The record industry should hire a few economists. This is a great idea, but they've got the pricing completely backwards. The more popular songs shold get cheaper and the less popular more expensive. Why? That's easy.
The stock market works the way it does because supply is fixed and demand is the only variable. With digital music, the supply is infinite, and the demand is variable. Theoretically, that should mean that the songs could be free, except that the creation of the media has fixed up-front costs. That means that after a fixed amount of revenue is generated by a song, all additional revenue is going to be 100% profit. In order to make the maximum amount of money off any particular song, you want to increase it's appeal as much as you can through price lowering, while at the same time making sure you charge enough to recoup your costs before you break even, and as much as you can without pushing away customers after you break even. If there is a lot of demand for a song, you're going to make a profit on it, but you could potentially make even more money by lowering the price, because the drop in price could attract more than enough customers to make up for the loss in revenue. For songs without a wide acceptance, it doesn't work that way. You probably don't have many people out there who like the song but have price holding them back from a purchase, and the people that are buying the song are probably the ones that really like it and would be willing to pay a bit extra to have access to music that would otherwise be unprofitable to publish.
The only way charging more for popular songs is a good idea is if your goal is to punish your customers for being mainstream music listners, or if you have a complete lack of understanding of supply and demand. If the goal is to actually make money, they've got this plan completely backwards.
Yeah, but there's no vendor lock-in with the hardware. Lots of people make it.
If Microsoft made free viewers for other platforms you'd have a point, but if you have to give money to Microsoft to use their "free" viewer, that's hardly free.
People carry butane lighters and hip flasks all the time. Hell, they sell more concentrated alcohol than this on most long haul flights already. At least pouring it into your device instead of your mouth won't cause intoxification.
You do realize that carring a bit of diluted methanol is no less dangerous than carrying a nip of stron brandy, right? (Unless you drink it. Don't drink methanol. It'll be fun for a bit, but then you'll be blind, and nobody likes being blind.)
People like you need to work on changing your default position for things you don't understand from fear to curiosity. If you went and did the tinyist amount of research instead of just being afraid we'd be one step closer to a better world.
Most importantly though, they were made from chalk: "It washes right off, so it will be removed the next time it rains." Total non-issue.
Look for the next story about it. They didn't wash off. You had to scrub like hell, and if you didn't they stayed untill they wore off. They were intended to come right off, but they blew it.
17 seconds? Maybe on your little toy that you call a computer. A *real* computer that you use to run some highly available, resource intensive process on needs longer than that just to initialize it's ECC memory... Not to mention the time it takes to scan the SAN, mount drives, start up services... All during that time you're hundreds of users are waiting for your system to come back, and you may be losing thousands of transactions per second of downtime.
If it's only taking you 17 seconds, it's clear that you only have a measly few gig, and that it's crappy non-ECC stuff. (Plus you probably have only gotten to a "working desktop". Your machine is likely still starting up stuff for the next few minutes while you wonder whiy your computer is so slow.)
Windows *needs* to be able to update almost everything without a reboot if it ever wants to grow up and be something more than a toy in the data center.
Yup. Just like they ditched showcases, and the one ad on the main menu.
Seems to me this will just be a menu for ads instead of a link to a single ad... And people do watch them. I even find myself clicking on them now and then when they're something like a movie preview.
You know, you can get Zyxel routers with equally good hardware that also run linux in the $18-25 price range... You just have to actually know what you're doing to run linux on them (Well, your own linux... They run linux out of the box) instead of installing a script-kiddy friendly installer.
Plus you don't have to deal with the community of Linksys fanboys... which rivals the communities of game console fanboys.
The Exchange database is totally intolerant of any unplanned outages, and is laid out in such a way that failures are never compartmentalized. That's one of the big benefits to Notes (not a big enough benefit to make up for it's negatives though), that every user has their own database so whole-system downtime is practially non-existant.
Any light duty exchange server in an environment that doesn't justify eliminating all single points of failure (communication, power, etc) will experience corruption. That can be from pulling a cable by accident, having a SAN switch fail, having an extended power outage, losing a UPS, etc... And then if you have any signifigant number of users you're talking many hours of downtime to rebuild.
That's terrible. You shouldn't need a full time staff to run your groupware server.
