I guess that's a decision that only they can take.
Is it more likely that we missed something due to not doing enough analysis on the data, or due to not processing the right frequencies.
They have already increased the amount of data processing done on the raw data at least once with the introduction of the triplet analysis
To turn it around, I'd hate to think we missed something by wasting resources on ever more complex processing of a limited frequency band, when there was a very obvious signal out there but we just didn't bother to listen on the right frequency.
As a footnote, it's interesting to note IBM's position in the high-end workstation and mid-range server markets, where all disks are SCSI, most are 10000rpm, and Seagate is (IMHO) IBM's main competitor.
In this market IBM have only recently launched 72GB drives (and these are still pretty thin on the ground) -- previously their largest was 36GB -- despite the fact that Seagate have been shipping 72GB drives for some time, and are now shipping 180GB SCSI drives (though admitedly the latter only at 7200rpm).
It always struck me as odd that IBM could ship large IDE (consumer) drives, but limited themselves to 36GB SCSI (professional) drives.
Now, there could be a perfectly innocent explanation for this; namely that IBM couldn't produce large drives fast enought to satisy the server market, so chose not to bother. (This would make sense, since there does appear to be a trade-off between size and speed with state-of-the-art drives.)
Still, you have to wonder whether they chose to release an immature technology in their IDE drives in order to compete in the consumer market, whilst knowing better than to risk this in the high-end SCSI market.
[Aside: as a footnote to this footnote, I should perhaps dispell a few myths and explain (my take on) the main differences between IDE and SCSI drives. It has little to do with the differences between the IDE and SCSI interfaces and which is faster (though SCSI does have certain advantages in high-end and server environments). The main difference is that SCSI drives are higher performance (and perhaps higher quality) because they are intended for use in servers and high-end workstations. The other big difference, of course, is price. Obviously this is partly (perhaps even largely) because SCSI drives are produced in smaller quantities; but in part it is because you are buying a higher quality product. Summary: you get what you pay for]
It seems to me that hard drive customers should be spending more time buying quality drives instead of trying to find drives that give them the most GB for their $$.
But if that's how most customers measure value, that's what will drive manufactures.
Generally we use IBM Ultrastar drives (10000rpm, Ultra160 SCSI), typically 18GB, 36GB when needed. (The 72GB Ultrastars are still pretty difficult to get hold of....)
Those of you used to consumer drives will no doubt thing they are too small and too expensive, but their fast and (in my experience) reliable.
You get what you pay for
Re:What a well-written article!
on
Taming the Web
·
· Score: 1
Gee, what an intelligent statement. I know there are so many providers that drop my telnet traffic.... oh wait, I've never seen that happen, even when I was visiting china.
Don't knock it -- 90%+ of the consumer market has no use for telnet. If a major provider decided to do just that, there might be an outcry amongst those of us who know what telnet is, but the world (and the ISP) would go on...
The Internet will never be completely controlled
on
Taming the Web
·
· Score: 1
Arggg... There should be a clear warning to new users to enter the comment in HTML rather than English...:)
2nd attempt:
The Internet is becoming more regulated; that is a fact. This trend is more than likely to continue.
But this article, like so many others, appears to have been written by someone who doesn't understand (at a technical level) what either a computer of the Internet really is. The author picks three arguments which -- he claims -- are the arguments that the Internet cannot be controlled -- and then refutes them.
None of these articles ever attempt to technically justify that the Internet can be controlled; in some sense, it cannot -- but it depends what you mean.
As long as we have an end-to-end packet transport (aka the Internet), and general-purpose stored-program computers (eg PCs), we can do anything. Throw in a little strong cryptography, and it becomes impossible for anyone to know -- let alone control -- what we choose to do with this infrastructure.
It seems implausible to suggest that either the end-to-end network or the computer will cease to exist in the foreseeable future... What is possible, of course, is that either (or both) of these technologies will become less accessible to consumers.
