actually, you are incorrect about RFID tags. They don't simply say "CoatCo Coat, black, large". Every RFID has an item-specific (note: not product-specific, that's individual item-specific) code number. IIRC it's a 128-bit tag, which is considered big enough.
Reading an RFID tag isn't merely indicating what the item is, but which specific item and it is traceable to you if it was entered into a database at the time of purchase.
As has been pointed out previously, a series of floor mounted RFID readers checking the tag in your shoes can track you.
Couple that with the power of customer tracking databases, and imagine that as soon as you walk into any store, the staff know not only your name, but your salary, sexual orientation and whether you have a criminal record. Ok with that? Sorry, but I'm not.
As for TiVo, I have one, I was fully aware of their tracking when I bought it, and no, I have no problem with that. From what I know of TiVo (and I have some contacts) they're smart enough to know that if they crossed the line and started tracking individual data, it could mean the end of their company.
why does this say anything about how hot the CPU is running? The colors are showing speed of airflow, not temperature or heat dissipation, as the caption on the picture you linked to says. The high airflow is because it's inside the duct.
the tests were all run on a 64-bit machine.
The argument is not so much about whether 32-bit or 64-bit binaries run faster, but which is the faster architecture.
I'm pretty sure we don't have any apples-to-apples test platforms for that one, though
Are they really sure what they're getting into here? The laws are different in the UK than they are in the US.
What they're doing here is demanding cash as "protection money" to cover the risk that the target company might get sued over infringement of copyright, the ownership (and infringement) of which is currently in dispute.
Any UK lawyers here? Does the UK definition of "extortion" cover this?
if it ever gets released at all. As an expat Brit, I find it incredibly frustrating to find UK movies and TV series released only on DVD, only in the UK, and region coded to be only viewable in the UK. Why?
Many people make bullshit presentations that twist the facts and obfuscate the most important points. Most people these days use PowerPoint (or something like it) to make presentations. From this we conclude that it's PowerPoint's fault.
I agree. This isn't saying that Dell endorses spyware, or that they think it's a good thing. The very fact that they call it "spyware" would seem to indicate they don't exactly have a positive view. This is just a legal liability thing; they don't want to get sued by a spyware provider which accuses Dell of assisting one of their customers in the violation of an EULA.
1. & 2. Yes, they were running WinXP pro. That's because that's what the machine ships with. WinXP 64 is still in beta. While you might like this to be an AMD/Intel shootout, it isn't. It's a comparison of two shipping machines, tested as configured. So why would that have been a better comparison? Incidentally, there's no reason to think 64 bit code would run faster on any of these benchmarks. In fact, there's good reason to think it would run slower. 64 bit code has a larger footprint; if you don't actually need the 64 bit address space, it usually hurts performance.
3. The memory bandwidth description isn't wrong. It is incorrect to compare the hypertransport bandwidth to the FSB bandwidth, because they're not the same. One of the key points about the Opteron/Athlon64 architecture is that the memory controller is integrated. In the Intel architecture, the FSB carries all of the CPU/Memory transfer load, but none of the graphics load (which goes across AGP into the northbridge). In the AMD architecture, the CPU/Memory transfer load is internal, and the HT links carry the graphics load. The HT links don't go to the memory. Memory bandwidth is limited by the memory modules in both cases, not the FSB/HT links. It just isn't a meaningful comparison to compare HT bandwidth and FSB bandwidth. They carry different things.
If this had been AMD announcing successfully fabricating a 4Mbit SRAM in 65nm technology in a full size fab, everybody would be leaping up and down screaming about how this would mean the end of the road for Intel.
As it is, rather predictably, we have a million comments about how it all must be BS.
yes, I'm well aware of that (I am a silicon engineer myself, btw).
However, it's also reasonable to expect that, at least in terms of circuit design, 65nm of any flavor is going to be much different that 90nm circuit design of any flavor. And certainly both will look nothing like.35u or.25u. When I was at college, 3u was considered leading edge, and circuit design today is barely recognizable from then.
remember what "development fab" means here. This is the development fab designated "D1D". It's at Ronler Acres in Hillsboro, OR. This is not just some lab, guys. It's a full scale $2B fab. The same size that AMD just broke ground on in Dresden for their fab. Only Intel's is already running.
Incidentally, the previous development fab (D1C) which was the development fab for the 90nm process and is on the same site right next to D1D, is now a full production fab. I believe it's making Prescott and Dothan parts in preparation for next quarter's launch
"a few SDRAM cells"? 4 million is more than a few. And they're SRAM cells, not SDRAM. Kind of stuff you need for cache. It does give a pretty good indication of how well along the process is.
BTW, this is apparently being done at the fab known as D1D in Hillsboro - this isn't a small scale research lab, it's a full size production fab. That it is being done there indicates it isn't as far away as you might think.