Maybe I'll give you that it's the best one that's generally available, but it's total shit. It's unstable, unreliable, and counterintuitive in many cases. It also does a poor job of third party interoperability, and has a legendary record of insecurity.
There may not be many better choices out there, but you're giving Exchange *way* too much credit.
You have to have a reasonable cost/benefit analysis. The Vonage 911 system works well in the vast majority of the cases. With millions of customers benefiting from this service, and what, one? Maybe two deaths that could even be remotely associated with having Vonage instead of POTS... Is it really worth eliminating this useful technology that could benefit hundreds of millions of people on the off chance that it may save a couple lives?
Our society needs some sanity injected into it. Life is valuable, but we can't continue as a modern society if the standard becomes that we have to protect every life at any cost. There are some cases where you have to decide what is good enough and start putting your resources elsewhere. There's an exponential curve of investment level that correlates to a linear increase in service quality here. Vonage should be allowed to pick a sane point on the curve.
You think the package deals are price gouging? Wait until you see the a la carte pricing... How many channels do you think you'll be able to subscribe to before you've exceeded the price of the package that contains them? 3? 4? It's going to be something rediculous like that.
First of all, Insightful my ass. The moderators of your comment should be shot.
Second, your comment is interesting, considering two things:
First, that this is an article about how Diebold can't profide North Carolina with source escrow because it can't provide the Windows code. (You did read the article, right? Or perhaps you'd like to borrow some clue?) Regardless, the shared source license it part of the marketing bullshit that Microsoft uses to create their special case, and you've completely bought into it. Source escrow typically guarantees your right to continue to redistribute and advance development a third party product should the producer cease to exist or to terminate support for a particular product. Find that guarantee in the Microsoft Shared Source license. If Microsoft terminated production of CE, people who make products based on it would be screwed. The shared source license is not even close to equivalent to source escrow.
Second that it's naive to think that any developer list is complete, or that there is even a remote chance of proving it either way. Require it all you want, but in the end you're going to end up with a worthless list of names with no way to know if it's complete, or correct.
Usually when you develop an embedded system, you demand code escrow from your suppliers. Microsoft is a special case though, because when they enter the conversation everybody seems to become stupid. If they had gone with any other vendor (I'm not just talking Linux here... They could have used VxWorks, QNX, BSD, one of the various DOSes...) they would have had code escrow. I bet they do for every other third party bit of software on their machine.
The list of developer names is pretty unreasonable, but code escrow is something that happens all the time, and only Microsoft manages to get out of it.
The fact is it simply can't happen in a language like Java.
There are a lot of things that can't happen in a language like Java. Some of them are considered features in some circumstances. At least you have the option to manage your memory or have it managed for you in C, instead of being stuck with one way or the other.
There are so many poor practices involved in that situation you just described that I don't even know where to begin.
There are dozens of simple rules you can follow when you write C code, any one of which would have prevented that problem. Either way, having your memory managed for you doesn't imply that you don't have access to the raw data anyway. Protection is only enforced in some languages.
so C++ can have the same types of so-called "hard to debug" leaks you blame on garbage collection
I said they become more difficult to debug because of garbage collection. They're certainly not caused by the garbage collection. They're caused (usually) by poor programming.
Garbage collection is a tool. It makes your job as a programmer easier, but it does not free you from the need to understand things like scope. Just because you don't have to worry about the mechanics of managing your memory, you still need to understand how it works, or you are going to write crappy code. Code that leaks.
But on top of that, if you have a true memory leak, C++ doesn't even tell you what's stored in all that leaked memory.
That's because you don't need the programming language to tell you that. There are plenty of ways you can make it easy to tell what's in your leaked memory. And then when you're done debugging, you can turn them off and reduce your memory footprint.
Actually, it seems to me that if you want reliability, maintainability, and perhaps most important, debugability, you want to manage your memory yourself.
When debugging a program with a leak (Yes, garbage collected programs have leaks too, they're just nastier, because they don't look like bugs because a reference is persisting somewhere.) if memory is program managed, finding the leak is a deterministic process. You're guaranteed success in a well-defined, and finite amount of time (The amount of time it takes to reproduce the leak plus the amount of time it takes to build the application, plus the amount of time it takes to get a basic understanding of the application structure). When you're debugging a leak in a memory managed environment, you're limited to the tools provided, which are pretty much universally not good at finding the worst kinds of leaks, because to the tools your leak looks like perfectly valid utilization. You're really doomed if your leak is in a third party library. You may never track it down, or even find the culprit. (Any Jakarta users out there?)