General-purpose computers might become expensive specialist items if the mass-market tended towards low-cost dedicated devices that encompassed the popular PC functionality. Imagine a low-cost dedicated word-processor, dedicated email/web terminal, and a games console. Either as three separate devices, or as a single integrated device. All of these things exist, but with the exception of the games console, none has yet been successful. This failure, however, is more to do with price and functionality that with the fundamental idea. If the dedicated appliances were as featureful as their current PC incarnations, and substantially cheaper, then the home PC market would crash. Over time, 90% of PC users would cease to be PC users. Volumes would drop, prices would rise, and 90% of the remainder would not be able to justify the cost of a PC.
The other part of the equation is that consumers will over time lose easy access to the end-to-end packet network that is the Internet. This is already starting to happen -- we have ISPs which proxy not just HTTP but also SMTP. They filter incoming access to many services, and perhaps outing access to some, too. Already they are not ISPs in the the pure sense of IP connectivity providers. If this trend continues then, over a period of time, consumer ISPs will cease to be Internet Service Providers, but will become Information Serivice Providers. You will continue to talk to your ISP with IP, but you will no longer be connecting to the Internet. You'll still be able to surf the web, send e-mail, watch streaming video, but you won't -- truly -- be connecting to the Internet.
The point is that whilst both the computer and the Internet might cease to be consumer items in the pure forms, they will continue to exist, just as they did long before most consumers had access to them.
The consequences of this could be quite interesting -- general purpose computers and end-to-end packet transport might once again be limited to computer-science departments and the research departments of IT-oriented commercial enterprises. The wheel would have turned full circle....
The Internet will never be completely controlled
on
Taming the Web
·
· Score: 1
The Internet is becoming more regulated; that is a fact. This trend is more than likely to continue.
But this article, like so many others, appears to have been written by someone who doesn't understand (at a technical level) what either a computer of the Internet really is. The author picks three arguments which -- he claims -- are the arguments that the Internet cannot be controlled -- and then refutes them.
None of these articles ever attempt to technically justify that the Internet can be controlled; in some sense, it cannot -- but it depends what you mean.
As long as we have an end-to-end packet transport (aka the Internet), and general-purpose stored-program computers (eg PCs), we can do anything. Throw in a little strong cryptography, and it becomes impossible for anyone to know -- let alone control -- what we choose to do with this infrastructure.
It seems implausible to suggest that either the end-to-end network or the computer will cease to exist in the foreseeable future... What is possible, of course, is that either (or both) of these technologies will become less accessible to consumers.
General-purpose computers might become expensive specialist items if the mass-market tended towards low-cost dedicated devices that encompassed the popular PC functionality. Imagine a low-cost dedicated word-processor, dedicated email/web terminal, and a games console. Either as three separate devices, or as a single integrated device. All of these things exist, but with the exception of the games console, none has yet been successful. This failure, however, is more to do with price and functionality that with the fundamental idea. If the dedicated appliances were as featureful as their current PC incarnations, and substantially cheaper, then the home PC market would crash. Over time, 90% of PC users would cease to be PC users. Volumes would drop, prices would rise, and 90% of the remainder would not be able to justify the cost of a PC.
The other part of the equation is that consumers will over time lose easy access to the end-to-end packet network that is the Internet. This is already starting to happen -- we have ISPs which proxy not just HTTP but also SMTP. They filter incoming access to many services, and perhaps outing access to some, too. Already they are not ISPs in the the pure sense of IP connectivity providers. If this trend continues then, over a period of time, consumer ISPs will cease to be Internet Service Providers, but will become Information Serivice Providers. You will continue to talk to your ISP with IP, but you will no longer be connecting to the Internet. You'll still be able to surf the web, send e-mail, watch streaming video, but you won't -- truly -- be connecting to the Internet.
The point is that whilst both the computer and the Internet might cease to be consumer items in the pure forms, they will continue to exist, just as they did long before most consumers had access to them.
The consequences of this could be quite interesting -- general purpose computers and end-to-end packet transport might once again be limited to computer-science departments and the research departments of IT-oriented commercial enterprises. The wheel would have turned full circle....
