Not all memory is in the DIMMS. There are caches everywhere.
Now, assuming a very conservative 2ns to decode the data, that's 8ns, which is a 25% performance hit.
You are forgetting what happens when you run out of memory. You page fault, and have to access the block off of disk, which takes something like 9 milliseconds. Getting rid of one page fault for every 1000 memory accesses is going to completely wipe out that 25% performance hit.
In other words, if you are using any sort of paging file, this will almost certainly improve performance, not hurt it.
The difference is that while doubling your hard drive gives you more room for stuff, doubling your memory space dramatically improves system speed as you use less disk space for paging.
(Yes, I know that the "double" part is best case, but given the relative speeds of memory and disk, even small amount of compression will likely improve speed.)
With the disk compression utilities, one of the biggest troubles is that the amount that can fit on the disk suddenly depends on the type of data. You get amazing compression rates if you are story textfiles, horrible ones if you are storing GIFs. (The most common compressed file when disk compression was all the rage.)
This will be similar. Suddenly the amount of memory an application uses is going to be less predictable. Perhaps this isn't much of an issue as with virtual memory, people are increasingly disconnected from application memory usage.
Because of virtual memory, this is likely to greatly improve the apparent speed of the system, at least in cases where memory is moderately tight. (<128 Meg on a windows box, for example). Disk access is something like three orders of magnitude slower than memory access. If compression avoids even a few page faults, the lower page-file requirements will more than make up for the extra time to compress.
Not true. Those 24-25 users would have to be attempting to use as much badnwidth as possible. It's not like the dialup case, where users don't share each other's bandwidth.
I once worked for a company with something like 600 employees with net access and a single T1. I often got download speeds over 1Mbps.
Downloading a huge software package, a bunch of audio files, listening to streaming audio and downloading a video file, and websurfing, all at the same time.
Yeah, that sounds what the typical user does...
Hell, I rarely saturate the 384 kbps line I use, and I download quite a bit myself.
The trouble is that you are confusing your maximum use with your average use. I'd be willing to bet your average use is <50kbps.
When you add more people to the pipe, how adequate it is depends on what the average person does. If the average person only downloads large files 1% of the time, then 640 Kbps will only seem marginally slower with fifty people than with one. And in fact, a 1 Mbps line with fifty people may well seem faster than a 640 Kbps line, depending on what those users are doing.
3 Mbps is perfectly adequate for 3600 users if those 3600 users are doing what a typical student would need to do for their schoolwork. It only starts becoming inadequate if those students start doing stuff that has nothing to do with education, like downloading lots of music files, or pron.
Yes, I know that some of those students will need to download Linux distributions. However, the number of those students that need to will be low. The number of students out of 3600 trying to download a distribution at any one time will likely be on the order of 1 or 2. That's fine for a 3600 Mbps line, as long as 800 other students aren't trying to simulataneously download the latest Metallica album.
Well, personally I think it is a little worse than that in that it is solving a temporary problem. Hard drives are currently beating Moore's law quite handily these days. I just saw a 40 gig drive for $215. 60 gig drives are reasonable.
How long before we see 100 gig drives, 200 gig drives available for the general public? Not long, I'd say. And a 100 gig drive can hold the contents of 130 CDs in completely raw format.
Seems to me that people should start thinking about lossless compression. I think mp3s are just an iterim format, useful only because of disk-space cost. Why would anyone worry about compressing a song if terabyte drives were only $250? At the rate we're going, that's about eight years off.
Hell, I'm willing to bet that in ten years, people are going to start talking about file formats that produce better than CD quality.
I find that really hard to believe, at least for an average user. I did tests myself between a CD, an 192bit mp3 played through a SoundblasterLive and an 192bit mp3 played through a Soundblaster 16. I could not tell the difference.
I found the cabling itself to be far more important.
Well, that and a slightly faster CPU. You do benefit a little from the fact that your PC isn't doing the playing.
Though not a whole heck of a lot. I've found I have trouble playing music and compiling at the same time on a K6-2. I get a few skips. My other, faster box does both happily.
I've also found that you don't need a particularly top of the line sound-card either. My six-year old Soundblaster 16 does fine.
Though I personally keep the patch cord short and use long speaker cables.
I've also noted that Intiut seems to be increasingly putting in bloat rather than new functionality. The last version had lots of graphical wing-dings, but no real improvements in usefull functionality.