Sure, languages without garbage collection may be more prone to leaks, but I'd rather have more leaks that I can fix, than fewer leaks that I can't...
Not that garbage collection doesn't have it's place. If you're application is complex enough though, allocating your own memory is the least of your worries.
If your program uses lots of RAM and you need every last drop either find an expert at RAM management to get every last bit
I don't care what kind of programmer you are, you should have a good foundation in certain things. Data structures, obviously. Discrete logic. Algorithmic complexity, including the ability to read and understand Big O notation (wspecially true of java programmers)... And a complete knowledge of basic memory management. Even if you're going to use garbage collection. Without these things, you're doomed to write poor programs. If you don't understand how a utility library works, you shouldn't be using it.
Which returns us to the question of what the actual benefit of this device is over a traditional electric on-demand water heater. The only one I could see is size.
Why bother timing this stuff? It's pointless. You'll have a bunch of useless data. It's not like they only have 8k of memory and don't have room to store the procedures for the hotkeys or something. They've got modern machines. Support both. That way the users can do it how they want to do it instead of how the arrogant UI designer that knows better thinks they should do it.
Like $40? So, you don't actually know. How much else are you guessing here?
It's an educated guess. There was an article (listed on slashdot not to long ago even) about how it costs intel about $40 to make a processor, regardless of it's speed, etc...
Also, intel doesn't just keep it's old fabs around. It upgrades them. I'd bet money that the P3s in the recent Xboxes use the latest fab technology, for two reasons. First, because the old fabs have been upgraded to use the newer processes. They may not use the absolute cutting edge, but they probably use one generation back, which is a few generatinos ahead of what they were using when the Xbox was released. Second because the cost of a silicon wafer and the processing of said wafer is fixed, and it's cheaper for intel to produce the processors on the smaller process to get more chips out of a single wafer.
Finally, there's no guessing about the fact that the prices went up, the only guessing is by how much, since Microsoft isn't at liberty to disclose that information. You can look it up rather easily because there was a legal battle between Microsoft and Nvidia about it from when Nvidia jacked up the prices. It eventually led to Microsoft using ATI for the 360 instead. So I'm guessing as to exactly what the prices are, but if can be confirmed as fact that they're higher than they were on day one.
It's hard for me to see why you're getting so emotional about this. Do you really care how much it cost Microsoft to build a console that they now no longer make?
The chips that you see for $10 are ones that were left in the retail channel after production stopped. They're discounted so deeply because the retailers are stuck with them and need to cut their losses.
And no, price for chips do not increase over time. They continue to decrease. 8080, 8086, and 486 chips are still being made, sold, and purchased to this day. Only in retail do they cost more than a few dollars.
Those chips are still in wide use, and are incredibly simple. The P3 700 isn't really that simple though. It uses the same silicon and fabs that are used to produce Intel's higest profit chips. There isn't a big market for the chip except from microsoft, so unlike the 8080, etc, the core hasn't been licensed and produced elsewhere. The only place you can get it is from Intel. They pay wholesale, and the get the maximum discount (which they got right from the begining, BTW), but since then the fees for continued production of an obsolete product have been applied. The price *isn't* $10, because there isn't some unfortunate retailer in the middle to eat a loss. They have to pay more like $40, which is not much more than it costs intel to actually manufacture the chip. Remember, not only is intel making an old part for Microsoft, but they're doing it with the same materials and equipment that they could be churning out Pentium-Ms on at a high margin. (Incidentally, that probably makes the P3s in recent Xboxes incredibly overclockable.)
They had the same problem with the graphics chips. This is probably the biggest reason the 360 is coming out *now*. Other console manufacturers get to the point where they make a profit on the hardware (if they don't right at the beginning), but the fact that microsoft has to pay a premium for these obsolete chips means they still lose money on the Xbox hardware. You can tell they learned their lesson from the design of the 360. Now they licened the dies, so they can source to a different manufacturer if their current suppliers decide to jack up the prices.
Slashdot Mirror
The record industry should hire a few economists. This is a great idea, but they've got the pricing completely backwards. The more popular songs shold get cheaper and the less popular more expensive. Why? That's easy.
