I talked to my parents last night who had both come to the conclusion that they could not connect to the internet (they have DSL) because AOL had shut down to avoid the worm. The funny thing was, it turned out the SirCam virus was responible for their scewed up TCP/IP stack.
Anyhow, speed will be the deciding factor on this one.
I'll be pretty surprised if speed ends up being the deciding factor, at least for the first few places. This is a pretty complex problems with many ways you can optimize certain aspects, like overlap inversion, color nestsing, redundency elimination, etc. But to really minimize the size of the output file (the most important factor after qualification and validity) contestants have to come up with a great balance of optimizing all of those factors together. Because it of time constraints, contestants cannot effectively run multiple attempts across all optimizations to find the single smallest output (this would run in exponential time, so small inputs might work, but large ones would run for days). The winners will likely have the best algorithms for optimizing the output in some kind of polynomial time, regardless of the language used.
You have to give your social security number to get a cell phone? Is $40/month such a risk that they have to run a full credit check? I've never purchased wireless in my name, so this is news to me.
Well, you have to consider two things. One is that the consumer would probably have had to purchase a system with a 1.7GHz CPU because of the higher cost of RDRAM. The performance difference between a 1.7GHz system with RDRAM and a 1.8GHz with SDRAM is probably pretty close for most consumer applications. In fact, most consumer applications will be faster with the SDRAM system. If nothing else, average users will feel better about buying a faster clocked machine. The second thing to consider is that there will likely be interleved DDR chipsets available for low end servers (several Socket A MP boards with this capability are out or will be released soon). Sure you can interleave RDRAM too, but the incremental cost is much higher due to the tighter engineering contraints of RDRAM channels. I think that overall consumers will win (or at least feel like they won, without spending more money).
Plus, they designed the P4 from the ground up to support RDRAM. It'd be just stupid to quit supporting them now, that it's becoming almost affordable to buy Rambus memory.
I don't recall which Intel executive (one of the chief architects of the Pentium IV architecture) it was that lead to this common misconception, but he later refuted that this was what he meant. He had said that the Pentium IV was designed to take advantage of large bandwidth chipsets, like Rambus. He then went on to remind everyone that like most microprocessors, the Pentium IV was designed to depend on the supporting chipset to interface with the memory technology. The Pentium IV was therefor not optimized for a specific memory technology. Instead the chipsets were designed to optimally work with the Pentium IV and whatever memory they supported.
Receive faxes and voicemail in your email inbox, and store them on your PC. No fax machine required.
Send faxes from your email, the web, or a Palm VII. Record your voicemail message and send it as an email attachment to one person or broadcast to a group.
Don't take it out on me dude. I was agreeing with you. I was disagreeing with the guy who called YOU an idiot. Don't be so quick to punch. My position still stands that you are right about the error correction capabilities, and that the guy who says there are none other than checksums and calls you an idiot is the ignorant one. I fully studied the Reed Solomon codes in a math course appropriately titled "Error Correcting Codes", in which the professor even handed out the exact SIAM article you linked to along with the Syllabus on the first day of class.
I remember when I was in about sixth grade, after Final Fantasy (I) had been out for a little while, rumors of the Final Fantasy III being available in Japan got exaggerated (as young kids do so well) into Final Fantasy 10 is available in Japan! So perhaps the PlayStation 9 is really available in Japan, and Final Fantasy 35 will be coming to the US soon...
From what I remember, Shawn has limited control because of his uncles larger stake in the company. So although Shawn got a nice stash from it, I don't think he's the one losing the most money (or with the most potential to make money).
though I haven't played with OSX-Server, so I can't compare.
Do yourself a favor and avoid it. OS-X server should not have been sold as a "Final" product but as a beta. It crashed frequently, didn't support many devices (very very few!), didn't support SCSI disconnect (read: could not eject tapes from a TBU without crashing), didn't play nice with NFS on non-OS-X clients, and had virtually no software other than the on-board software. The one backup solution was "recalled" due to reliability and copyright issues, and the AppleShare services did not support AppleTalk. Worst of all, you had to pay hundreds of dollars for the piece of junk.
