Leapfrogs products are not about the most expensive polished electronics. They are about platforms that cost as little as possible to deliver compelling educational content for kids. Instead of whining, go play with one of these. I was blown away after 5 minutes and totally intrigued as to what the developers will come up with once they have got comfy with such a radically different computing paradigm. And to those that say, hey why the hell are you not teaching your own kids instead of buying them gadgets, I agree totally, I will be totally involoved in my kids eduction at home...however that will make us members of a very exclusive club, because that isn't what is happening in the vast majority of American homes for a host of reasons.
I not sure I'd agree with that statement, ARM was already a sure thing in mobile devices at that point but it takes a long time for that to bubble through the product pipeline and get into peoples hands. The ARM610 was interesting though, having a special MMU designed specificly to work well with Newtons OO OS. I've never bothered to look to see if any of those features still live in ARMs newer MMU designs. BTW the first silicon for the ARM610 that Apple received came from Plessey Semiconductors not VLSI though it was only by a matter of weeks. I know that because I worked on that design team.
I don't think the cost is significant at all compared with other equipment each content provider needs. Interestingly MPEG2->MPEG4 transcoders are indeed turning out to be a bit of a hot market at the moment for exactly the transitional reasons you highlighted. MPEG2 enocded material will still be with us for many years, though H.264 enocded signals may be placed into the ATSC mux relatively soon. Obviously transcoding MPEG2 SD to MPEG4 HD would as you correctly say be bogus.
I go with the original posters statement. Whilst this new satellite has a great deal of new technology it does not actually have any mpeg4 codec technology flying actually on it, the content is uplinked already compressed. BTW To complement the other bitstream information posted as replies, H.264/MPEG4 part10 will give quality equal to ATSC HD at about 7-8 MBps
Firstly GPS is unlikely to be ever turned off at a global level for civilian users. There are so many unanticipated benefits of it that it is now an intimate part of our every day lives. GPS for instance provides the clock synchronization for most modern communications systems, it guides automated machinery on farms, it is (recently) approved for aircraft navigation by the FAA in the US. There are so many more examples. The US DoD now persues a strategy of selective denial where GPS will be denied only locally in a war theatre to non-authorized military/gov users.
On another pro-euro point, the EU is gearing up to launch its own GALILEO system, which is very comprehensive and modern and will include special provisions for "life critical" applications where reliability is key.
My somewhat limited sailing knowledge suggests that in Northen Europe other vesels pose a much greater navigation threat than the coastline these days, especially in the area of the english channel. Frankly boats without, radar, LORAN and GPS etc have no place to be out in conditions that require instrument navigation, these technologies are not so expensive, especially weighed against the cost of a boat.
So saying all of that, keep the light houses, hell there historical value alone merits 400K euro's year.
Hard to answer without understanding how its used. Maybe he wants to browse with his latop anywhere on his land. Or maybe he has sophisticated and critical monitoring and control of his greenhouse operations run by this. Regardless, my first approach is try what is simple. I presume its reasonably flat and obsticle free land. A simple, adequately tall mast (http://www.hyperlinktech.com/web/antenna_height.p hp) near the centre of his land with a high gain omnidirectional antenna should give pretty impressive range. If his property is say 50 acres square then thats 0.5km on a side, so not really that far. The only question then is does that provide enough signal strength to be directly received by comodity embeded antennas or will each greenhouse need a directional antenna pointing at the omni on the mast. I'm not sure at what point the FCC would regulate mast height for an ISM band. In a few years if this is the low bandwidth control application I imgine it really is then it might be solved by a zigbee mesh solution at very low cost and maintainance.
Come one, and I quote " This is because research at Samsung showed that context switches and IPCs (inter-process communications) are faster under uClinux on processors that have virtually indexed caches and a TLB (translation lookaside buffer) without address space tags." That is exactly what this piece of "research" said wasn't true! And I say "research" because whilst I heartily approve of any engineer quantifying any theory, this is not rocket science, every embeded OS developer understands this point. Most embeded deivces use CPU's with MMU's, for example every cell phone that has an ARM7b running the call control stack would account for several hundered million examples.... Not to be down on what looks like a god product.