Unfortunately, until Gnucash can transparently talk to my bank, it'd be difficult to switch.
Quicken does keep track of acrrued interest, though. At least, it does for my loans... Admittedly, it can be a royal pain-in-the-ass to figure out.
Quite the opposite. He was important as a theorist, not an inventor. Probably one of the most important theorists of the twentieth century.
Yes, Turing was a far more important figure in history than Woz. And Albert Einstein was a more important figure in history than Alexander Graham Bell. That doesn't mean that either of them are inventors.
I assume you mean by "first time" that there'd be a fixed time limit. Otherwise, BPCB could have their own version about two weeks after BCPA did, and there'd be almost no profit.
Personally, I'd like to see copyrights that only last twenty years, or until the death of the author, whichever comes first. That would stop many of the worst abuses.
Remember, a world without copyrights would probably result in people having a greater desire for works, not less. (after all, it's trivial; many books will be available for cost + a razor thin profit) There's still lots of money to be made, it's just an intensely more competitive environment.
Lots of money for the publishers, but not the authors. The publishers are selling a physical object that has a cost, just like a shirt or a frisbee. Yes, they'll make a profit. The author, on the other hand, is making something with a huge initial investment that can be easily copied. Without some guarantee that this investment won't go for nought, there is little incentive to make it.
In other words, Neal tries to sell the manuscript to Cryptonomican II to BPCA for $500,000. BPCA says, "why should we? We'll just wait for you to find some other publisher to publish it and then copy it. We won't have to pay you a cent". Given that they can likely ramp up printing in a couple of weeks, they'll likely not gain much by being the first to get it.
Without copyright laws, there is little incentive to find new authors. Just find the ones people like, and publish their work. Even more than today, publishing will be about marketting, product placement and the like. If anyone can publish "Cryptonomicon II", then the one who makes the most money is the one who markets it best, and at the lowest cost. Obviously the one who pays the author will have trouble being the cheapest... Which do you think will boost profits more? Paying $500,000 to the author to get it a few weeks early, or paying $500,000 to Barnes and Noble to get it on the stand by the front door of all there stores?
Only a name author makes much money the first few weeks of release...
Beal would almost certainly not get a couple million for "Cryptonomicon II". At very best, he'd get the profits that BCPA made before BCPB managed to get it out. BCPA would have no incentive to pay more. And given that BCPB could likely get it out in a few weeks, that's not much money.
And for a non-famous author, it is practically zero.
If copyright were done away with, the most viable software model would being to grab all the source code you could get your hands on from other people, while using draconian and fascistic means to prevent any source code from leaking out of your own development house.
People seem to have this wierd-ass niave view that if you could legally publish the source code to Microsoft Excel, that all you'd have to do is go up to Redmond and ask Bill Gates for it. Sorry. Just because you have the legal right to something if you come upon it doesn't mean you'll ever come upon it.
Discarding copyright law would kill free software. Why? Because no author of free software would be able to prevent others from closing the source back up.
Copyleft was invented to prevent people from using other people's source code in ways that the authors didn't approve of.
This has nothing, nada, not-a-thing, to do with countering copyright laws. It uses copyright law.
Typical copyrights restrict people from selling or giving away software without paying the original author. Copyleft licenses restrict people from selling or giving away software without providing the source code.
Both place restrictions on the software. Neither contradict each other or even effect each other.
...but all it does is allow the concentration of media control.
If there were no copyright, big-huge-media-megacorp could take the latest Stephen King novel, publish it themselves, undercutting the price Mr. King's publishing house charges, and not pay one him read cent.
The question left to the reader: does this give more power to:
Stephen King?
big-huge-media-megacorp?
The answer should be obvious. Yes, artists get screwed by big corps. right now, but they have one single thing in their favor. They have the legal right to determine who publishes their work. That is why popular ones can command such high prices.
Take away that right and you take away the only sort of power an artist has.
Do you really think that a large corporation would pay, say, Neal Stephenson, lots of money for a novel if a rival corporation could just copy the result without paying? How much do you think Mr. Stephenson is going to earn in such a world?
If copyright were done away with, the GPL would be unecessary.
If copyright were done away with, I could take the Slackware distribution, make a bunch of source code changes, compile those changes and sell the resulting binaries while refusing to give anyone my source code changes.
Is that what you mean by "Unnecessary"?
