we supposed to get back to the former model of paying per performance? On one hand, it isn't fair to pay the artist every time you listen to their work, because they aren't performing it again. On the other hand, it isn't really fair to the artist that those who never paid for the performance can enjoy it as if they had. And then there are the rights of the listeners, who would like to enjoy the music they buy in the manner they like, without being restricted unnecessarily by DRM. And then there are the students, who cannot learn about current music because it is still copyrighted, and the teachers cannot circumvent the DMCA for any purpose.
I don't have the answer to the problem, but I know that the current copyright system is broken - the only ones who benefit are the corporations. Instead of actually enlarging the public domain and enriching culture and providing a living for the artist, today's copyright only impoverishes artists and, thanks to the DMCA, will never functionally enter the public domain. Our culture is simply lost to our fading memories. Copyright as it exists today doesn't benefit either the artist or the fan.
The impression I get from YouTube is that it is more about sharing home videos than infringing content. Everything about the site seems to indicate their intention was to encourage amateur filmmaking:
I really don't want to live in a society where public debate can be effectively stifled by the mere accusation of copyright infringement. I understand that media companies want to maintain their profits, but there comes a point at which the restriction of distribution has no marginal commercial value . YouTube is that point - their format is low res, low framerate, and generally unsuitable for anything but public discussion.
I think we'd all be better off if copyright infringement was restricted to only to cases where the plaintiff could prove substantial monetary damages. After all, (to quote Mark Cuban) everybody knows that nobody is losing money over clips on YouTube, so it shouldn't matter what gets posted.
** Yes, I know some people watch tv episodes via YouTube, but quite frankly, these folks wouldn't buy cable anyway. I mean, why would someone pay $50 a month for broadband internet so they can watch YouTube at 1/4 the resolution and 1/2 the framerate of regular tv? They'd be better off buying cable because at least they'd get to watch their shows without squinting.
If the goal is to protect children?
Rather than argue over what is and what isn't pornography, why not just setup a .kids domain which is explicitly for children?
That way, those seeking to register a .kids domain would have the onus of proving their material was appropriate for kids. (Not that this is difficult). With the .xxx domain, every .com .net .org, etc... site has the burden of proving they don't belong in the .xxx domain. But, if the opposite approach is taken, only those sites specifically meant for children will have the burden of proving their content appropriate for children, and we can leave the rest of the internet as is.
I'm not sure why politicians keep on beating a dead horse when neither liberals nor conservatives want a .xxx domain. And "protecting the children" is as simple as giving them their own TLD, rather than trying to disrupt the internet as we know it.
1/10000 performance difference to Core 2 Duo is actually quite true.
Not remotely. Even the 8086 didn't spend more than a few dozen clock cycles executing the most complicated instructions. IIRC, the worst instruction could take 28 cycles. But the typical instruction took between 2 and 3 cycles. So an 8086 clocked at Core 2 speeds would only be slower by perhaps a factor of 10 or 20 in the average case.
But I think my point was more along the lines that the silicon on the die would be better used building CPU logic than making up for the shortcomings of the memory subsystem. As someone else has posted, the majority of the transistors on the die are used for the cache and memory controller, and glue logic. Why not make the CPU simpler, and use a better memory architecture?
The goal is to establish an environment of fear, such that most users are afraid to anything other than what the media conglomerates say is okay.
And that, I believe, is the problem. I can no longer buy RIAA-parent-company DVDs and CDs in good conscience, because I know a portion of the proceeds will be used for suing women and children. And I think the impressions being formed are overwhelmingly negative. Where are the college students protesting, "Save the RIAA!"? Teenagers and college students are starting to believe that buying CDs will only fund lawsuits against the defenseless and poor.
I understand their position regarding copyright infringement. But it is infringement, not theft, and certainly not murder on the high seas. Their strong-arm tactics make the entire industry look bad, and call into question the legitimacy of their cause. When such an entity chooses to pursue lawsuits on such frivolous evidence, one can only conclude that the real goal is not justice but merely the acquisition of additional wealth. First, they steal from the artist through oppressive and one-sided contracts, and now they are trying to extort money from those whom they believe will not have the resources to resist. When was the last time the RIAA filed a lawsuit against a millionaire?
