The best way to weed out candidates for a program is to admit as many as possible and let some (maybe even the majority) fail and drop out. Any other method, especially methods like selecting candidates on the basis of high school GPA is basically the same - except you randomly eliminate some of the best candidates beforehand.
If a school has some constraint where they expect the number of admitted engineering students to be the same as the number of engineering students that they graduate then they're simply being unrealistic. No matter how much you pre-filter your candidates, you're still going to loose a chunk to multivariable calculus or the first programming class that uses pointers.
The FSF has no ability to alter the text of existing versions of the GPL. If you want to use only the GPLv1 and no other version, you can say that and the FSF has no way to magically change the license text that you distribute with your code. If you want to say "Version X or later", then you're leaving an opening to the FSF to change things (which might even be a good idea), but including that text or not is your choice.
The creation of a new copyleft license at this point in time is simply not-invented-here syndrome on the part of some corporate lawyers, and the result is license compatibility issues. Any full copyleft license is innately incompatible with any other, and that's caused enough hassle over the last 18 years that there's no reason to do it again now.
Non-copyleft permissive licenses aren't really a problem - the only annoyance is having to read yet another license that's basically equivalent to the X11 license and be sure that that's what it really says.
That video doesn't so much present the details as introduce the issue (and hype it as a serious problem). It leaves a number of questions unanswered - specifically: Does a debt currency system actually result in a non-trivial systematic economic unbalance in favor of bankers, or do factors like inflation, publicly held bank stock, and bad debt tend to balance that out in practice? What are the political ramifications of a centrally manged banking system with the amount of government oversight that our system has? What are the legitimate other choices? What are the practical tradeoffs between them? What evaluation criteria should be used in comparing this issue to other political issues? What are the results of such an evaluation? If some other system would be better, how can we transition to that system?
If you borrow $1000 from the bank, then the bank basically ends up $10,000 to spend.
It's a little more complicated than that. If it were that simple, I'd be opening a bank tomorrow to exploit that loophole.
As it is, the biggest problem I see with the banking system is simply that I don't have a clear understanding of how it works and what the resulting consequences are. Money is pretty important - everyone should be able to have a basic understanding of it, it shouldn't be so complicated that someone who's interested and has a math background isn't clear what's going on.
I've found H.264 is a tad bit fuzzier on my LCD TV during high-action scenes than Xvid (my reference is with the show Firefly) but given the difference in filesize, I'll never be one to complain!
Saying that the quality is different at different file sizes is not interesting. What is the quality like at *the same* file sizes? What are the file sizes like at the same quality?
The main reason multithread programming is so hard is that it is non-deterministic.
Sure, if you're futzing with locks in a shared memory model that's true. If you implement any of the reasonably simple abstraction techniques that have been well understood for something like 20 years now (say... CSP for example) then your concurrent program ends up being deterministic and reasonably easy to follow.
Basically the "multithreaded programming is hard" argument is no more interesting than the "programming is hard because you have to manually free allocated memory all the time and if you don't your program breaks horribly" argument. You fix that by implementing some sort of garbage collector. The solutions to concurrent programming woes are a lot like the solutions to memory allocation woes - you end up having to write programs differently and there are measurable performance losses compared to the theoretical optimal manual solution, but in the end you have to chose between that and memory leaks / deadlocks.
Indeed; for personal computers there is hardly ever need for more than, say, two cores.
Right now. With the software available to you. Mostly because programmers wouldn't get very far trying to make you need things that aren't available to you.
On the other hand, programmers tend to find a way to make use of whatever hardware is available to them. It's sort of funny watching some of them bitch and moan about how hard multi-core programming is, but once 8+ core systems are common you can be sure that your compute intensive apps won't even run on less than 4.
Programming for multi-core systems isn't so much hard as it is different, and people *hate* having to learn new things.
Anyway, for me at least an OS is pretty ephemeral.
That's a very strange way to look at things, and not a way that would work very well if you actually had significant local software you were running on a machine. In general, intentionally selecting a software platform (including OS) and then selecting hardware that allows that software to function optimally is a more productive method than selecting hardware based on some arbitrary other criteria (like what?!?) and then letting your software selection be determined as an arbitrary result.