Obviously consumers want free calls, but thats of no interest to the cell companies. For many years however the wireless companies have struggled with a growing problem of network capacity, and wanting to offer a local wireless loop to compete with the wireline. There are many teams at big and small companies working on the problem. Essentially what they ideally would like to do is place a basestation (what they tend to call a picocell) in your home or buisness. It has very limited range, and that is deliberate, they only have so much RF spectrum so to increase capacity there must be more cells and they must therefore have less physical coverage area. Bear in mind that there is no reason why this is only for your benefit, it can offer service to anyone. The challenge is to make it "friendly" to the rest of the cell network. Traditional cells are planned in great detail and tested, they must offer no service gaps and conflicitingly must not range to far so that they don't interfere with other nearby cells. Thus their power, position, and frequency usage is carefully thought through. Now jump foward to the "home base station", it needs to be like a self install DSL modem, it gets delivered by mail or you buy it at a store and simply plug it into your broadband. Now it has to slef provision its frequency plan to enhance coverage rather than screw it up. It also needs to open a secure tunnel into the cell networks control and voice backhaul infrastructure presumably via your broadband connection which has some serious security concerns for the cell operator. End result they get you to install there infrastructure and provide the real estate.
I'm also in the SF bay area and my perspective is whilst a job advert generates any number of resumes that appear to be well qualified, at interview people I would consider hiring, especially for key development/engineering posts are few and far between. As someone working in small and competative tech companys our ability to surive and thrive is driven by the few people we can afford to employ, and thus we need the very best to have a chance. That might sound elitest, but its also common sense. I have hired more H1B workers than I can even count at this point, and everyone has been hired on merit alone. They have never been payed a dime less than there peers, and frankly the legal costs and complications have often cost a great deal in time and $$$. I frankly don't see that many "Americans" across the interview table, they are a small percentage of the applicants and they are no different in terms of the ratio of briliant vs average from other imigrant ethnic groups.
OK I smell FUD. The first two generations of IPOD used a 1230mAH battery (http://www.ipodbatteryfaq.com/files/sonybattery.p df) and the 3rd generation an 850mAH battery (http://www.ipodbatteryfaq.com/files/battery_docka ble.jpg). The latest IPOD still doesn't use manufacturing technology that is as advanced as the latest cell phone chip sets. You can be sure a convergence IPOD-Phone won't be much of a compromise. And on that topic I suggest anyone who thinks that Apple is so incredibly stupid that they don't imagine any of this should take a look at current Apple jobs....they are recruiting people with cell phone design smarts (and video....)...now what could that possibly mean????
OK I haven't thought this out too much so don't flame me too much if I missed something obvious but: Geostionary orbit is the distance from the earth that an object with angular velocity that keeps in stationary orbit over the same point on the earth is at equilibrium from centrepetal aceleration and gravational attraction. Assuming that statement is correct then to support the weight of the 'cable' and payload which are deeper into the gravity well of earth then the endpoint needs to be much farther out than a normal geostationary orbit so that the force generated by centrepetal acceleration minus gravitational attraction of the whole system still has a resulting force vector that tensions the cable.........does that make sense to anyone!?!?!
You will need new equipment, and please note that Galileo differs from GPS in that it offers a 'for money' commercial high accuracy signal also. But don't get upset and just blame the Europeans because the US is also about to change the GPS system, adding a new second civilan signal to the L2 band where there was formerly only only a secure (p-code) signal and also adding the new L5 frequency. In short if you want to work with all the new signals available by the end of the decade then you will be buying new hardware.
You know the one thing I feel I lack when I read (with great interest) peoples concerns about electronic polling is just how bad were previous low tech systems. We all remember the Florida presidential debicle, but I wonder if there is somewhere much more info collected across many elections and systems that gives us something to compare with.