I suspect that this "GPL unnecessary with no copyright" meme arose because some people have decided that because some other people misuse copyrights, that all copyrights must therefore be bad. This is black-and-white thinking, and shows a fundamental misunderstanding of both copyright, the GPL and the purpose of the GPL.
If there were no copyright, then all software would be treated pretty much like the BSD license. Anyone could do anything with whatever they could get their mits on. There's a reason that many chose the GPL instead. Because it requires that published changes include source code changes, it encourages people to provide source.
Without copyright law, this falls apart. Rather than being required to publish the source, people are encouraged to keep it secret. Because there is no copyright law to allow the original owner to put requirements on source code use, they cannot prevent people from burying the code changes in a vault to get a leg up on the competition.
Again, if there were no copyright, I could take the Linux kernal, add some stuff like "SuperDuperBus" support, and sell the resulting package without source. I could also bury little changes to make it incompatible with certain other competitor's version, but of course you wouldn't know that, because the source would only reside, in heavily encrypted form, on my hard drive. Soon, since everyone wants "SuperDuperBus" support to use the latest hardware, I've got my sourceless software out everywhere. And with those subtle incompatibilities (unexamined, remember, no one else has that source) I've gained effective control over the OS. And as long as I can keep throwing in useful features, I'll likely retain it.
Sure, I couldn't easily do it, but a big corp. could. Do you think "Corel Linux" would have come with source without being forced to by copyright law? You're niave if you do...
This is what the GPL prevents. If copyright law goes away, the GPL ceases to have meaning, and the problem it was designed to prevent are no longer prevented.
Now please go learn a little about the GPL, it's purpose, and copyright law before continuing to spread what is IMHO an extremely dangerous meme.
Yes, a "brief clip" would almost certainly fall under "fair use". You can sure as hell quote a paragraph or two of a book without permission. Isn't that the same as a 30 second clip?
Slashdot Mirror
Not to mention that it can be used in places software can't, like disk or CPU caches.
. $800 of PC133 right now is about 640MB of RAM.
Not all memory is in the DIMMS. There are caches everywhere.
Now, assuming a very conservative 2ns to decode the data, that's 8ns, which is a 25% performance hit.
You are forgetting what happens when you run out of memory. You page fault, and have to access the block off of disk, which takes something like 9 milliseconds. Getting rid of one page fault for every 1000 memory accesses is going to completely wipe out that 25% performance hit.
In other words, if you are using any sort of paging file, this will almost certainly improve performance, not hurt it.
The difference is that while doubling your hard drive gives you more room for stuff, doubling your memory space dramatically improves system speed as you use less disk space for paging.
(Yes, I know that the "double" part is best case, but given the relative speeds of memory and disk, even small amount of compression will likely improve speed.)
With the disk compression utilities, one of the biggest troubles is that the amount that can fit on the disk suddenly depends on the type of data. You get amazing compression rates if you are story textfiles, horrible ones if you are storing GIFs. (The most common compressed file when disk compression was all the rage.)
This will be similar. Suddenly the amount of memory an application uses is going to be less predictable. Perhaps this isn't much of an issue as with virtual memory, people are increasingly disconnected from application memory usage.
Because of virtual memory, this is likely to greatly improve the apparent speed of the system, at least in cases where memory is moderately tight. (<128 Meg on a windows box, for example). Disk access is something like three orders of magnitude slower than memory access. If compression avoids even a few page faults, the lower page-file requirements will more than make up for the extra time to compress.
Not true. Those 24-25 users would have to be attempting to use as much badnwidth as possible. It's not like the dialup case, where users don't share each other's bandwidth.
I once worked for a company with something like 600 employees with net access and a single T1. I often got download speeds over 1Mbps.
Downloading a huge software package, a bunch of audio files, listening to streaming audio and downloading a video file, and websurfing, all at the same time.
Yeah, that sounds what the typical user does...
Hell, I rarely saturate the 384 kbps line I use, and I download quite a bit myself.
The trouble is that you are confusing your maximum use with your average use. I'd be willing to bet your average use is <50kbps.
When you add more people to the pipe, how adequate it is depends on what the average person does. If the average person only downloads large files 1% of the time, then 640 Kbps will only seem marginally slower with fifty people than with one. And in fact, a 1 Mbps line with fifty people may well seem faster than a 640 Kbps line, depending on what those users are doing.