Sorry to say, but you strike me as someone who's just parroting something they read 5 years ago, and hasn't adapted with the times.
Not at all. I'm just trying to reintroduce some common sense to the whole hardware compatibility discussion. The *reason* you are buying hardware is to run software; given that, it would make sense to select hardware with your software requirements in mind. Anything else is simply poor planning.
Of course, many people would respond by saying I should have bought a Palm device instead, since those Linux tools are more mature. Maybe they're right, but I like my Pocket PC too much, in terms of hardware and certain software that it runs, to have chosen something else, or to change now.
Just as long as you realize that this is your own choice to be locked in to a single vendor's product line rather than some flaw in Linux. You could have built a similar setup based around Linux instead of Windows, but you didn't.
Oh, and I have a Pocket PC, which does just work in Windows, and barely works at all, at least not without a lot of pain, in Linux.
Could that be because the hardware was explicitly designed to run Windows Mobile perhaps?
If you want a Linux machine, buy a Linux machine. Buying a Windows machine and then complaining that it doesn't run Linux perfectly is absurd - I bet an old Mac won't run PS3 games perfectly either (even though they both have very similar hardware architectures).
Whenever the next Ubuntu version comes out I'll try it out on the workstation and see if sleep mode actually works.
I really don't understand this at all. Why would you be willing to base your choice of operating system on random chance? Did you happen to pick a piece of hardware that is supported? Did it happen to magically become supported since then? I sure as hell wouldn't pick an OS based on that.
There is hardware that is fully supported by Ubuntu. It provides the same performance and functionality as unsupported hardware at the same price - this isn't like a PPC mac where the hardware is radically different. If you buy supported hardware, your software will work. If you buy arbitrary hardware and then later hope that "it will work with the next release of Ubuntu" you're using the most absurd form of OS selection (and hardware selection) that I've ever heard of.
People would never think this way for any OS other than Linux. People don't buy random processors and hope it works on Windows (ooh, I got an UltraSparc this time - crap, looks like Vista doesn't work; I guess I'll just keep running Solaris). People don't buy random hardware to run AIX, OS X, or Solaris, or Symbian - they buy hardware to run the software they want to use. Why would you treat your desktop system any differently?
Where the line has gotten fuzzy--and in large part because of Terrorism tactics--is when you have what appears to be a band of civilians hell bent on blowing up a shopping mall.
What "terrorism tactics"?
There have been so few significant terrorist attacks in the United States in recent history that you could count them on one hand. Of those attacks, every last one of them was the direct result of the United States government deciding to bomb the shit out of (or assassinate the democratic leader of, or impose unpopular laws on) some political hornets nest or other.
The fact that those attacks have been used as an excuse to continue the exact policies that caused them is absurd.
It's just not practical - there's no reason to use much more complex and dangerous designs to get 10% of extra efficiency.
You're utterly missing my point. Just because some specific device design today is a bad deal compared to some other specific device design today absolutely does not imply that a given technique is innately impractical.
If you want to be accurate, say this: There exists a high temperature reactor design that seems to be a bad deal compared to more traditional reactor designs.
We can use some insane things like high temperature (thousands degrees) reactors with gas cooling to get another 10%-15% of efficiency, but it is just not practical.
Never, ever, ever dismiss that sort of technique (i.e. engineering problems) as "not practical". It may not currently be a good business decision under the evaluation rules being used by the management at a particular company, but that doesn't mean that a significant performance improvement that requires new techniques isn't a viable (and essential to evaluate) solution to a given problem.
That's one way things could happen, but it seems like a wildly unlikely scenario to me. Some people have been blocking web ads for as long as they have existed (that's more than 10 years now), and nothing horrible has happened to the web yet.