I've seen this kind of flame-fest ever since I started using the internet. Take usenet as an example. Outside of pr0n, I'd say more than half the posts in many unmoderated technical newsgroups are childish chatter. People call each other nasty stuff and say stupid things all the time. I think it's probably the whole anonymity of the experience. I'm certain that most of those people wouldn't use language like that to someone's face.
You're exactly right. This stuff even predates the widespread use of the internet, but as the access becomes easier, cheaper, and more wide-spread, it becomes more of a problem.
They are referred to as lots of words that I would happily use in friendly conversation with a friend, but never post in a public forum read by strangers.
Taco's point tells us a little bit about why: Many of these people don't have friendly conversations with a friend, at least not face-to-face friends. They have been socialized (or desocialiazed) by the Internet and their computer. They don't understand what tacht is, and they see others using this behavior and accept it as OK. The sad thing is that this type of behavior will eventually decrease the signal to noise ratio to the point that there will be so little useful information on some of these sites that people will stop using them.
It is also invalid because he is doing make menuconfig and the saving the default.config settings. The problem is, these vary a good bit between PPC and x86, and so the G4 is likely compiling things like ADB support and the PC is probably compiling things like VESA FB support, which both the other platform does not enable by default. So it isn't even different instructions, it's different sources too!
My experience with PPC Linux on a G4 400 was that it compiled many things very quickly, much faster than the PIII 600 sitting next to it. But the true measure would be application performance, such as Apache, gimp, blender, etc.
Do they think Napster users want to share music if they have to pay for it? They just need to bury it and give in. The.nap format is just the final blow to stop the faint pulse.
1: Usually by the colo people; at Exodus, they tell us to export whatever we want backed up to an NFS share over a private internal network (requires a second NIC).
It doesn't necessarily require a second NIC. You can always just do ifconfig eth0:1 192.168.1.1 and voila you have a virtual adapter. Then setup a static route if necessary. I believe a virtual adapter requires kernel support in Linux (anyone remember the option?) or whatever the OS of choice is. If you're worried about sniffer attacks, a second adapter may have better security.
IANAL either, but the partner would probably incorporate in Sealand as an independent subsidiary of the parent company, which loosens the grip of the nation the parent company is incorpoated in.
The beauty of PCs is how customizable they are, it would be a great loss to loose that.
This is why PowerPC machines are much more rare than Intel counterparts. For many purposes, a PowerPC system is much faster than a similarly clocked PII/III. But the price of a PowerPC is artificially high because you can only buy a couple of motherboard models at a time, and there is no flexability as to which case you can get, and little flexability what options you can get. While they have PCI and AGP slots, you have to buy whatever AGP adapter Apple want to put in there first. And IBM has made PowerPC workstations, but those are marketed at the higher end, and at installations where variation is not necessary. Because of this limited competition and flexability, prices are artificially high and user acceptance is much lower
If it is $30,000 I don't think they'll need to worry. Adding some measures to decrease piracy incidence to a $30,000 machine will add perhaps $500 to the cost and a few cents to the cost of each book. It isn't like people are going to have these hooked up to their home computer, and they won't be standard with all new iMacs. hehe
You are correct that machines are overpowered and a 0.01 second increase in the 0.1 second spellcheck is negligable to the user. The defense needs to push hard to prove that the prosecution is wrong by asserting that this caused damages. It is likely they will point out that it made the 350MHz PII like a 333MHz PII and the damages should be relative to the price difference.
As for the networking, he may be screwed with those charges and forced to settle once they throw out or significantly reduce the charges of damages for cpu consumption.
Secondly, you'll need an expert witness familiar with process scheduling to explain why the dnet client doesn't reduce the computing power of the machines, and thus there was no cost incurred by diminishing the value of the machines for their intended use.
