One thing missing from the summary is that 3.5 TeV/beam is only (hoïpefully) a very temporary setup. The ramp up to 5TeV/beam, or 10 TeV centre of mass energy should be quick rapid if everything works. Going to the full 7 TeV will take longer though.
That definition of that meaning of "computer" is correct
You have to allow for meanings to change with time. That might have been a common definition 50+ years ago when humans needed to do that sort of thing. However they have been replaced with machines that carry the same name. Stop someone in the highstreet and ask them what a computer is and I can just about guarentee you (unless I'm really unlucky and stop you!) that they will describe a computer as a machine. That doesn't mean that the original meaning is now wrong it just means that the most common meaning of the term has changed.
Sorry for quoting Wikipedia - I'm not particularly fond of it either - but it was top of the heap of a quick google search and is often quick and convenient.
It does seem like a bit more applied science to work out the kinks in this 'relatively new bit of technology' would be in order
I think the issue is more a lack of experience with the technology rather than kinks in it. There really was a huge amount of research, development and testing that went on behind the scenes. The problem is that superconducting magnets is a new technology for CERN - all previous accelerator magnets were "warm" - but superconductor technology was the only way to generate the field required within an affordable power budget.
The problem is that with superconducting you get a whole different series of issues that you need to be careful about. For example repairing the power bar connection between warm magnets is a delay of 1-2 days with superconducting technology it is 1-2 months (warm, fix, cool) so ensuring that the connections are good takes on a whole new level of importance! Things like this are what you get from experience - it is not so much a kink in the technology as simply that is behaves differently from the previous technology. Unfortunately deploying superconducting magnets on such a huge scale has never been done before - the closest would be the Tevatron which is ~10 times smaller and CERN did consult extensively with Fermilab. The scale is also part of the reason for the slow fix times - there are thousands of connections to be checked and fixed.
In contrast, the way that you want to use the term "analogue computer" shows there to be no logical distinction....
Actually, if you reread my original post, I do not want to use the term that way. My argument was in fact that you you accept the Antikythera machine as a computer you also have to accept a stick in the sand - basically reductio ad absurdum.
My definition of a computer would be limited to devices capable of following a programmed series of instructions by themselves. Indeed Wikipedia says that "The ability to store and execute lists of instructions called programs makes computers extremely versatile, distinguishing them from calculators." The Antikythera machine is a calculator, not a computer.
The primary aim of the upgrade is to increase luminosity, not energy (although in some respects the two are very similar). Increasing to 7TeV would not be an upgrade - it is intended as a fix that will be applied when we can have sufficient down time. For the moment it was deemed that it was better for the physics to get some data now, even if only at 10TeV.
If good elecrical connections sufficient to meet the needs of the LHC are within the realm of what is known by humans, then I'll cry "incompetence".
I can see your point of view but what you need to know is that these are vastly different from ordinary electrical connections. The "wire" consists of a superconducting metal core which is surrounded by copper. The copper is there so that in the case of a quench (which happens extremely rapidly since it is a phase change) the copper can carry the almost 10kA current for long enough that it can be dumped into a resistor which is the size of a small room which gets very hot. This design is needed because copper is not a superconductor but has a far lower resistivity than things which do superconduct (obviously only when they are not in their superconducting! phase!). The problems were caused by the splicing process between lengths of this "wire" during assembly which caused the layers to separate near the join.
Perhaps this sheds a little light onto why it is so complex? We are not talking about soldering two wires together we are talking about the safety backup system for a 27km long power bar that carries almost 10kA of current at a temperature of 2K (for reference outer space is warmer at 2.7K). Superconductivity itself was only discovered in 1911 and, as recently at 1962, people were winning Nobel prizes for making theoretical predictions about its behaviour at junctions.
So in reality this is a relatively new bit of technology that has never been applied on this scale before. Testing was done in advance - considerable testing - but a lot of the problems were either not visible in short scale tests or were only present when large sections were assembled. In addition, even if you do completely understand the physics involved in building something it is not always possible to predict all possible outcomes. Even today planes still crash for understandable, but unpredicted, reasons. Of course as we gain experience we learn how they fail and build them so they don't fail in the same way again but, practically speaking, you will never be able to predict all the ways that a complex bit of machinery can fail.
... best of what's still around. I've noticed a distinct decline in the quality of professional services in the last decade.