It just makes me sick. And the artists, of all people, stand to lose the most. Instead of buying from big labels, I've begun looking at the smaller, independent artists precisely because of the RIAA tactics.
I wonder if they even considered the fact that dropping CD sales might be related to people unhappy with the fact that they are suing their customers. Unhappy customers tend not to be repeat buyers.
Do you have any idea how large a 1GB dual-port stick of ram would be? It'd probably cost 2 to 3x as much to make
I realize that many people share your opinion, and that it is a valid point. However, I would add that it is people like you who hold back the desktop PC. When everything is about low cost, does it surprise anyone that the standards and hardware produced are optimized not for performance, but for low price?
One thing which impressed me about the designs of IBM mainframes and of the Apple computers (prior to Intel) was that they were absolutely beautiful from an architectural standpoint. When you have clients willing to pay a decent price for your computers, you can build them the way they should be built. (Of course, this is just my opinion).
Instead, we all buy PCs because they're cheaper, and then gripe about the low performance.
Granted, their opinions do not affect their legal standing. I'm not sure why you interpreted my post as stating so. My point was that there's a cognitive dissonance in claiming the wrongfulness of copyright infringement of your own work while simultaneaously infringing the copyright of others. IOW, are these the same students who illegally share music? And if so, why are they being hippocrits?
So I wonder if these same students consider the "cataloging" and sharing of copyrighted music just as infringing?
If it isn't infringing to share music via P2P, then why would it be infringing for school staff to share papers? Especially when in the latter case, they are doing it with the explicit purpose of preventing copying of the students work? Presumably, a student who finds his hard work copied by another student would have a copyright infringement case against the other student, provided that it was indeed copied without permission. After all, high school students would never stoop so low as to allow others to copy their work.
I'm thinking that more than a few of them have downloaded and shared music under the justification of "Maybe I'll buy it later... if I like it." But for some reason, it's only considered copyright infringement when it is their work being copied.
After all, if it isn't your work being copied, it's sharing, right?
And your 5000 transistor processor would have 1/10000th the MIPS of that core 2 duo.
BTW the core 2 duo comes with 4MB of L2 and 128KB of L1. That's 207,618,048 transistors for the SRAM cells
OMG go take a course on processor architectures
I'm stunned, really, I am. I'm not quite sure if you are deliberately trolling or just didn't bother to think before posting.
In the first place, as someone who has presumably had a course on computer architecture, you'd know that even a non-pipelined processor can reasonably execute one half to one instruction per clock cycle. With each core on the Core 2 duo issuing up to 4 instructions per clock cycle, the MIPS of those little cores would be closer to 1/8 to 1/4 that of Core 2 Duo core. The 1/10000th figure you suggest is off by at least 3 orders of magnitude.
Next, there are only 151 million transistors on the die, so I doubt Intel is using 6 transistors per bit for SRAM, as your numbers indicate. Functional SRAM cells can be made with 2 transistors per cell, which is probably closer to the design used by Intel.
And the OMG part just wasn't warranted. Please do explain how having a memory controller on die reduces latency between the processor and memory. Either way, the processor has to go through the memory controller, so I'm not sure how having it on the die, as opposed to on the motherboard, changes inherent latency. (OTOH, if having it on die allows the processor to schedule bus access more appropriately, then it would reduce latency, albeit through maintaining better control of the bus - the reason I thought I made clear in the original post).
And I agree with you that main memory is slow, and perhaps we should be looking into solving that problem, rather than figuring out how to add yet more complexity to the processor.
I'm well aware of the cost issue, but if I had to choose between 256 MB of dual ported RAM and 1 GB of single ported, I'd pick the dual ported RAM every time.
Do you have any idea how large a 1GB dual-port stick of ram would be? It'd probably cost 2 to 3x as much to make.
Versus, say, putting another core on the die? The Core 2 Duo is 151 million transistors. According to Tanenbaum, a rudimentary (non-pipelined, non-cached) microprocessor can be put together for about 5000 transistors. You can do the math: with the transistor budget of the Core 2 Duo, you could have 3000 processor cores.