Slashdot Mirror
I guess that's a decision that only they can take.
Is it more likely that we missed something due to not doing enough analysis on the data, or due to not processing the right frequencies.
They have already increased the amount of data processing done on the raw data at least once with the introduction of the triplet analysis
To turn it around, I'd hate to think we missed something by wasting resources on ever more complex processing of a limited frequency band, when there was a very obvious signal out there but we just didn't bother to listen on the right frequency.
Probably because no-one was buying their data recorders, so they could aford to give them away.
Seriously, though, this donation is good publicity, and undoubtedly good value for money for HP (otherwise they wouldn't have done it)
-roy
I'm very cynical of reviews (of all computer components, not just hard drives).
No review tests reliabilty, because to do so would involve buying several dozen samples of the drive and running them all for serveral years.
So all reviewers can test in reality is performance. So rely on what you learn from other people, rather than from reviews.
Usenet (and Slashdot) are your friends....
-roy
As a footnote, it's interesting to note IBM's position in the high-end workstation and mid-range server markets, where all disks are SCSI, most are 10000rpm, and Seagate is (IMHO) IBM's main competitor.
In this market IBM have only recently launched 72GB drives (and these are still pretty thin on the ground) -- previously their largest was 36GB -- despite the fact that Seagate have been shipping 72GB drives for some time, and are now shipping 180GB SCSI drives (though admitedly the latter only at 7200rpm).
It always struck me as odd that IBM could ship large IDE (consumer) drives, but limited themselves to 36GB SCSI (professional) drives.
Now, there could be a perfectly innocent explanation for this; namely that IBM couldn't produce large drives fast enought to satisy the server market, so chose not to bother. (This would make sense, since there does appear to be a trade-off between size and speed with state-of-the-art drives.)
Still, you have to wonder whether they chose to release an immature technology in their IDE drives in order to compete in the consumer market, whilst knowing better than to risk this in the high-end SCSI market.
[Aside: as a footnote to this footnote, I should perhaps dispell a few myths and explain (my take on) the main differences between IDE and SCSI drives. It has little to do with the differences between the IDE and SCSI interfaces and which is faster (though SCSI does have certain advantages in high-end and server environments). The main difference is that SCSI drives are higher performance (and perhaps higher quality) because they are intended for use in servers and high-end workstations. The other big difference, of course, is price. Obviously this is partly (perhaps even largely) because SCSI drives are produced in smaller quantities; but in part it is because you are buying a higher quality product. Summary: you get what you pay for]
-roy
But if that's how most customers measure value, that's what will drive manufactures.
Generally we use IBM Ultrastar drives (10000rpm, Ultra160 SCSI), typically 18GB, 36GB when needed. (The 72GB Ultrastars are still pretty difficult to get hold of....)
Those of you used to consumer drives will no doubt thing they are too small and too expensive, but their fast and (in my experience) reliable.
You get what you pay for
Don't knock it -- 90%+ of the consumer market has no use for telnet. If a major provider decided to do just that, there might be an outcry amongst those of us who know what telnet is, but the world (and the ISP) would go on...
Arggg... There should be a clear warning to new users to enter the comment in HTML rather than English... :)
2nd attempt:
The Internet is becoming more regulated; that is a fact. This trend is more than likely to continue.
But this article, like so many others, appears to have been written by someone who doesn't understand (at a technical level) what either a computer of the Internet really is. The author picks three arguments which -- he claims -- are the arguments that the Internet cannot be controlled -- and then refutes them.
None of these articles ever attempt to technically justify that the Internet can be controlled; in some sense, it cannot -- but it depends what you mean.
As long as we have an end-to-end packet transport (aka the Internet), and general-purpose stored-program computers (eg PCs), we can do anything. Throw in a little strong cryptography, and it becomes impossible for anyone to know -- let alone control -- what we choose to do with this infrastructure.
It seems implausible to suggest that either the end-to-end network or the computer will cease to exist in the foreseeable future... What is possible, of course, is that either (or both) of these technologies will become less accessible to consumers.