Unrelated. The LHC failures have all been caused by unforeseen consequences of standard techniques applied in completely unique situations or new techniques developed to suit the situation. When you are doing something that has literally never, ever been done before things like this are common. Prior experience can only take you so far after that you are learning how to do the thing because you are the first person to ever do it. This is a far cry from installing a sink or rewiring a house which has been done thousands of times before and for which the ways in which it can fail are well known and can be avoided.
The people involved in the work are not just a few plumbers and electricians that were called up from the local yellow pages (or Pages Jaunes at CERN) but are either CERN employees or employers of contractors. My experience has been that while they are extremely "union" orientated (they are very particular about their breaks, starting/stopping work etc) they are also extremely professional to the point where they have come and shown be the right way to do something so it did not make their work look unprofessional!
...and yet the Run II of the Tevatron was similarly delayed and that was for a machine which was only upgraded. So having established that 5 weeks of holiday a year does not seem to affect the outcome perhaps "sensible" would be a better description.
The aim is still to go to 7TeV/beam this is only a temporary reduction in energy. In addition all the evidence so far points to a low mass higgs, not up at the hard ~1TeV/c2 limit where the energy is actually important. This is not unprecedented - the Tevatron which was supposed to be 1TeV/beam ran at 0.8 TeV for the first run and increased it to 0.96 TeV for the second run.
However, That being said it was never really the case that would would turn the machine on and the Higgs would magically pop out of the ether for all to see. The most likely scenario is a low mass Higgs which decays to b-quarks. Unfortunately the LHC will be EXTREMELY good at producing b quarks from known physic processes (there is even a entire experiment devoted to studying them - LHCb). The result is that a lot of hard, painstaking work will be needed before we can spot the b quarks from a Higgs from background "ordinary" b quarks. Of course there is still a chance that the Higgs might have enough mass to decay to two Z bosons which would be very easy to see early on but, if the Standard Model Higgs exists, the chance looks slim.
In contrast, the hardware of the Antikythera mechanism embodies a method of solution, so that it's user only needs to understand the problem and how to enter it into the computer, but not the detailed method of solution.
It depends entirely how you view it. The Antikythera machine is certainly a more advanced computer: the required actions for solving the problem are certainly less complex than a stick and some sand. However if you regard the stick and sand as a machine with a set of actions you must take the two are alike (you do not have to understand why you do them - you just do them): the only difference is the complexity of the instructions. Did you understand the complete reasoning behind the method of long division and multiplication when you first learnt it as school?
You do need to know more details about the method with the stick but you undoubtedly need to know a reasonable amount with the Antikythera machine as well. I see no way to define a defendable line between the Antikythera machine and a stick in the sand which would allow you to call one a computer and not the other. Both are computational tools - one is more advanced, simpler to operate but more restricted, than the other.
Build a small explosively pumped flux compression generator as a physics project, ensure the schools computers and powered down, and voila no cell phones. Educational AND effective.
What if you're a doctor or emergency worker on call?
If they are a doctor who is still attending school then I'd actually think it would be unethical NOT to block their cell phone!....and if they are there for a visit then they should not be on call. Doctors managed perfectly fine without cell phones before they were invented. they might be useful for a qiuck consult or to call in extra help if needed but the health system will not collapse without them.
I graduated before Columbine.....the nerd smacking the jock in the head with a metal chair repeatedly until he was down....The lunch monitor didn't even flinch.
Lessons learnt: you can get away with just about anything (e.g. beating someone around the head with a metal implement!!) if you think it is justified, if you are going to nick something make sure you leave the victim in no fit state to come after you and you can save money by sacking the "lunch monitors". All excellent preparation for entry into a civilized society.
Makes you wonder how Columbine could possibly have happened doesn't it?
If someone buys stolen property in good faith, never believing that it was stolen then the police inform them, they have absolutely no right to keep it
Well done for believing the RIAA's propaganda campaign. What Amazon did was provide unauthorized copies of a text. This is copyright infringement NOT stealing. You are completely correct if Amazon had stolen physical books from someone and then sold them because there would be a prior owner who would want them back. However Amazon illegally created copies of the book and sold those. Hence there is no prior owner to return them too. All the copyright owner can do is sue Amazon for damages.
....or maybe he is willing to stand up and fight for his rights, and the rights of others.
In that case he would be campaigning to have the law changed to make such actions illegal and not trying to sue Amazon. Even if he is successful all they will do is change the licensing agreement to allow them to do this without being sued in the future. If he is really trying to stand up for his (and others) rights this is a stupid way to go about it....its also a stupid way to try and get some money. The only people who will get rich from this are the lawyers.