What it comes down to is that a large part of the die is being used to manage shortcomings elsewhere in the system. For example, the SRAM cache is used to compensate for the fact the DRAM families of main memory require frequent refreshes by the memory controller, and are frequently unavailable during DMA transfers from elsewhere in the system. The latency of DRAM and SDRAM type memories can be in the dozens of clock cycles, even under ideal circumstances (the design trades latency for throughput). And placing the memory controller on the die was done to address precisely the problem I mentioned: something, somewhere else in the system commandeers the bus during a critical time, and the processor ends up stalling, waiting on access to main memory. The problem isn't with the processor; it is doing its job - it's with the design of the rest of the system, in which the CPU is literally, just another peripheral.
But we shouldn't fix a problem in the memory system by adding a kludge to the processor. And that is exactly what this is. Consumers buy the cheapest RAM they can find, without regard for how fast, or how well designed, the rest of the system is.
And yes, I understand the cost issue. But who really uses 1 GB of memory? Maybe 5 percent of users out there. The rest of them just buy the 1 GB box because, (Joe Sixpack voice:) "Hey, it's got more than that other box..." I doubt if most users could even give you a ballpark figure for the memory used by their applications.
In fact, if most users knew how much memory they actually used, we wouldn't be having this discussion. Instead, users would demand faster memory, rather than more of it.
It is interesting to note that Intel has now decided to put the memory controller on the die, after AMD showed the advantages of doing so.
However, I'm a little dismayed that Intel hasn't yet addressed the number one bottleneck for system throughput: the (shared) memory bus itself.
In the 90's, researchers at MIT were putting memory on the same die as the processor. These processors had unrestricted access to its own, internal RAM. There was no waiting on a relatively slow IDE drive or Ethernet card to complete a DMA transaction; no stalls during memory access, etc...
What is really needed is a redesign of the basic PC memory architecture. We really need dual ported RAM, so that a memory transfer to or from a peripheral doesn't take over the memory bus used by the processor. Having an onboard memory controller helps, but it doesn't address the fundamental issue that a 10 ms IDE DMA transfer effectively stalls the CPU for those 10 milliseconds. In this regard, the PC of today is no more efficient than the PC of 20 years ago.
It has long been known that certain racial and ethnic groups have a certain gait, as a matter of simple conformance with their culture. It shouldn't surprise anyone that a child grows up imitating their parents, including their manner of walking.
So, by teaching computers to recognize a person's gait, and single out people based on their gait, we have, in essence, taught computers to give greater scrutiny to individuals of a particular ethnic or racial group.
In other words, we've programmed computers to be racist.
While it is arguable whether this was the intent or not, the fact is the effect is the same. It was bad enough when a person was racist, but at least there was the hope of a defense in exposing the racism of an officer. How do you expose racism in a computer program, when computers are presumably incapable of racial bias or personal feelings?
Score one for racism.
Politicians are busy trying to create people that are more than 15% sheep...
At this stage, it is uncertain who has made better progress, but it's not looking good for the scientists...
Pthreads has been out for a while. It is open source, and runs on Linux, Windows, and Mac(?).
Whether or not you believe concurrency should be an explicit library or a matter of compiler extension is a bit of a religious argument. But pthreads does offer the functionality, and works fairly well.
First of all, I, and many others before me, have been writing multithreaded applications for years in the likes of Linux and UNIX. I have had to maintain multithreaded applications created by others. My collective experience tells me:
It is not trivial.
Let me repeat: It is not a trivial task. Even if you have libraries and an API which abstracts out the ugly stuff, you still have the problem of concurrency, proper locking, deadlocks, etc...
The majority of problems with using multithreaded programming come not from "ugly" parts of the OS/API layer, but from a misunderstanding of the problem. A few problems in computer science - particularly in the physical sciences - do benefit from multithreading. And it is easier to use threads when writing a game than just to execute all of the IO in one big loop (Hello DOS!). But for most applications, using threads is not only unnecessary, but overkill, and introduces the possibility of yet another class of bugs for which the application must be tested. Furthermore, as deadlock and race conditions are often timing related, they are the most difficult type of bug to find and fix. Finding and fixing this class of bugs is still somewhat of a black art in the industry, and is highly dependent on the skill and experience of the programmer.
In short, unless your system/application design cannot do without multithreaded programming, it is best not to use it. Even with a glossy API, you still cannot escape the fact that debugging a multithreaded application is an order of magnitude more difficult than a single threaded one. In any case, you shouldn't be using threads just because you can.