Please do......I would love to hear what any stray charge from it does to your Ipod electronics who's ESD protection circuitry would be utterly overwhelmed by this kind of voltage
Intestingly enough I recall when I bought my IPOD I chuckled at Apples custom laser engraving service thinking hell why not spend even more $$$ on this thing!! Suddenly I see that indelable name and address etc burned into the steel case in a whole different light, making these things at least a little trackable should they ever be recovered by police or offered for sale on EBay etc...
It will be interesting to see how the outages look geographicly. Just reading the description of the layout of the tunnel system and the limted egress points I would expect long distance trunks to be far more affected than local exchanges as it seems BT uses it as a ways of bringing in new backhauls to the city centre without civil works. In many ways I was expecting the affects on the city centre telephone service to be quite limited. It will also be interesting to see if the data outage is a reflection on peoples focus on logical redundancy at the planning stage rather than physical redundancy (all fibres in the same hole...)
Manchester has a population comfortably in excess of 1 Million people and a large buisness centre. 100K dead telephones represents only a small but significant amount of the city.
In there marketing movie the talk about about compressed air as there means of storing energy....then they say that after the aircraft has floated aloft as lighter than air vehicle they use the stored compressed air to drive turbines to compress air to increase the mass of the vehicle so it can become heavier than air and commence to glide (downwards) as a heavier than air vehicle....Now lets get this straight they can use X quantity of compressed air to compress Y quantity of air where the mass of Y is presumably much greater than X.....Err perpetual energy machine anyone???? Surely what they really do is compress the helium bags reducing there volume and hence allowing the air they would displace to reenter the craft?? ION
The right idea, but the wrong technology, GPS is virtually useless indoors, and the evolution A-GPS from Qualcomm works by supplying aiding information based on approximate knowledge of your whereabouts. Thats fine when a cell tower takes the call and can pinpoint your location to a couple of hundred metres but no use for VoIP because you litterally could be anywhere on the planet and your IP address gives no reliable consistant fix on your location. See www.rosum.com if want to a more viable solution.
Lets see 20 years ago this is what I remember being taught in undergrad CS (Though I'd been an assembly programmer for 7 years already then): 1st Year: Pascal, PDP-11 assembly 2nd Year: 68000 Assembly, 3rd Year: Ada, Eiffel, Modulo-2, Smalltalk, VHDL, EDIF
Interestingly despite using mostly UNIX hardware C was left to be self taught, though it could be used in place of Pascal. The point really was use appropriate tools for appropriate jobs: Thus we learnt OS/Kernel basics in assembler, likewise IRQ concepts and HAL stuff. For algorithmic stuff we worked in a mainstream strongly typed high level language. For trends yet to propagate to the commerical world (some doomed ones!) we learned in exotic academic languages. Assembly is as valid today as it was in the 1950's, its just that its most appropriate for only some tasks that a minority of programmers do: Deeply embeded microcontroller SW, OS kernel work (take a look in our beloved linux kernel code), hardware bootstrap, hard real time etc etc etc
Its not an issue of Intel having not "designed" for 10GHz operation, this idea that clock speeds can be dramticaly increased at drasticly reduced temeratures fails on two simple counts: 1) Material science. The transistor operating speed and signal propagations just don't scale so dramaticly due to temperature. If I run an analysis of a chip at say a temperature of 0C and again at 70C (typical operating temperature range for a commercial use chip) then I will see something like a difference of 20% in the max clock speed. 2) Non-Linearity of operation. Different signal paths will change there propagation delays differently for a given change in temperature. Eventually the circuit will just not work correctly, typically for low temperature due to a so called race or short path error. If the (die) junction temperature was sucessfully reduced to the kind of temperatures that spawned this discussion then the chances of correct operation would frankly be very slim. I would guess that even in this case there is still a very steep temperature gradient within the package and that the die is operating at a far higher temperature than people understand.