When quoting the emission at a particular wavelength the convention is to use the differential spectrum (dN/dlambda). Technically this is the number of photons per unit time per wavelength but physical meaning is that it gives the measure of the number of photons per unit time at a particular wavelength which is why it is stated as such. See your A level physics course at school (UK) or 2nd year university physics course (US/Canada) for details.
It is treatable but not cureable. You are stuck with it for the rest of your life and it can cause complications later if it flares up when you are older and already fighting off another disease.
A stick and some sand is a mechanical device which can deterministically compute planetary data based on user input too. If your definition of a computer has such a low basis then it must also include basic writing implements which precede just about everything else. Why does the fact that it is free form exlcude it?
a form of computer that uses the continuously-changeable aspects of physical phenomena such as electrical,[1] mechanical, or hydraulic quantities to model the problem being solved.
Sounds exactly like a stick in sand is an analogue computer. You change the height of the sand (a physical phenomenom) to model your problem using a specific representation (letters and numbers). Perhaps you ought to have read it before posting?
It's not a translation, because the output has nothing to do with the source code; you cannot inspect the binary and translate it to the same source code.
Clearly the output must have something to do with the source code! Since when does a translation have to be a one-to-one, reversible mapping? A clear example is Babelfish. That is an obvious mechanical translation and yet it drops the gender from adjectives when translating from French to English so that both "Je suis grand" and "Je suis grande" map to "I am tall". When you ask it to translate back it chooses "Je suis grand".
This is the same as a compiler. Like English in the above example, machine code carries around less information and so some data is dropped when converting from, say C++. The machine code generated will depend on the compiler, just like a language translation will depend on the translator program you use (although for very simple examples there are not many options for difference). Seems exactly like a mechanical translation to me....
It is a very impressive find but if that counts as a computer then so does a stick and some sand which we know is far older and is probably what was used to design the device that was discovered.
Slashdot Mirror
One thing missing from the summary is that 3.5 TeV/beam is only (hoïpefully) a very temporary setup. The ramp up to 5TeV/beam, or 10 TeV centre of mass energy should be quick rapid if everything works. Going to the full 7 TeV will take longer though.
You left out 'yore.' I see that one about every other week.
That definition of that meaning of "computer" is correct
You have to allow for meanings to change with time. That might have been a common definition 50+ years ago when humans needed to do that sort of thing. However they have been replaced with machines that carry the same name. Stop someone in the highstreet and ask them what a computer is and I can just about guarentee you (unless I'm really unlucky and stop you!) that they will describe a computer as a machine. That doesn't mean that the original meaning is now wrong it just means that the most common meaning of the term has changed.
Sorry for quoting Wikipedia - I'm not particularly fond of it either - but it was top of the heap of a quick google search and is often quick and convenient.
It does seem like a bit more applied science to work out the kinks in this 'relatively new bit of technology' would be in order
I think the issue is more a lack of experience with the technology rather than kinks in it. There really was a huge amount of research, development and testing that went on behind the scenes. The problem is that superconducting magnets is a new technology for CERN - all previous accelerator magnets were "warm" - but superconductor technology was the only way to generate the field required within an affordable power budget.
The problem is that with superconducting you get a whole different series of issues that you need to be careful about. For example repairing the power bar connection between warm magnets is a delay of 1-2 days with superconducting technology it is 1-2 months (warm, fix, cool) so ensuring that the connections are good takes on a whole new level of importance! Things like this are what you get from experience - it is not so much a kink in the technology as simply that is behaves differently from the previous technology. Unfortunately deploying superconducting magnets on such a huge scale has never been done before - the closest would be the Tevatron which is ~10 times smaller and CERN did consult extensively with Fermilab. The scale is also part of the reason for the slow fix times - there are thousands of connections to be checked and fixed.
In contrast, the way that you want to use the term "analogue computer" shows there to be no logical distinction....
Actually, if you reread my original post, I do not want to use the term that way. My argument was in fact that you you accept the Antikythera machine as a computer you also have to accept a stick in the sand - basically reductio ad absurdum.
My definition of a computer would be limited to devices capable of following a programmed series of instructions by themselves. Indeed Wikipedia says that "The ability to store and execute lists of instructions called programs makes computers extremely versatile, distinguishing them from calculators." The Antikythera machine is a calculator, not a computer.