The stock market works the way it does because supply is fixed and demand is the only variable. With digital music, the supply is infinite, and the demand is variable. Theoretically, that should mean that the songs could be free, except that the creation of the media has fixed up-front costs. That means that after a fixed amount of revenue is generated by a song, all additional revenue is going to be 100% profit. In order to make the maximum amount of money off any particular song, you want to increase it's appeal as much as you can through price lowering, while at the same time making sure you charge enough to recoup your costs before you break even, and as much as you can without pushing away customers after you break even. If there is a lot of demand for a song, you're going to make a profit on it, but you could potentially make even more money by lowering the price, because the drop in price could attract more than enough customers to make up for the loss in revenue. For songs without a wide acceptance, it doesn't work that way. You probably don't have many people out there who like the song but have price holding them back from a purchase, and the people that are buying the song are probably the ones that really like it and would be willing to pay a bit extra to have access to music that would otherwise be unprofitable to publish.
The only way charging more for popular songs is a good idea is if your goal is to punish your customers for being mainstream music listners, or if you have a complete lack of understanding of supply and demand. If the goal is to actually make money, they've got this plan completely backwards.
Yeah, but there's no vendor lock-in with the hardware. Lots of people make it.
If Microsoft made free viewers for other platforms you'd have a point, but if you have to give money to Microsoft to use their "free" viewer, that's hardly free.
Readers are still free?
Where?
It doesn't count as free if you have to buy Windows to run it.
If they spent 200m before, and spend 180m now, it should be pretty easy to tell...
People carry butane lighters and hip flasks all the time. Hell, they sell more concentrated alcohol than this on most long haul flights already. At least pouring it into your device instead of your mouth won't cause intoxification.
You do realize that carring a bit of diluted methanol is no less dangerous than carrying a nip of stron brandy, right? (Unless you drink it. Don't drink methanol. It'll be fun for a bit, but then you'll be blind, and nobody likes being blind.)
People like you need to work on changing your default position for things you don't understand from fear to curiosity. If you went and did the tinyist amount of research instead of just being afraid we'd be one step closer to a better world.
Most importantly though, they were made from chalk: "It washes right off, so it will be removed the next time it rains." Total non-issue.
Look for the next story about it. They didn't wash off. You had to scrub like hell, and if you didn't they stayed untill they wore off. They were intended to come right off, but they blew it.
17 seconds? Maybe on your little toy that you call a computer. A *real* computer that you use to run some highly available, resource intensive process on needs longer than that just to initialize it's ECC memory... Not to mention the time it takes to scan the SAN, mount drives, start up services... All during that time you're hundreds of users are waiting for your system to come back, and you may be losing thousands of transactions per second of downtime.
If it's only taking you 17 seconds, it's clear that you only have a measly few gig, and that it's crappy non-ECC stuff. (Plus you probably have only gotten to a "working desktop". Your machine is likely still starting up stuff for the next few minutes while you wonder whiy your computer is so slow.)
Windows *needs* to be able to update almost everything without a reboot if it ever wants to grow up and be something more than a toy in the data center.
You mean like the BSA?
Yup. Just like they ditched showcases, and the one ad on the main menu.
Seems to me this will just be a menu for ads instead of a link to a single ad... And people do watch them. I even find myself clicking on them now and then when they're something like a movie preview.
You know, you can get Zyxel routers with equally good hardware that also run linux in the $18-25 price range... You just have to actually know what you're doing to run linux on them (Well, your own linux... They run linux out of the box) instead of installing a script-kiddy friendly installer.
Plus you don't have to deal with the community of Linksys fanboys... which rivals the communities of game console fanboys.
The Exchange database is totally intolerant of any unplanned outages, and is laid out in such a way that failures are never compartmentalized. That's one of the big benefits to Notes (not a big enough benefit to make up for it's negatives though), that every user has their own database so whole-system downtime is practially non-existant.
Any light duty exchange server in an environment that doesn't justify eliminating all single points of failure (communication, power, etc) will experience corruption. That can be from pulling a cable by accident, having a SAN switch fail, having an extended power outage, losing a UPS, etc... And then if you have any signifigant number of users you're talking many hours of downtime to rebuild.
That's terrible. You shouldn't need a full time staff to run your groupware server.
Great?