As for your comment about SOI, why does it need to be so black and white? It's always a judgement about the benefit versus cost, and it's always possible to get the same result more than one way. IBM made a strategic decision to go down the SOI path some time ago. Intel has gone down the strained silicon path. Each has its advantages and costs, and either camp could switch to the other if they saw an advantage in doing so. But given that they have made different choices, it's unlikely that one is "wrong" and the other is "right"
and of course, the Archimedes' RISC processor, the Archimedes RISC Machine, or ARM, is very much still in existence. It's one of the most widely used embedded processors.
the reason the Linux community was asking SCO to point to Linux code which they claim infringes is because SCO had previously claimed that they couldn't point to Unix code that is infringed because that would render their trade secrets no longer protectable (or something like that).
IBM can ask for the Unix code, and that is the right thing to ask for.
the problem isn't that passwords are sent over the network in the clear. They aren't. The problem is that any security system* that relies upon passwords as the basic secret is vulnerable to an offline dictionary attack; collect a sample of an exchange where the password was used somehow to encrypt a nonce also sent over the network, and try all the words in the dictionary that might be used as a password in an attempt to replicate the same exchange. When you get a match, bingo - you have the password.
* there are some known systems that use passwords but which are not susceptible to this attack, invoving a carefully bound combination of passwords and a Diffie-Hellman exchange. I don't have a reference to hand, but such a system exists. Kerberos isn't it, though.
A CMOS chip has several layers of interconnect. In current technology, all but the lowest layer are metal (either Aluminum or Copper). That's what the whole "copper" thing was all about.
This is different; it's talking about the bottom layer of interconnect which also is used as the transistor gates. Up until now, these have almost always been some kind of polycrystalline silicon, recently with all kinds of modifications. Various manufacturers have experimented at one time or another with high melting point metals (that bit is important - you can't use Aluminum here because of processing steps that occur after the gate layer deposition) such as Tungsten or Nickel.
The use of metal for these gates is incicdental, in that the real change is in the use of a higher k dielectric. That is what makes the difference. The problem is that the polysilicon variants are chemically incompatible with the new dielectric, in the sense that the old deposition methods for growing the polysilicon layer don't work on the new dielectrics. Hence the new super-secret metal formula - the breakthrough is being able to deposit this layer on the dielectric, thus enabling the dielectric to be used - and that is what changes the leakage/Vt relationship to allow low leakage at the same time as fast switching speeds.
I think the biggest issue here is hardware. I don't believe for one second that Diebold considers the software to be their primary means of making a profit - rather, they make a profit on manufacturing and selling their custom hardware. The software, being secret, is merely the lock to keep revenue from hardware sales flowing. Design some hardware first, then software.
I'd also suggest that the hardware is the source of many of the problems.
Here's how I'd architect the hardware system at the terminal end (and yes, I'm a hardware engineer... well, ok, silicon engineer):
Hardware consists of a single CPU system, display, printer and optical reader (optical reader being of the variety that reads lottery ticket slips).
Software voting system permits voter to select entries from the voting list in order to create their voter template. When done, the voter hits "print" button which provides a laser printed vote form with boxes blacked out denoting selections. No vote is cast on printing, and user may print as many such forms as they choose.
When voter verifies printed ballot is correct, it is inserted into the reader slot. Optical reader reads paper ballot, counting result, and stores paper ballot in lockbox.
Initial vote counted from optical reader, printed forms in lockbox can be counted by machine or manually if a recount is required.
This all seems obvious enough to me. What I don't understand is why Diebold didn't build it this way. Who exactly was responsible for architectural design reviews, and why weren't the flaws pointed out then? Does Diebold just design the machine and turn up with a closed box, and say "trust us"? There's no way it should be done that way
/.ers getting wound up about a patent. Some important things you need to know:
1. The patent only covers anything that does everything in the claim, just like it says. It cannot be generalized. If you don't do all of the steps, you don't infringe.
2. Claims are often narrowed by stuff you don't see in the patent itself, but which are contained in the file wrapper, that is, the documents exchanged by the applicant (or applicant's attorneys) and the patent office. These documents must be obtained from the patent office, and are often very revealing. Typically every patent gets rejected in the first instance, on grounds of insufficient novelty, and will be appealed by the applicant saying "...but we only intended application in this narrow set of circumstances, which are different from the prior art...". All those documents are recorded and form part of the validity of the patent. They are admissible in court. It usually turns out that patents are much narrower than a reading of the patent alone implies.
Read the claim carefully. It applies only to "web" transfers, and you must upload from an existing computer and download to a "new" computer. I'd be willing to bet that there are specific circumstances contained in the file wrapper also.
I don't think this is very alarming, in summary. It's probably a much narrower patent than it first appears. Really, this kind of thing happens all the time and it isn't a big deal. Move along, nothing to see here.