Exactly. Any money left after the lawyer's fees and expenses have been covered will get donated.
The primary aim of the upgrade is to increase luminosity, not energy (although in some respects the two are very similar). Increasing to 7TeV would not be an upgrade - it is intended as a fix that will be applied when we can have sufficient down time. For the moment it was deemed that it was better for the physics to get some data now, even if only at 10TeV.
If good elecrical connections sufficient to meet the needs of the LHC are within the realm of what is known by humans, then I'll cry "incompetence".
I can see your point of view but what you need to know is that these are vastly different from ordinary electrical connections. The "wire" consists of a superconducting metal core which is surrounded by copper. The copper is there so that in the case of a quench (which happens extremely rapidly since it is a phase change) the copper can carry the almost 10kA current for long enough that it can be dumped into a resistor which is the size of a small room which gets very hot. This design is needed because copper is not a superconductor but has a far lower resistivity than things which do superconduct (obviously only when they are not in their superconducting! phase!). The problems were caused by the splicing process between lengths of this "wire" during assembly which caused the layers to separate near the join.
Perhaps this sheds a little light onto why it is so complex? We are not talking about soldering two wires together we are talking about the safety backup system for a 27km long power bar that carries almost 10kA of current at a temperature of 2K (for reference outer space is warmer at 2.7K). Superconductivity itself was only discovered in 1911 and, as recently at 1962, people were winning Nobel prizes for making theoretical predictions about its behaviour at junctions.
So in reality this is a relatively new bit of technology that has never been applied on this scale before. Testing was done in advance - considerable testing - but a lot of the problems were either not visible in short scale tests or were only present when large sections were assembled. In addition, even if you do completely understand the physics involved in building something it is not always possible to predict all possible outcomes. Even today planes still crash for understandable, but unpredicted, reasons. Of course as we gain experience we learn how they fail and build them so they don't fail in the same way again but, practically speaking, you will never be able to predict all the ways that a complex bit of machinery can fail.
... best of what's still around. I've noticed a distinct decline in the quality of professional services in the last decade.
Unrelated. The LHC failures have all been caused by unforeseen consequences of standard techniques applied in completely unique situations or new techniques developed to suit the situation. When you are doing something that has literally never, ever been done before things like this are common. Prior experience can only take you so far after that you are learning how to do the thing because you are the first person to ever do it. This is a far cry from installing a sink or rewiring a house which has been done thousands of times before and for which the ways in which it can fail are well known and can be avoided.
The people involved in the work are not just a few plumbers and electricians that were called up from the local yellow pages (or Pages Jaunes at CERN) but are either CERN employees or employers of contractors. My experience has been that while they are extremely "union" orientated (they are very particular about their breaks, starting/stopping work etc) they are also extremely professional to the point where they have come and shown be the right way to do something so it did not make their work look unprofessional!
...and yet the Run II of the Tevatron was similarly delayed and that was for a machine which was only upgraded. So having established that 5 weeks of holiday a year does not seem to affect the outcome perhaps "sensible" would be a better description.
The aim is still to go to 7TeV/beam this is only a temporary reduction in energy. In addition all the evidence so far points to a low mass higgs, not up at the hard ~1TeV/c2 limit where the energy is actually important. This is not unprecedented - the Tevatron which was supposed to be 1TeV/beam ran at 0.8 TeV for the first run and increased it to 0.96 TeV for the second run.
However, That being said it was never really the case that would would turn the machine on and the Higgs would magically pop out of the ether for all to see. The most likely scenario is a low mass Higgs which decays to b-quarks. Unfortunately the LHC will be EXTREMELY good at producing b quarks from known physic processes (there is even a entire experiment devoted to studying them - LHCb). The result is that a lot of hard, painstaking work will be needed before we can spot the b quarks from a Higgs from background "ordinary" b quarks. Of course there is still a chance that the Higgs might have enough mass to decay to two Z bosons which would be very easy to see early on but, if the Standard Model Higgs exists, the chance looks slim.
In contrast, the hardware of the Antikythera mechanism embodies a method of solution, so that it's user only needs to understand the problem and how to enter it into the computer, but not the detailed method of solution.
It depends entirely how you view it. The Antikythera machine is certainly a more advanced computer: the required actions for solving the problem are certainly less complex than a stick and some sand. However if you regard the stick and sand as a machine with a set of actions you must take the two are alike (you do not have to understand why you do them - you just do them): the only difference is the complexity of the instructions. Did you understand the complete reasoning behind the method of long division and multiplication when you first learnt it as school?