A much more likely scenario is this: Most people will remain completely unaware that web ads are optional, and they'll keep clicking on them and buying things. Those of us "in the know" will keep blocking annoying ads (even as we occasionally make ad driven purchases anyway; I recently bought a wicked lasers laser pointer due to a web ad I saw). I expect it to go on this way for a while, until the web gradually changes into whatever it will be next - and neither you or I has any idea what that will be.
some stuff cant be left to 'market economy' in stupidly overdarwinian approaches.
True enough. I don't think that any given ad-supported website is one of those things. And no, ad supported websites do not make up the entire internet. Further, declaring the threat to ad revenues from ad-blockers to be an economic threat on par with the great depression is absurd.
In any case, advertising revenues on the web aren't going anywhere. Advertisers will have to balance the benefits of obnoxious ads with the possibility of getting blocked by end users - and that's a good thing.
This is very simple: Just because a business model happens to work today does not mean that anyone is entitled to eternal profits from that business model. As long as ad revenues continue to work, sites will continue to fund themselves that way. If ads get too annoying and everyone blocks them then website owners will have to find other revenue sources. Sorry. That's a market economy for you.
But even the advertising companies aren't legitimately worried about that. Instead, the most profitable such company has simply made their ads less intrusive. I run adblock plus, and - guess what... the text ads on google.com aren't blocked because even the adblock guys aren't bothered by them.
User interfaces confuse people about the difference between code and data, and the spread of trojans is the unfortunate result.
In reality, the difference between code and data ends up being even more confusing than modern user interfaces would imply. Software developers seem to really enjoy the idea of embedding code in data or the other way around - so a good percentage of files are both.
This presents something of a problem. Not because there aren't solutions - I can think of four or five good ones off hand - but because there's so much inertia. People want better security, as long as it doesn't break their Microsoft Word documents and self-extracting ZIP files that automatically execute device driver installers.
Unix *is* open source. Between *BSD and Solaris, pretty much all the Unix code you might want is available. Seriously - what useful code is in some version of Unix that Novel may hold copyrights for that isn't in *BSD or Solaris?
Not unless you clicked through a "these packages aren't signed" warning. The package signing system is specifically designed to handle compromised repositories.
GPL has *nothing* to say about in-memory modification of a preloaded binary. That is not even covered by copyright law so it couldn't.
That's an interesting claim. I'm not a lawyer, but the details of exactly what copyright covers are actually pretty vague. I wouldn't be so quick to say that copyright law doesn't cover modifications of copyrighted works.
If schools where competitive businesses rather than a socialized service, you'd be absolutely right. Unfortunately, if a school is providing poor service the students can't go anywhere else. They are legally obligated to go to school, and in many cases choosing another school is not feasible.
Everything is a tradeoff for this sort of thing. A VIA C7 w/ Unichrome Pro graphics probably gives better performance than a PPC board at the same price point. If that matters (and it may) then this board meets the design requirements better for the product. If not then the answer is different.
Slashdot Mirror
The best way to weed out candidates for a program is to admit as many as possible and let some (maybe even the majority) fail and drop out. Any other method, especially methods like selecting candidates on the basis of high school GPA is basically the same - except you randomly eliminate some of the best candidates beforehand.
If a school has some constraint where they expect the number of admitted engineering students to be the same as the number of engineering students that they graduate then they're simply being unrealistic. No matter how much you pre-filter your candidates, you're still going to loose a chunk to multivariable calculus or the first programming class that uses pointers.
Umm... no.
The FSF has no ability to alter the text of existing versions of the GPL. If you want to use only the GPLv1 and no other version, you can say that and the FSF has no way to magically change the license text that you distribute with your code. If you want to say "Version X or later", then you're leaving an opening to the FSF to change things (which might even be a good idea), but including that text or not is your choice.
The creation of a new copyleft license at this point in time is simply not-invented-here syndrome on the part of some corporate lawyers, and the result is license compatibility issues. Any full copyleft license is innately incompatible with any other, and that's caused enough hassle over the last 18 years that there's no reason to do it again now.
Non-copyleft permissive licenses aren't really a problem - the only annoyance is having to read yet another license that's basically equivalent to the X11 license and be sure that that's what it really says.