Its hard in such limited space to explain just to what lengths (we) chip designers go to model the operation of our designs, but suffice to say ultimately we model the propagation of signal transitions across a device to a degree of detail that would utterly amaze (and baffle) most people who consider them selves very tech savy. Hand on heart, I can honnestly admit that even though I have designed more chips than I have fingers and toes, that I would be fairly clueless at this point to explain the physics that the software tools that I use model. The crux of chip design has always been management of overwhelming complexity, and computation and design tools never get any closer to overtaking the designs we work on as ever year we hit the next set of atomic physical effects with a chip 4x more complex than last years. Put simply we don't set out to design chips that work under conditions we don't care about, we do every thing we can to simplfy our job and that includes(In Intels case) deciding a very precise operating voltage and temperature range for the correct operation of there microprocessors otherwise the design task would become intractably complex. As some other posters have noted the true panacea of such low temperatures is the possibilty of the use of super conducting materials, however at least for IC's that still belongs in the realm of esoteric research.
It is a concern. NAND flash which is generally the technology that leads the density curve needs special algorithms called 'wear leveling' in the device driver/file system to try to prevent 'hot spots' that cause bits to fail prematurely. the lower density NOR flash devices don't have the problem and tend to be used in application where this is expected to be a problem. That being said NAND flash has been used for many years in this type of application so I would describe it as an already solved problem. Given that there is no order of magnitude storage advantage for these miniturized HDD's over flash I can't imagine that flash would not still dominate in both the existing and new applciations it is touted for. I know I would think twice about have an HDD in a camera etc, I always treat my IPOD with extra care knowing what mechanical magic lurks within....
Slashdot Mirror
Leapfrogs products are not about the most expensive polished electronics. They are about platforms that cost as little as possible to deliver compelling educational content for kids. Instead of whining, go play with one of these. I was blown away after 5 minutes and totally intrigued as to what the developers will come up with once they have got comfy with such a radically different computing paradigm.
And to those that say, hey why the hell are you not teaching your own kids instead of buying them gadgets, I agree totally, I will be totally involoved in my kids eduction at home...however that will make us members of a very exclusive club, because that isn't what is happening in the vast majority of American homes for a host of reasons.
I not sure I'd agree with that statement, ARM was already a sure thing in mobile devices at that point but it takes a long time for that to bubble through the product pipeline and get into peoples hands. The ARM610 was interesting though, having a special MMU designed specificly to work well with Newtons OO OS. I've never bothered to look to see if any of those features still live in ARMs newer MMU designs. BTW the first silicon for the ARM610 that Apple received came from Plessey Semiconductors not VLSI though it was only by a matter of weeks. I know that because I worked on that design team.
I don't think the cost is significant at all compared with other equipment each content provider needs. Interestingly MPEG2->MPEG4 transcoders are indeed turning out to be a bit of a hot market at the moment for exactly the transitional reasons you highlighted. MPEG2 enocded material will still be with us for many years, though H.264 enocded signals may be placed into the ATSC mux relatively soon. Obviously transcoding MPEG2 SD to MPEG4 HD would as you correctly say be bogus.
I go with the original posters statement. Whilst this new satellite has a great deal of new technology it does not actually have any mpeg4 codec technology flying actually on it, the content is uplinked already compressed.
BTW To complement the other bitstream information posted as replies, H.264/MPEG4 part10 will give quality equal to ATSC HD at about 7-8 MBps
Firstly GPS is unlikely to be ever turned off at a global level for civilian users. There are so many unanticipated benefits of it that it is now an intimate part of our every day lives. GPS for instance provides the clock synchronization for most modern communications systems, it guides automated machinery on farms, it is (recently) approved for aircraft navigation by the FAA in the US. There are so many more examples. The US DoD now persues a strategy of selective denial where GPS will be denied only locally in a war theatre to non-authorized military/gov users.
On another pro-euro point, the EU is gearing up to launch its own GALILEO system, which is very comprehensive and modern and will include special provisions for "life critical" applications where reliability is key.
My somewhat limited sailing knowledge suggests that in Northen Europe other vesels pose a much greater navigation threat than the coastline these days, especially in the area of the english channel. Frankly boats without, radar, LORAN and GPS etc have no place to be out in conditions that require instrument navigation, these technologies are not so expensive, especially weighed against the cost of a boat.