An expert witness is going to be cross examined. This computer lab was most likely Windows 9x or NT, so the scheduler will likely allow the dnet client to consume approximately 50% of the CPU while other non-kernel tasks use as much CPU as they can. Even with the priority set low, with any scheduler, you have overhead since the dnet client is using memory bandwidth, cache, and increasing the time of scheduling. Context switches further decrease efficiency since there is one more set of context switches necessary to tend to the dnet client. Even with the most nice setting (or whatever NT calls it, I forget) there is some slowdown, even with a nearly perfect scheduler due to cache misses/page faults and context switches. So while it can be argued that there was minimal disruption, the dnet client did detract from the responsiveness of the machines, and therefor their value. The defense will have to show that this detraction was no where near enough to warrant a felony. The jury is in for a long ride trying to understand all of this.
By contrast, more than a few companies have hot spare buildings. You heard that right: If, one day, the office should cease to exist, everyone may go to another.
The hotspare building where I work is called The Trace, and it is a nice bar several blocks from here. Whenever the power fails for more than an hour, we frequently migrate to our hot spare.
Chances are Georgia has a WAN connecting schools via 128Kbps ISDN (remember this started 2 years ago) and then a gateway with a T1. To save money, they probably were being billed in a 90th or 95th percentile for usage. So the excess bandwidth used all day by many computers constantly downloading new keys and uploading results may have been enough to break them out of their billing tier into another, more expensive tier. Anyone who is familiar with the school district have any insider perspective on their LAN/WAN configuration?
Did we not all see this one coming? Come on, anyone who's ever taken accounting knows that the margins on groceries are tiny (4%-5% vs the ideal 9%-12% for public corporations). Through in the number of times Webvan almost fell, and it is no surprise that they are closing up shop. What is surprising is that they made it this long.
Slashdot Mirror
I talked to my parents last night who had both come to the conclusion that they could not connect to the internet (they have DSL) because AOL had shut down to avoid the worm. The funny thing was, it turned out the SirCam virus was responible for their scewed up TCP/IP stack.
Anyhow, speed will be the deciding factor on this one.
I'll be pretty surprised if speed ends up being the deciding factor, at least for the first few places. This is a pretty complex problems with many ways you can optimize certain aspects, like overlap inversion, color nestsing, redundency elimination, etc. But to really minimize the size of the output file (the most important factor after qualification and validity) contestants have to come up with a great balance of optimizing all of those factors together. Because it of time constraints, contestants cannot effectively run multiple attempts across all optimizations to find the single smallest output (this would run in exponential time, so small inputs might work, but large ones would run for days). The winners will likely have the best algorithms for optimizing the output in some kind of polynomial time, regardless of the language used.
Nobody's It is is my employer's name.
You have to give your social security number to get a cell phone? Is $40/month such a risk that they have to run a full credit check? I've never purchased wireless in my name, so this is news to me.
Well, you have to consider two things. One is that the consumer would probably have had to purchase a system with a 1.7GHz CPU because of the higher cost of RDRAM. The performance difference between a 1.7GHz system with RDRAM and a 1.8GHz with SDRAM is probably pretty close for most consumer applications. In fact, most consumer applications will be faster with the SDRAM system. If nothing else, average users will feel better about buying a faster clocked machine. The second thing to consider is that there will likely be interleved DDR chipsets available for low end servers (several Socket A MP boards with this capability are out or will be released soon). Sure you can interleave RDRAM too, but the incremental cost is much higher due to the tighter engineering contraints of RDRAM channels. I think that overall consumers will win (or at least feel like they won, without spending more money).
Plus, they designed the P4 from the ground up to support RDRAM. It'd be just stupid to quit supporting them now, that it's becoming almost affordable to buy Rambus memory.
I don't recall which Intel executive (one of the chief architects of the Pentium IV architecture) it was that lead to this common misconception, but he later refuted that this was what he meant. He had said that the Pentium IV was designed to take advantage of large bandwidth chipsets, like Rambus. He then went on to remind everyone that like most microprocessors, the Pentium IV was designed to depend on the supporting chipset to interface with the memory technology. The Pentium IV was therefor not optimized for a specific memory technology. Instead the chipsets were designed to optimally work with the Pentium IV and whatever memory they supported.