3 Mbps is perfectly adequate for 3600 users if those 3600 users are doing what a typical student would need to do for their schoolwork. It only starts becoming inadequate if those students start doing stuff that has nothing to do with education, like downloading lots of music files, or pron.
Yes, I know that some of those students will need to download Linux distributions. However, the number of those students that need to will be low. The number of students out of 3600 trying to download a distribution at any one time will likely be on the order of 1 or 2. That's fine for a 3600 Mbps line, as long as 800 other students aren't trying to simulataneously download the latest Metallica album.
Well, personally I think it is a little worse than that in that it is solving a temporary problem. Hard drives are currently beating Moore's law quite handily these days. I just saw a 40 gig drive for $215. 60 gig drives are reasonable.
How long before we see 100 gig drives, 200 gig drives available for the general public? Not long, I'd say. And a 100 gig drive can hold the contents of 130 CDs in completely raw format.
Seems to me that people should start thinking about lossless compression. I think mp3s are just an iterim format, useful only because of disk-space cost. Why would anyone worry about compressing a song if terabyte drives were only $250? At the rate we're going, that's about eight years off.
Hell, I'm willing to bet that in ten years, people are going to start talking about file formats that produce better than CD quality.
Because I don't want slower compiles.
I find that really hard to believe, at least for an average user. I did tests myself between a CD, an 192bit mp3 played through a SoundblasterLive and an 192bit mp3 played through a Soundblaster 16. I could not tell the difference.
I found the cabling itself to be far more important.
Well, that and a slightly faster CPU. You do benefit a little from the fact that your PC isn't doing the playing.
Though not a whole heck of a lot. I've found I have trouble playing music and compiling at the same time on a K6-2. I get a few skips. My other, faster box does both happily.
I've also found that you don't need a particularly top of the line sound-card either. My six-year old Soundblaster 16 does fine.
Though I personally keep the patch cord short and use long speaker cables.
That's exactly my experience upon first purchasing Quicken 1.0 for DOS back around 1990.
Good financial software is damn important.
Unfortunately, Quicken seems to be bloating rather than improving these days.
Quicken's budgeting still pretty much sucks.
I've also noted that Intiut seems to be increasingly putting in bloat rather than new functionality. The last version had lots of graphical wing-dings, but no real improvements in usefull functionality.
Unfortunately, until Gnucash can transparently talk to my bank, it'd be difficult to switch.
Quicken does keep track of acrrued interest, though. At least, it does for my loans... Admittedly, it can be a royal pain-in-the-ass to figure out.
What is really sad is that very few people knew then that he did more than any other man to defeat the Nazis because of war secrecy rules.
I don't dismiss is theoretical contributions.
Quite the opposite. He was important as a theorist, not an inventor. Probably one of the most important theorists of the twentieth century.
Yes, Turing was a far more important figure in history than Woz. And Albert Einstein was a more important figure in history than Alexander Graham Bell. That doesn't mean that either of them are inventors.
Turing likely isn't there. Neither are Albert Einstein or Isaac Newton.
Being a great theorist does not make you an inventor. (Though Turing did do some hands-on stuff during the war.)
As I said, the norm throughout most of history has been for copyright to be a thoroughly alien concept.
Yes, and throughout history, artists were barely compensated for their works while the people who published made fortunes.
Non-famous artists today make some money, though not a lot. Without copyright, they'd make nothing at all.
Personally, I'd like to see copyrights that only last twenty years, or until the death of the author, whichever comes first. That would stop many of the worst abuses.
Remember, a world without copyrights would probably result in people having a greater desire for works, not less. (after all, it's trivial; many books will be available for cost + a razor thin profit) There's still lots of money to be made, it's just an intensely more competitive environment.
Lots of money for the publishers, but not the authors. The publishers are selling a physical object that has a cost, just like a shirt or a frisbee. Yes, they'll make a profit. The author, on the other hand, is making something with a huge initial investment that can be easily copied. Without some guarantee that this investment won't go for nought, there is little incentive to make it.
In other words, Neal tries to sell the manuscript to Cryptonomican II to BPCA for $500,000. BPCA says, "why should we? We'll just wait for you to find some other publisher to publish it and then copy it. We won't have to pay you a cent". Given that they can likely ramp up printing in a couple of weeks, they'll likely not gain much by being the first to get it.