The general trend of change in a capitalist system is to make things which used to be expensive cheap, because ideas and technologies that don't do that disappear.
This puts businesses that can't adjust out of business.
And that is scary.
I don't have any solutions, and that is scary. If I did, it wouldn't be scary anymore.
Pay me to speak at your convention/wedding/bar mitzvah, and I can imbibe you with my enormous wisdom at explaining how scary the future is.
Ned Ludd is my hero. He saw how scary all this was. Sheesh, some people just don't think before they invent things.
please, guys, grow up. You can't expect to sue somebody or throw them in jail (or even get away with calling them a thief) just because they point out that your product is defective.
Let's recap two essential points of your argument:
1. The critique is flawed because it is based on "incorrect assumptions" because the author "didn't read your white paper explaining how the system works before writing it". and
2. He probably committed a felony by comprehensively breaking the system.
Guys, you can't possibly have both of these. They are mutually exclusive. If he did, as you say, actually circumvent the system, then his conclusion that the system doesn't work can't be invalid, can it now?
no, that isn't it. The deal here is that both OSs run *simultaneously on virtual machines*, while both having access to the same disk. Not the same as dual booting...
Slashdot Mirror
actually, you are incorrect about RFID tags. They don't simply say "CoatCo Coat, black, large". Every RFID has an item-specific (note: not product-specific, that's individual item-specific) code number. IIRC it's a 128-bit tag, which is considered big enough. Reading an RFID tag isn't merely indicating what the item is, but which specific item and it is traceable to you if it was entered into a database at the time of purchase. As has been pointed out previously, a series of floor mounted RFID readers checking the tag in your shoes can track you. Couple that with the power of customer tracking databases, and imagine that as soon as you walk into any store, the staff know not only your name, but your salary, sexual orientation and whether you have a criminal record. Ok with that? Sorry, but I'm not. As for TiVo, I have one, I was fully aware of their tracking when I bought it, and no, I have no problem with that. From what I know of TiVo (and I have some contacts) they're smart enough to know that if they crossed the line and started tracking individual data, it could mean the end of their company.
why does this say anything about how hot the CPU is running? The colors are showing speed of airflow, not temperature or heat dissipation, as the caption on the picture you linked to says. The high airflow is because it's inside the duct.
the tests were all run on a 64-bit machine. The argument is not so much about whether 32-bit or 64-bit binaries run faster, but which is the faster architecture. I'm pretty sure we don't have any apples-to-apples test platforms for that one, though
Are they really sure what they're getting into here? The laws are different in the UK than they are in the US. What they're doing here is demanding cash as "protection money" to cover the risk that the target company might get sued over infringement of copyright, the ownership (and infringement) of which is currently in dispute. Any UK lawyers here? Does the UK definition of "extortion" cover this?
Krill
Many people make bullshit presentations that twist the facts and obfuscate the most important points. Most people these days use PowerPoint (or something like it) to make presentations. From this we conclude that it's PowerPoint's fault.
Please
Krill
Krill
3. The memory bandwidth description isn't wrong. It is incorrect to compare the hypertransport bandwidth to the FSB bandwidth, because they're not the same. One of the key points about the Opteron/Athlon64 architecture is that the memory controller is integrated. In the Intel architecture, the FSB carries all of the CPU/Memory transfer load, but none of the graphics load (which goes across AGP into the northbridge). In the AMD architecture, the CPU/Memory transfer load is internal, and the HT links carry the graphics load. The HT links don't go to the memory. Memory bandwidth is limited by the memory modules in both cases, not the FSB/HT links. It just isn't a meaningful comparison to compare HT bandwidth and FSB bandwidth. They carry different things.
As it is, rather predictably, we have a million comments about how it all must be BS.
However, it's also reasonable to expect that, at least in terms of circuit design, 65nm of any flavor is going to be much different that 90nm circuit design of any flavor. And certainly both will look nothing like .35u or .25u. When I was at college, 3u was considered leading edge, and circuit design today is barely recognizable from then.
Incidentally, the previous development fab (D1C) which was the development fab for the 90nm process and is on the same site right next to D1D, is now a full production fab. I believe it's making Prescott and Dothan parts in preparation for next quarter's launch
Why do you imagine that the processor design stops just because the process advances to the next gen.
BTW, this is apparently being done at the fab known as D1D in Hillsboro - this isn't a small scale research lab, it's a full size production fab. That it is being done there indicates it isn't as far away as you might think.
As for your comment about SOI, why does it need to be so black and white? It's always a judgement about the benefit versus cost, and it's always possible to get the same result more than one way. IBM made a strategic decision to go down the SOI path some time ago. Intel has gone down the strained silicon path. Each has its advantages and costs, and either camp could switch to the other if they saw an advantage in doing so. But given that they have made different choices, it's unlikely that one is "wrong" and the other is "right"
and of course, the Archimedes' RISC processor, the Archimedes RISC Machine, or ARM, is very much still in existence. It's one of the most widely used embedded processors.