Why not work toward happiness?
If you aren't happy with where you're at, at least figure out where you'd be most happy working. Talk to people - new people. Give yourself a year to figure out where you want to be in 10 or 20 years from now, and work out a plan to get there.
But just dropping employment probably won't be a good idea. It may very well limit your employment options later on. Besides, not having money will depress you anyway. And even if you aren't happy with your current job, it might be alright if you know you're working toward something better. The trick is finding that better thing.
I can tell you the reason why the industry needs a subsidy:
No one wants to pay for DRM.
The market hasn't failed. Rather, the content companies have begun to realize that people don't want to pay more to get less.
I mean, why would I buy a tv with fewer features?
The content companies have begun to realize that they need to provide some kind of short-term incentive to get the customer to give up the rights to which they have become accustomed. Once the first generation grows up without the ability to record tv, they'll think it's normal. And the worst of it is that it isn't the content companies paying the bill, but the American taxpayer!
With DTV, the public domain goes away. DRM isn't there to prevent some content from being rebroadcast; it is there to keep all content away from the net. Even things legally in the public domain.
Call your Senator and tell him to oppose this bill. Tell him you don't want Congress wasting money...
There are a few people who are focused on "how do I make this better?
That's because a computer programmer's time is a lot more expensive than the machine on which his code runs. In some of my computer science classes, I was told flatly to ignore efficiency because it was cheaper for business to buy a faster computer than it was for me to optimize the algorithm in assembly.
This is just the classic conflict between engineering and business - the business wants a shed, and the engineer wants to build a cathedral.
The problem with his, analysis, and yours, is the mistaken assumption that the majority of the opinions expressed constitute the majority of opinion.
There are a lot folks like me who have found Linux useful, use it for what it is good at, and refrain from getting involved in the religious wars in the mailing lists.
If you can't rationally discuss something in the "Linux community", you're in the wrong community. Consider security, for example. I happen to know that both the NSA and RedHat chose to offer alternative security models in their Linux versions. Yet I don't recall any religious arguments over the relative merits of the respective models. Instead, people seemed to comment on the pros and cons of each model when the issue came up - or maybe I just didn't read slashdot one day...
And I think the best part about Linux, that everyone seems to miss, is the honesty.
When someone tells me that something works in Linux, I can trust that statement to be true, because they usually speak from a position of experience and impartiality. Contrast this with the proprietary model of software, where someone's livelihood is at stake if they don't sell enough licenses. I simply can't trust any advice given about Windows compatibility because there is always exists a monetary incentive for Microsoft to lie about the quality of their software.
I can much more easily accept software that doesn't work quite right if I know what I'm getting from the outset. OTOH, it is particularly galling for a large company, like Microsoft, to claim their OS works, and experience otherwise after having spent a considerable amount of time and money in useless frustration. At least there wasn't any deception with Linux, and I don't bang my head against a wall trying to force it to do something for which it wasn't designed.
I can show the moderators what a sense of humor is! Come on, folks - you take yourselves too seriously...
See how you can spend an extra 130 british pounds, and still have no working webcam!
I can show you how you can spend no money, install Linux, and not have a working webcam, WiFi, and be stuck at 1024 x 768 because X doesn't recognize your video card!
And for a bonus, I can show you how to spend half the night recompiling SCSI drivers into your kernel so that you can actually use that old SCSI drive which has been collecting dust.
I understand that the search engines are private entities, and can do as they please regarding what they choose to advertise. However, I'm not entirely comfortable with private ownership of the medium used for public discourse. There are opinions which are never heard in America simply because the media is corporate owned and simply refuses to publish said opinions.
I could care less what private entities do with their own networks, but our public discourse should not be limited to a privately-owned medium.
When religion is wrong, it's not possible to discover and fix it.
Tell that to Jesus. He fixed the misconceptions in the religion of his day.
It's not that "fixing" religion is impossible, it's just that some religions - like Christianity - have had a long time to hone their doctrine. Finding a mistake in Christian doctrine would mean that you have more insight than the 2,000 years of scholars who have come before you. Possible, but not likely.
And, from time to time, some religions with errors do arise. Remember Jim Jones? Well, when things like this happen, they usually correct themselves within a few generations.