J2 (www.j2.com) has a the following services:
Receive faxes and voicemail in your email inbox, and store them on your PC. No fax machine required.
Send faxes from your email, the web, or a Palm VII. Record your voicemail message and send it as an email attachment to one person or broadcast to a group.
They have several other services too.
Don't take it out on me dude. I was agreeing with you. I was disagreeing with the guy who called YOU an idiot. Don't be so quick to punch. My position still stands that you are right about the error correction capabilities, and that the guy who says there are none other than checksums and calls you an idiot is the ignorant one. I fully studied the Reed Solomon codes in a math course appropriately titled "Error Correcting Codes", in which the professor even handed out the exact SIAM article you linked to along with the Syllabus on the first day of class.
I remember when I was in about sixth grade, after Final Fantasy (I) had been out for a little while, rumors of the Final Fantasy III being available in Japan got exaggerated (as young kids do so well) into Final Fantasy 10 is available in Japan! So perhaps the PlayStation 9 is really available in Japan, and Final Fantasy 35 will be coming to the US soon...
From what I remember, Shawn has limited control because of his uncles larger stake in the company. So although Shawn got a nice stash from it, I don't think he's the one losing the most money (or with the most potential to make money).
though I haven't played with OSX-Server, so I can't compare.
Do yourself a favor and avoid it. OS-X server should not have been sold as a "Final" product but as a beta. It crashed frequently, didn't support many devices (very very few!), didn't support SCSI disconnect (read: could not eject tapes from a TBU without crashing), didn't play nice with NFS on non-OS-X clients, and had virtually no software other than the on-board software. The one backup solution was "recalled" due to reliability and copyright issues, and the AppleShare services did not support AppleTalk. Worst of all, you had to pay hundreds of dollars for the piece of junk.
I've seen this kind of flame-fest ever since I started using the internet. Take usenet as an example. Outside of pr0n, I'd say more than half the posts in many unmoderated technical newsgroups are childish chatter. People call each other nasty stuff and say stupid things all the time. I think it's probably the whole anonymity of the experience. I'm certain that most of those people wouldn't use language like that to someone's face.
You're exactly right. This stuff even predates the widespread use of the internet, but as the access becomes easier, cheaper, and more wide-spread, it becomes more of a problem.
They are referred to as lots of words that I would happily use in friendly conversation with a friend, but never post in a public forum read by strangers.
Taco's point tells us a little bit about why: Many of these people don't have friendly conversations with a friend, at least not face-to-face friends. They have been socialized (or desocialiazed) by the Internet and their computer. They don't understand what tacht is, and they see others using this behavior and accept it as OK. The sad thing is that this type of behavior will eventually decrease the signal to noise ratio to the point that there will be so little useful information on some of these sites that people will stop using them.
It is also invalid because he is doing make menuconfig and the saving the default .config settings. The problem is, these vary a good bit between PPC and x86, and so the G4 is likely compiling things like ADB support and the PC is probably compiling things like VESA FB support, which both the other platform does not enable by default. So it isn't even different instructions, it's different sources too!
My experience with PPC Linux on a G4 400 was that it compiled many things very quickly, much faster than the PIII 600 sitting next to it. But the true measure would be application performance, such as Apache, gimp, blender, etc.
Do they think Napster users want to share music if they have to pay for it? They just need to bury it and give in. The .nap format is just the final blow to stop the faint pulse.
1: Usually by the colo people; at Exodus, they tell us to export whatever we want backed up to an NFS share over a private internal network (requires a second NIC).
It doesn't necessarily require a second NIC. You can always just do ifconfig eth0:1 192.168.1.1 and voila you have a virtual adapter. Then setup a static route if necessary. I believe a virtual adapter requires kernel support in Linux (anyone remember the option?) or whatever the OS of choice is. If you're worried about sniffer attacks, a second adapter may have better security.