That video doesn't so much present the details as introduce the issue (and hype it as a serious problem). It leaves a number of questions unanswered - specifically: Does a debt currency system actually result in a non-trivial systematic economic unbalance in favor of bankers, or do factors like inflation, publicly held bank stock, and bad debt tend to balance that out in practice? What are the political ramifications of a centrally manged banking system with the amount of government oversight that our system has? What are the legitimate other choices? What are the practical tradeoffs between them? What evaluation criteria should be used in comparing this issue to other political issues? What are the results of such an evaluation? If some other system would be better, how can we transition to that system?
It's a little more complicated than that. If it were that simple, I'd be opening a bank tomorrow to exploit that loophole.
As it is, the biggest problem I see with the banking system is simply that I don't have a clear understanding of how it works and what the resulting consequences are. Money is pretty important - everyone should be able to have a basic understanding of it, it shouldn't be so complicated that someone who's interested and has a math background isn't clear what's going on.
Saying that the quality is different at different file sizes is not interesting. What is the quality like at *the same* file sizes? What are the file sizes like at the same quality?
Sure, if you're futzing with locks in a shared memory model that's true. If you implement any of the reasonably simple abstraction techniques that have been well understood for something like 20 years now (say... CSP for example) then your concurrent program ends up being deterministic and reasonably easy to follow.
Basically the "multithreaded programming is hard" argument is no more interesting than the "programming is hard because you have to manually free allocated memory all the time and if you don't your program breaks horribly" argument. You fix that by implementing some sort of garbage collector. The solutions to concurrent programming woes are a lot like the solutions to memory allocation woes - you end up having to write programs differently and there are measurable performance losses compared to the theoretical optimal manual solution, but in the end you have to chose between that and memory leaks / deadlocks.
Right now. With the software available to you. Mostly because programmers wouldn't get very far trying to make you need things that aren't available to you.
On the other hand, programmers tend to find a way to make use of whatever hardware is available to them. It's sort of funny watching some of them bitch and moan about how hard multi-core programming is, but once 8+ core systems are common you can be sure that your compute intensive apps won't even run on less than 4.
Programming for multi-core systems isn't so much hard as it is different, and people *hate* having to learn new things.
That's a very strange way to look at things, and not a way that would work very well if you actually had significant local software you were running on a machine. In general, intentionally selecting a software platform (including OS) and then selecting hardware that allows that software to function optimally is a more productive method than selecting hardware based on some arbitrary other criteria (like what?!?) and then letting your software selection be determined as an arbitrary result.
Not at all. I'm just trying to reintroduce some common sense to the whole hardware compatibility discussion. The *reason* you are buying hardware is to run software; given that, it would make sense to select hardware with your software requirements in mind. Anything else is simply poor planning.
Just as long as you realize that this is your own choice to be locked in to a single vendor's product line rather than some flaw in Linux. You could have built a similar setup based around Linux instead of Windows, but you didn't.
Could that be because the hardware was explicitly designed to run Windows Mobile perhaps?
If you want a Linux machine, buy a Linux machine. Buying a Windows machine and then complaining that it doesn't run Linux perfectly is absurd - I bet an old Mac won't run PS3 games perfectly either (even though they both have very similar hardware architectures).
I really don't understand this at all. Why would you be willing to base your choice of operating system on random chance? Did you happen to pick a piece of hardware that is supported? Did it happen to magically become supported since then? I sure as hell wouldn't pick an OS based on that.
There is hardware that is fully supported by Ubuntu. It provides the same performance and functionality as unsupported hardware at the same price - this isn't like a PPC mac where the hardware is radically different. If you buy supported hardware, your software will work. If you buy arbitrary hardware and then later hope that "it will work with the next release of Ubuntu" you're using the most absurd form of OS selection (and hardware selection) that I've ever heard of.
People would never think this way for any OS other than Linux. People don't buy random processors and hope it works on Windows (ooh, I got an UltraSparc this time - crap, looks like Vista doesn't work; I guess I'll just keep running Solaris). People don't buy random hardware to run AIX, OS X, or Solaris, or Symbian - they buy hardware to run the software they want to use. Why would you treat your desktop system any differently?