Maybe I'll give you that it's the best one that's generally available, but it's total shit. It's unstable, unreliable, and counterintuitive in many cases. It also does a poor job of third party interoperability, and has a legendary record of insecurity.
There may not be many better choices out there, but you're giving Exchange *way* too much credit.
You have to have a reasonable cost/benefit analysis. The Vonage 911 system works well in the vast majority of the cases. With millions of customers benefiting from this service, and what, one? Maybe two deaths that could even be remotely associated with having Vonage instead of POTS... Is it really worth eliminating this useful technology that could benefit hundreds of millions of people on the off chance that it may save a couple lives?
Our society needs some sanity injected into it. Life is valuable, but we can't continue as a modern society if the standard becomes that we have to protect every life at any cost. There are some cases where you have to decide what is good enough and start putting your resources elsewhere. There's an exponential curve of investment level that correlates to a linear increase in service quality here. Vonage should be allowed to pick a sane point on the curve.
You think the package deals are price gouging? Wait until you see the a la carte pricing... How many channels do you think you'll be able to subscribe to before you've exceeded the price of the package that contains them? 3? 4? It's going to be something rediculous like that.
Oh, one more things.
Only some of the code from the operating systems your listed is available under the shared source license.
First of all, Insightful my ass. The moderators of your comment should be shot.
Second, your comment is interesting, considering two things:
First, that this is an article about how Diebold can't profide North Carolina with source escrow because it can't provide the Windows code. (You did read the article, right? Or perhaps you'd like to borrow some clue?) Regardless, the shared source license it part of the marketing bullshit that Microsoft uses to create their special case, and you've completely bought into it. Source escrow typically guarantees your right to continue to redistribute and advance development a third party product should the producer cease to exist or to terminate support for a particular product. Find that guarantee in the Microsoft Shared Source license. If Microsoft terminated production of CE, people who make products based on it would be screwed. The shared source license is not even close to equivalent to source escrow.
Second that it's naive to think that any developer list is complete, or that there is even a remote chance of proving it either way. Require it all you want, but in the end you're going to end up with a worthless list of names with no way to know if it's complete, or correct.
Usually when you develop an embedded system, you demand code escrow from your suppliers. Microsoft is a special case though, because when they enter the conversation everybody seems to become stupid. If they had gone with any other vendor (I'm not just talking Linux here... They could have used VxWorks, QNX, BSD, one of the various DOSes...) they would have had code escrow. I bet they do for every other third party bit of software on their machine.
The list of developer names is pretty unreasonable, but code escrow is something that happens all the time, and only Microsoft manages to get out of it.
The fact is it simply can't happen in a language like Java.
There are a lot of things that can't happen in a language like Java. Some of them are considered features in some circumstances. At least you have the option to manage your memory or have it managed for you in C, instead of being stuck with one way or the other.
There are so many poor practices involved in that situation you just described that I don't even know where to begin.
There are dozens of simple rules you can follow when you write C code, any one of which would have prevented that problem. Either way, having your memory managed for you doesn't imply that you don't have access to the raw data anyway. Protection is only enforced in some languages.
so C++ can have the same types of so-called "hard to debug" leaks you blame on garbage collection
I said they become more difficult to debug because of garbage collection. They're certainly not caused by the garbage collection. They're caused (usually) by poor programming.
Garbage collection is a tool. It makes your job as a programmer easier, but it does not free you from the need to understand things like scope. Just because you don't have to worry about the mechanics of managing your memory, you still need to understand how it works, or you are going to write crappy code. Code that leaks.
But on top of that, if you have a true memory leak, C++ doesn't even tell you what's stored in all that leaked memory.
That's because you don't need the programming language to tell you that. There are plenty of ways you can make it easy to tell what's in your leaked memory. And then when you're done debugging, you can turn them off and reduce your memory footprint.
Actually, it seems to me that if you want reliability, maintainability, and perhaps most important, debugability, you want to manage your memory yourself.
When debugging a program with a leak (Yes, garbage collected programs have leaks too, they're just nastier, because they don't look like bugs because a reference is persisting somewhere.) if memory is program managed, finding the leak is a deterministic process. You're guaranteed success in a well-defined, and finite amount of time (The amount of time it takes to reproduce the leak plus the amount of time it takes to build the application, plus the amount of time it takes to get a basic understanding of the application structure). When you're debugging a leak in a memory managed environment, you're limited to the tools provided, which are pretty much universally not good at finding the worst kinds of leaks, because to the tools your leak looks like perfectly valid utilization. You're really doomed if your leak is in a third party library. You may never track it down, or even find the culprit. (Any Jakarta users out there?)