You do need to know more details about the method with the stick but you undoubtedly need to know a reasonable amount with the Antikythera machine as well. I see no way to define a defendable line between the Antikythera machine and a stick in the sand which would allow you to call one a computer and not the other. Both are computational tools - one is more advanced, simpler to operate but more restricted, than the other.
Build a small explosively pumped flux compression generator as a physics project, ensure the schools computers and powered down, and voila no cell phones. Educational AND effective.
Then they could use the phone or go out of the building to make the call, or do it in the ambulance on the way to the hospital etc.
What if you're a doctor or emergency worker on call?
If they are a doctor who is still attending school then I'd actually think it would be unethical NOT to block their cell phone! ....and if they are there for a visit then they should not be on call. Doctors managed perfectly fine without cell phones before they were invented. they might be useful for a qiuck consult or to call in extra help if needed but the health system will not collapse without them.
I graduated before Columbine.....the nerd smacking the jock in the head with a metal chair repeatedly until he was down....The lunch monitor didn't even flinch.
Lessons learnt: you can get away with just about anything (e.g. beating someone around the head with a metal implement!!) if you think it is justified, if you are going to nick something make sure you leave the victim in no fit state to come after you and you can save money by sacking the "lunch monitors". All excellent preparation for entry into a civilized society.
Makes you wonder how Columbine could possibly have happened doesn't it?
If someone buys stolen property in good faith, never believing that it was stolen then the police inform them, they have absolutely no right to keep it
Well done for believing the RIAA's propaganda campaign. What Amazon did was provide unauthorized copies of a text. This is copyright infringement NOT stealing. You are completely correct if Amazon had stolen physical books from someone and then sold them because there would be a prior owner who would want them back. However Amazon illegally created copies of the book and sold those. Hence there is no prior owner to return them too. All the copyright owner can do is sue Amazon for damages.
And the kid's attorneys have already said their profits (if any) will go to charity.
Where, of course, "profit" means any money left after their costs, i.e. their extremely cheap and completely reasonable fees, have been paid.
....or maybe he is willing to stand up and fight for his rights, and the rights of others.
In that case he would be campaigning to have the law changed to make such actions illegal and not trying to sue Amazon. Even if he is successful all they will do is change the licensing agreement to allow them to do this without being sued in the future. If he is really trying to stand up for his (and others) rights this is a stupid way to go about it....its also a stupid way to try and get some money. The only people who will get rich from this are the lawyers.
When quoting the emission at a particular wavelength the convention is to use the differential spectrum (dN/dlambda). Technically this is the number of photons per unit time per wavelength but physical meaning is that it gives the measure of the number of photons per unit time at a particular wavelength which is why it is stated as such. See your A level physics course at school (UK) or 2nd year university physics course (US/Canada) for details.
Not really. Malaria is treatable.
It is treatable but not cureable. You are stuck with it for the rest of your life and it can cause complications later if it flares up when you are older and already fighting off another disease.
A stick and some sand is a mechanical device which can deterministically compute planetary data based on user input too. If your definition of a computer has such a low basis then it must also include basic writing implements which precede just about everything else. Why does the fact that it is free form exlcude it?
a form of computer that uses the continuously-changeable aspects of physical phenomena such as electrical,[1] mechanical, or hydraulic quantities to model the problem being solved.
Sounds exactly like a stick in sand is an analogue computer. You change the height of the sand (a physical phenomenom) to model your problem using a specific representation (letters and numbers). Perhaps you ought to have read it before posting?
It's not a translation, because the output has nothing to do with the source code; you cannot inspect the binary and translate it to the same source code.
Clearly the output must have something to do with the source code! Since when does a translation have to be a one-to-one, reversible mapping? A clear example is Babelfish. That is an obvious mechanical translation and yet it drops the gender from adjectives when translating from French to English so that both "Je suis grand" and "Je suis grande" map to "I am tall". When you ask it to translate back it chooses "Je suis grand".
This is the same as a compiler. Like English in the above example, machine code carries around less information and so some data is dropped when converting from, say C++. The machine code generated will depend on the compiler, just like a language translation will depend on the translator program you use (although for very simple examples there are not many options for difference). Seems exactly like a mechanical translation to me....
It is a very impressive find but if that counts as a computer then so does a stick and some sand which we know is far older and is probably what was used to design the device that was discovered.