IANAL either, but the partner would probably incorporate in Sealand as an independent subsidiary of the parent company, which loosens the grip of the nation the parent company is incorpoated in.
The beauty of PCs is how customizable they are, it would be a great loss to loose that.
This is why PowerPC machines are much more rare than Intel counterparts. For many purposes, a PowerPC system is much faster than a similarly clocked PII/III. But the price of a PowerPC is artificially high because you can only buy a couple of motherboard models at a time, and there is no flexability as to which case you can get, and little flexability what options you can get. While they have PCI and AGP slots, you have to buy whatever AGP adapter Apple want to put in there first. And IBM has made PowerPC workstations, but those are marketed at the higher end, and at installations where variation is not necessary. Because of this limited competition and flexability, prices are artificially high and user acceptance is much lower
If it is $30,000 I don't think they'll need to worry. Adding some measures to decrease piracy incidence to a $30,000 machine will add perhaps $500 to the cost and a few cents to the cost of each book. It isn't like people are going to have these hooked up to their home computer, and they won't be standard with all new iMacs. hehe
You are correct that machines are overpowered and a 0.01 second increase in the 0.1 second spellcheck is negligable to the user. The defense needs to push hard to prove that the prosecution is wrong by asserting that this caused damages. It is likely they will point out that it made the 350MHz PII like a 333MHz PII and the damages should be relative to the price difference. As for the networking, he may be screwed with those charges and forced to settle once they throw out or significantly reduce the charges of damages for cpu consumption.
Secondly, you'll need an expert witness familiar with process scheduling to explain why the dnet client doesn't reduce the computing power of the machines, and thus there was no cost incurred by diminishing the value of the machines for their intended use.
An expert witness is going to be cross examined. This computer lab was most likely Windows 9x or NT, so the scheduler will likely allow the dnet client to consume approximately 50% of the CPU while other non-kernel tasks use as much CPU as they can. Even with the priority set low, with any scheduler, you have overhead since the dnet client is using memory bandwidth, cache, and increasing the time of scheduling. Context switches further decrease efficiency since there is one more set of context switches necessary to tend to the dnet client. Even with the most nice setting (or whatever NT calls it, I forget) there is some slowdown, even with a nearly perfect scheduler due to cache misses/page faults and context switches. So while it can be argued that there was minimal disruption, the dnet client did detract from the responsiveness of the machines, and therefor their value. The defense will have to show that this detraction was no where near enough to warrant a felony. The jury is in for a long ride trying to understand all of this.
By contrast, more than a few companies have hot spare buildings. You heard that right: If, one day, the office should cease to exist, everyone may go to another.
The hotspare building where I work is called The Trace, and it is a nice bar several blocks from here. Whenever the power fails for more than an hour, we frequently migrate to our hot spare.
I wonder what the statue of limitations is for this type of crime in Georgia. It is possible they may get the first year thrown out...
Chances are Georgia has a WAN connecting schools via 128Kbps ISDN (remember this started 2 years ago) and then a gateway with a T1. To save money, they probably were being billed in a 90th or 95th percentile for usage. So the excess bandwidth used all day by many computers constantly downloading new keys and uploading results may have been enough to break them out of their billing tier into another, more expensive tier. Anyone who is familiar with the school district have any insider perspective on their LAN/WAN configuration?
Did we not all see this one coming? Come on, anyone who's ever taken accounting knows that the margins on groceries are tiny (4%-5% vs the ideal 9%-12% for public corporations). Through in the number of times Webvan almost fell, and it is no surprise that they are closing up shop. What is surprising is that they made it this long.
I believe there is a reference to it in this book: Cultural Logics and Global Economies: Maya Identity in Thought and Practice. In any case you could contact the author of that book, Edward Fischer, and ask him if he knows of any publiations on the subject. He is the one who discussed this in a class I took.