IBM can ask for the Unix code, and that is the right thing to ask for.
Krill
* there are some known systems that use passwords but which are not susceptible to this attack, invoving a carefully bound combination of passwords and a Diffie-Hellman exchange. I don't have a reference to hand, but such a system exists. Kerberos isn't it, though.
Krill
A CMOS chip has several layers of interconnect. In current technology, all but the lowest layer are metal (either Aluminum or Copper). That's what the whole "copper" thing was all about.
This is different; it's talking about the bottom layer of interconnect which also is used as the transistor gates. Up until now, these have almost always been some kind of polycrystalline silicon, recently with all kinds of modifications. Various manufacturers have experimented at one time or another with high melting point metals (that bit is important - you can't use Aluminum here because of processing steps that occur after the gate layer deposition) such as Tungsten or Nickel.
The use of metal for these gates is incicdental, in that the real change is in the use of a higher k dielectric. That is what makes the difference. The problem is that the polysilicon variants are chemically incompatible with the new dielectric, in the sense that the old deposition methods for growing the polysilicon layer don't work on the new dielectrics. Hence the new super-secret metal formula - the breakthrough is being able to deposit this layer on the dielectric, thus enabling the dielectric to be used - and that is what changes the leakage/Vt relationship to allow low leakage at the same time as fast switching speeds.
Krill
I'd also suggest that the hardware is the source of many of the problems.
Here's how I'd architect the hardware system at the terminal end (and yes, I'm a hardware engineer... well, ok, silicon engineer):
Hardware consists of a single CPU system, display, printer and optical reader (optical reader being of the variety that reads lottery ticket slips).
Software voting system permits voter to select entries from the voting list in order to create their voter template. When done, the voter hits "print" button which provides a laser printed vote form with boxes blacked out denoting selections. No vote is cast on printing, and user may print as many such forms as they choose.
When voter verifies printed ballot is correct, it is inserted into the reader slot. Optical reader reads paper ballot, counting result, and stores paper ballot in lockbox.
Initial vote counted from optical reader, printed forms in lockbox can be counted by machine or manually if a recount is required.
This all seems obvious enough to me. What I don't understand is why Diebold didn't build it this way. Who exactly was responsible for architectural design reviews, and why weren't the flaws pointed out then? Does Diebold just design the machine and turn up with a closed box, and say "trust us"? There's no way it should be done that way
Krill
1. The patent only covers anything that does everything in the claim, just like it says. It cannot be generalized. If you don't do all of the steps, you don't infringe.
2. Claims are often narrowed by stuff you don't see in the patent itself, but which are contained in the file wrapper, that is, the documents exchanged by the applicant (or applicant's attorneys) and the patent office. These documents must be obtained from the patent office, and are often very revealing. Typically every patent gets rejected in the first instance, on grounds of insufficient novelty, and will be appealed by the applicant saying "...but we only intended application in this narrow set of circumstances, which are different from the prior art...". All those documents are recorded and form part of the validity of the patent. They are admissible in court. It usually turns out that patents are much narrower than a reading of the patent alone implies.
Read the claim carefully. It applies only to "web" transfers, and you must upload from an existing computer and download to a "new" computer. I'd be willing to bet that there are specific circumstances contained in the file wrapper also.
I don't think this is very alarming, in summary. It's probably a much narrower patent than it first appears. Really, this kind of thing happens all the time and it isn't a big deal. Move along, nothing to see here.
Krill
The general trend of change in a capitalist system is to make things which used to be expensive cheap, because ideas and technologies that don't do that disappear.
This puts businesses that can't adjust out of business.
And that is scary.
I don't have any solutions, and that is scary. If I did, it wouldn't be scary anymore.
Pay me to speak at your convention/wedding/bar mitzvah, and I can imbibe you with my enormous wisdom at explaining how scary the future is.
Ned Ludd is my hero. He saw how scary all this was. Sheesh, some people just don't think before they invent things.
Krill
Krill
Krill
Let's recap two essential points of your argument:
1. The critique is flawed because it is based on "incorrect assumptions" because the author "didn't read your white paper explaining how the system works before writing it". and
2. He probably committed a felony by comprehensively breaking the system.
Guys, you can't possibly have both of these. They are mutually exclusive. If he did, as you say, actually circumvent the system, then his conclusion that the system doesn't work can't be invalid, can it now?
Krill
"it's good for riaa. it's good for us. it's good for webster's. heck, it's not like any of the kids online use them these days anyhow"
no, that isn't it. The deal here is that both OSs run *simultaneously on virtual machines*, while both having access to the same disk. Not the same as dual booting...