Sure, languages without garbage collection may be more prone to leaks, but I'd rather have more leaks that I can fix, than fewer leaks that I can't...
Not that garbage collection doesn't have it's place. If you're application is complex enough though, allocating your own memory is the least of your worries.
If your program uses lots of RAM and you need every last drop either find an expert at RAM management to get every last bit
I don't care what kind of programmer you are, you should have a good foundation in certain things. Data structures, obviously. Discrete logic. Algorithmic complexity, including the ability to read and understand Big O notation (wspecially true of java programmers)... And a complete knowledge of basic memory management. Even if you're going to use garbage collection. Without these things, you're doomed to write poor programs. If you don't understand how a utility library works, you shouldn't be using it.
Which returns us to the question of what the actual benefit of this device is over a traditional electric on-demand water heater. The only one I could see is size.
Why bother timing this stuff? It's pointless. You'll have a bunch of useless data. It's not like they only have 8k of memory and don't have room to store the procedures for the hotkeys or something. They've got modern machines. Support both. That way the users can do it how they want to do it instead of how the arrogant UI designer that knows better thinks they should do it.
Like $40? So, you don't actually know. How much else are you guessing here?
It's an educated guess. There was an article (listed on slashdot not to long ago even) about how it costs intel about $40 to make a processor, regardless of it's speed, etc...
Also, intel doesn't just keep it's old fabs around. It upgrades them. I'd bet money that the P3s in the recent Xboxes use the latest fab technology, for two reasons. First, because the old fabs have been upgraded to use the newer processes. They may not use the absolute cutting edge, but they probably use one generation back, which is a few generatinos ahead of what they were using when the Xbox was released. Second because the cost of a silicon wafer and the processing of said wafer is fixed, and it's cheaper for intel to produce the processors on the smaller process to get more chips out of a single wafer.
Finally, there's no guessing about the fact that the prices went up, the only guessing is by how much, since Microsoft isn't at liberty to disclose that information. You can look it up rather easily because there was a legal battle between Microsoft and Nvidia about it from when Nvidia jacked up the prices. It eventually led to Microsoft using ATI for the 360 instead. So I'm guessing as to exactly what the prices are, but if can be confirmed as fact that they're higher than they were on day one.
It's hard for me to see why you're getting so emotional about this. Do you really care how much it cost Microsoft to build a console that they now no longer make?
The chips that you see for $10 are ones that were left in the retail channel after production stopped. They're discounted so deeply because the retailers are stuck with them and need to cut their losses.
And no, price for chips do not increase over time. They continue to decrease. 8080, 8086, and 486 chips are still being made, sold, and purchased to this day. Only in retail do they cost more than a few dollars.
Those chips are still in wide use, and are incredibly simple. The P3 700 isn't really that simple though. It uses the same silicon and fabs that are used to produce Intel's higest profit chips. There isn't a big market for the chip except from microsoft, so unlike the 8080, etc, the core hasn't been licensed and produced elsewhere. The only place you can get it is from Intel. They pay wholesale, and the get the maximum discount (which they got right from the begining, BTW), but since then the fees for continued production of an obsolete product have been applied. The price *isn't* $10, because there isn't some unfortunate retailer in the middle to eat a loss. They have to pay more like $40, which is not much more than it costs intel to actually manufacture the chip. Remember, not only is intel making an old part for Microsoft, but they're doing it with the same materials and equipment that they could be churning out Pentium-Ms on at a high margin. (Incidentally, that probably makes the P3s in recent Xboxes incredibly overclockable.)
They had the same problem with the graphics chips. This is probably the biggest reason the 360 is coming out *now*. Other console manufacturers get to the point where they make a profit on the hardware (if they don't right at the beginning), but the fact that microsoft has to pay a premium for these obsolete chips means they still lose money on the Xbox hardware. You can tell they learned their lesson from the design of the 360. Now they licened the dies, so they can source to a different manufacturer if their current suppliers decide to jack up the prices.