Without copyright laws, there is little incentive to find new authors. Just find the ones people like, and publish their work. Even more than today, publishing will be about marketting, product placement and the like. If anyone can publish "Cryptonomicon II", then the one who makes the most money is the one who markets it best, and at the lowest cost. Obviously the one who pays the author will have trouble being the cheapest... Which do you think will boost profits more? Paying $500,000 to the author to get it a few weeks early, or paying $500,000 to Barnes and Noble to get it on the stand by the front door of all there stores?
Only a name author makes much money the first few weeks of release...
Beal would almost certainly not get a couple million for "Cryptonomicon II". At very best, he'd get the profits that BCPA made before BCPB managed to get it out. BCPA would have no incentive to pay more. And given that BCPB could likely get it out in a few weeks, that's not much money.
And for a non-famous author, it is practically zero.
You're assuming I'm not using a copy-protection scheme...
Dongles. CD Keys. Horrible authentication schemes over TCP/IP. More horrible things than you can imagine.
People seem to have this wierd-ass niave view that if you could legally publish the source code to Microsoft Excel, that all you'd have to do is go up to Redmond and ask Bill Gates for it. Sorry. Just because you have the legal right to something if you come upon it doesn't mean you'll ever come upon it.
Discarding copyright law would kill free software. Why? Because no author of free software would be able to prevent others from closing the source back up.
This has nothing, nada, not-a-thing, to do with countering copyright laws. It uses copyright law.
Typical copyrights restrict people from selling or giving away software without paying the original author. Copyleft licenses restrict people from selling or giving away software without providing the source code.
Both place restrictions on the software. Neither contradict each other or even effect each other.
If there were no copyright, big-huge-media-megacorp could take the latest Stephen King novel, publish it themselves, undercutting the price Mr. King's publishing house charges, and not pay one him read cent.
The question left to the reader: does this give more power to:
Stephen King?
big-huge-media-megacorp?
The answer should be obvious. Yes, artists get screwed by big corps. right now, but they have one single thing in their favor. They have the legal right to determine who publishes their work. That is why popular ones can command such high prices.
Take away that right and you take away the only sort of power an artist has.
Do you really think that a large corporation would pay, say, Neal Stephenson, lots of money for a novel if a rival corporation could just copy the result without paying? How much do you think Mr. Stephenson is going to earn in such a world?
If copyright were done away with, I could take the Slackware distribution, make a bunch of source code changes, compile those changes and sell the resulting binaries while refusing to give anyone my source code changes.
Is that what you mean by "Unnecessary"?
I suspect that this "GPL unnecessary with no copyright" meme arose because some people have decided that because some other people misuse copyrights, that all copyrights must therefore be bad. This is black-and-white thinking, and shows a fundamental misunderstanding of both copyright, the GPL and the purpose of the GPL.
If there were no copyright, then all software would be treated pretty much like the BSD license. Anyone could do anything with whatever they could get their mits on. There's a reason that many chose the GPL instead. Because it requires that published changes include source code changes, it encourages people to provide source.
Without copyright law, this falls apart. Rather than being required to publish the source, people are encouraged to keep it secret. Because there is no copyright law to allow the original owner to put requirements on source code use, they cannot prevent people from burying the code changes in a vault to get a leg up on the competition.
Again, if there were no copyright, I could take the Linux kernal, add some stuff like "SuperDuperBus" support, and sell the resulting package without source. I could also bury little changes to make it incompatible with certain other competitor's version, but of course you wouldn't know that, because the source would only reside, in heavily encrypted form, on my hard drive. Soon, since everyone wants "SuperDuperBus" support to use the latest hardware, I've got my sourceless software out everywhere. And with those subtle incompatibilities (unexamined, remember, no one else has that source) I've gained effective control over the OS. And as long as I can keep throwing in useful features, I'll likely retain it.
Sure, I couldn't easily do it, but a big corp. could. Do you think "Corel Linux" would have come with source without being forced to by copyright law? You're niave if you do...
This is what the GPL prevents. If copyright law goes away, the GPL ceases to have meaning, and the problem it was designed to prevent are no longer prevented.
Now please go learn a little about the GPL, it's purpose, and copyright law before continuing to spread what is IMHO an extremely dangerous meme.
Yes, a "brief clip" would almost certainly fall under "fair use". You can sure as hell quote a paragraph or two of a book without permission. Isn't that the same as a 30 second clip?
Someone with moderator points can mod up things they've posted as an AC.