General-purpose computers might become expensive specialist items if the mass-market tended towards low-cost dedicated devices that encompassed the popular PC functionality. Imagine a low-cost dedicated word-processor, dedicated email/web terminal, and a games console. Either as three separate devices, or as a single integrated device. All of these things exist, but with the exception of the games console, none has yet been successful. This failure, however, is more to do with price and functionality that with the fundamental idea. If the dedicated appliances were as featureful as their current PC incarnations, and substantially cheaper, then the home PC market would crash. Over time, 90% of PC users would cease to be PC users. Volumes would drop, prices would rise, and 90% of the remainder would not be able to justify the cost of a PC.
The other part of the equation is that consumers will over time lose easy access to the end-to-end packet network that is the Internet. This is already starting to happen -- we have ISPs which proxy not just HTTP but also SMTP. They filter incoming access to many services, and perhaps outing access to some, too. Already they are not ISPs in the the pure sense of IP connectivity providers. If this trend continues then, over a period of time, consumer ISPs will cease to be Internet Service Providers, but will become Information Serivice Providers. You will continue to talk to your ISP with IP, but you will no longer be connecting to the Internet. You'll still be able to surf the web, send e-mail, watch streaming video, but you won't -- truly -- be connecting to the Internet.
The point is that whilst both the computer and the Internet might cease to be consumer items in the pure forms, they will continue to exist, just as they did long before most consumers had access to them.
The consequences of this could be quite interesting -- general purpose computers and end-to-end packet transport might once again be limited to computer-science departments and the research departments of IT-oriented commercial enterprises. The wheel would have turned full circle....
The Internet is becoming more regulated; that is a fact. This trend is more than likely to continue. But this article, like so many others, appears to have been written by someone who doesn't understand (at a technical level) what either a computer of the Internet really is. The author picks three arguments which -- he claims -- are the arguments that the Internet cannot be controlled -- and then refutes them. None of these articles ever attempt to technically justify that the Internet can be controlled; in some sense, it cannot -- but it depends what you mean. As long as we have an end-to-end packet transport (aka the Internet), and general-purpose stored-program computers (eg PCs), we can do anything. Throw in a little strong cryptography, and it becomes impossible for anyone to know -- let alone control -- what we choose to do with this infrastructure. It seems implausible to suggest that either the end-to-end network or the computer will cease to exist in the foreseeable future... What is possible, of course, is that either (or both) of these technologies will become less accessible to consumers. General-purpose computers might become expensive specialist items if the mass-market tended towards low-cost dedicated devices that encompassed the popular PC functionality. Imagine a low-cost dedicated word-processor, dedicated email/web terminal, and a games console. Either as three separate devices, or as a single integrated device. All of these things exist, but with the exception of the games console, none has yet been successful. This failure, however, is more to do with price and functionality that with the fundamental idea. If the dedicated appliances were as featureful as their current PC incarnations, and substantially cheaper, then the home PC market would crash. Over time, 90% of PC users would cease to be PC users. Volumes would drop, prices would rise, and 90% of the remainder would not be able to justify the cost of a PC. The other part of the equation is that consumers will over time lose easy access to the end-to-end packet network that is the Internet. This is already starting to happen -- we have ISPs which proxy not just HTTP but also SMTP. They filter incoming access to many services, and perhaps outing access to some, too. Already they are not ISPs in the the pure sense of IP connectivity providers. If this trend continues then, over a period of time, consumer ISPs will cease to be Internet Service Providers, but will become Information Serivice Providers. You will continue to talk to your ISP with IP, but you will no longer be connecting to the Internet. You'll still be able to surf the web, send e-mail, watch streaming video, but you won't -- truly -- be connecting to the Internet. The point is that whilst both the computer and the Internet might cease to be consumer items in the pure forms, they will continue to exist, just as they did long before most consumers had access to them. The consequences of this could be quite interesting -- general purpose computers and end-to-end packet transport might once again be limited to computer-science departments and the research departments of IT-oriented commercial enterprises. The wheel would have turned full circle....