What "terrorism tactics"?
There have been so few significant terrorist attacks in the United States in recent history that you could count them on one hand. Of those attacks, every last one of them was the direct result of the United States government deciding to bomb the shit out of (or assassinate the democratic leader of, or impose unpopular laws on) some political hornets nest or other.
The fact that those attacks have been used as an excuse to continue the exact policies that caused them is absurd.
You're utterly missing my point. Just because some specific device design today is a bad deal compared to some other specific device design today absolutely does not imply that a given technique is innately impractical.
If you want to be accurate, say this: There exists a high temperature reactor design that seems to be a bad deal compared to more traditional reactor designs.
Never, ever, ever dismiss that sort of technique (i.e. engineering problems) as "not practical". It may not currently be a good business decision under the evaluation rules being used by the management at a particular company, but that doesn't mean that a significant performance improvement that requires new techniques isn't a viable (and essential to evaluate) solution to a given problem.
I think you've found the basic bug. If we just fix that one, a lot of these other ones will quietly disappear.
That's great, until the ISP decides that they can block any UDP traffic that isn't DNS to their servers.
A much better idea would be to simply make the connections look as much like HTTP over SSL as possible. They can't block that.
That's one way things could happen, but it seems like a wildly unlikely scenario to me. Some people have been blocking web ads for as long as they have existed (that's more than 10 years now), and nothing horrible has happened to the web yet.
A much more likely scenario is this: Most people will remain completely unaware that web ads are optional, and they'll keep clicking on them and buying things. Those of us "in the know" will keep blocking annoying ads (even as we occasionally make ad driven purchases anyway; I recently bought a wicked lasers laser pointer due to a web ad I saw). I expect it to go on this way for a while, until the web gradually changes into whatever it will be next - and neither you or I has any idea what that will be.
True enough. I don't think that any given ad-supported website is one of those things. And no, ad supported websites do not make up the entire internet. Further, declaring the threat to ad revenues from ad-blockers to be an economic threat on par with the great depression is absurd.
In any case, advertising revenues on the web aren't going anywhere. Advertisers will have to balance the benefits of obnoxious ads with the possibility of getting blocked by end users - and that's a good thing.
This is very simple: Just because a business model happens to work today does not mean that anyone is entitled to eternal profits from that business model. As long as ad revenues continue to work, sites will continue to fund themselves that way. If ads get too annoying and everyone blocks them then website owners will have to find other revenue sources. Sorry. That's a market economy for you.
But even the advertising companies aren't legitimately worried about that. Instead, the most profitable such company has simply made their ads less intrusive. I run adblock plus, and - guess what... the text ads on google.com aren't blocked because even the adblock guys aren't bothered by them.
In reality, the difference between code and data ends up being even more confusing than modern user interfaces would imply. Software developers seem to really enjoy the idea of embedding code in data or the other way around - so a good percentage of files are both.
This presents something of a problem. Not because there aren't solutions - I can think of four or five good ones off hand - but because there's so much inertia. People want better security, as long as it doesn't break their Microsoft Word documents and self-extracting ZIP files that automatically execute device driver installers.
Unix *is* open source. Between *BSD and Solaris, pretty much all the Unix code you might want is available. Seriously - what useful code is in some version of Unix that Novel may hold copyrights for that isn't in *BSD or Solaris?
Not unless you clicked through a "these packages aren't signed" warning. The package signing system is specifically designed to handle compromised repositories.
That's an interesting claim. I'm not a lawyer, but the details of exactly what copyright covers are actually pretty vague. I wouldn't be so quick to say that copyright law doesn't cover modifications of copyrighted works.
If schools where competitive businesses rather than a socialized service, you'd be absolutely right. Unfortunately, if a school is providing poor service the students can't go anywhere else. They are legally obligated to go to school, and in many cases choosing another school is not feasible.
Everything is a tradeoff for this sort of thing. A VIA C7 w/ Unichrome Pro graphics probably gives better performance than a PPC board at the same price point. If that matters (and it may) then this board meets the design requirements better for the product. If not then the answer is different.