So saying all of that, keep the light houses, hell there historical value alone merits 400K euro's year.
Hard to answer without understanding how its used. Maybe he wants to browse with his latop anywhere on his land. Or maybe he has sophisticated and critical monitoring and control of his greenhouse operations run by this.p hp) near the centre of his land with a high gain omnidirectional antenna should give pretty impressive range. If his property is say 50 acres square then thats 0.5km on a side, so not really that far. The only question then is does that provide enough signal strength to be directly received by comodity embeded antennas or will each greenhouse need a directional antenna pointing at the omni on the mast. I'm not sure at what point the FCC would regulate mast height for an ISM band.
Regardless, my first approach is try what is simple. I presume its reasonably flat and obsticle free land. A simple, adequately tall mast (http://www.hyperlinktech.com/web/antenna_height.
In a few years if this is the low bandwidth control application I imgine it really is then it might be solved by a zigbee mesh solution at very low cost and maintainance.
This is a serious and valuable post.
Come one, and I quote " This is because research at Samsung showed that context switches and IPCs (inter-process communications) are faster under uClinux on processors that have virtually indexed caches and a TLB (translation lookaside buffer) without address space tags."
That is exactly what this piece of "research" said wasn't true!
And I say "research" because whilst I heartily approve of any engineer quantifying any theory, this is not rocket science, every embeded OS developer understands this point.
Most embeded deivces use CPU's with MMU's, for example every cell phone that has an ARM7b running the call control stack would account for several hundered million examples....
Not to be down on what looks like a god product.
Obviously consumers want free calls, but thats of no interest to the cell companies. For many years however the wireless companies have struggled with a growing problem of network capacity, and wanting to offer a local wireless loop to compete with the wireline. There are many teams at big and small companies working on the problem. Essentially what they ideally would like to do is place a basestation (what they tend to call a picocell) in your home or buisness. It has very limited range, and that is deliberate, they only have so much RF spectrum so to increase capacity there must be more cells and they must therefore have less physical coverage area. Bear in mind that there is no reason why this is only for your benefit, it can offer service to anyone.
The challenge is to make it "friendly" to the rest of the cell network. Traditional cells are planned in great detail and tested, they must offer no service gaps and conflicitingly must not range to far so that they don't interfere with other nearby cells. Thus their power, position, and frequency usage is carefully thought through. Now jump foward to the "home base station", it needs to be like a self install DSL modem, it gets delivered by mail or you buy it at a store and simply plug it into your broadband. Now it has to slef provision its frequency plan to enhance coverage rather than screw it up. It also needs to open a secure tunnel into the cell networks control and voice backhaul infrastructure presumably via your broadband connection which has some serious security concerns for the cell operator. End result they get you to install there infrastructure and provide the real estate.
I'm also in the SF bay area and my perspective is whilst a job advert generates any number of resumes that appear to be well qualified, at interview people I would consider hiring, especially for key development/engineering posts are few and far between. As someone working in small and competative tech companys our ability to surive and thrive is driven by the few people we can afford to employ, and thus we need the very best to have a chance. That might sound elitest, but its also common sense. I have hired more H1B workers than I can even count at this point, and everyone has been hired on merit alone. They have never been payed a dime less than there peers, and frankly the legal costs and complications have often cost a great deal in time and $$$. I frankly don't see that many "Americans" across the interview table, they are a small percentage of the applicants and they are no different in terms of the ratio of briliant vs average from other imigrant ethnic groups.
OK I smell FUD. The first two generations of IPOD used a 1230mAH battery (http://www.ipodbatteryfaq.com/files/sonybattery.p df) and the 3rd generation an 850mAH battery (http://www.ipodbatteryfaq.com/files/battery_docka ble.jpg). The latest IPOD still doesn't use manufacturing technology that is as advanced as the latest cell phone chip sets. You can be sure a convergence IPOD-Phone won't be much of a compromise. And on that topic I suggest anyone who thinks that Apple is so incredibly stupid that they don't imagine any of this should take a look at current Apple jobs ....they are recruiting people with cell phone design smarts (and video....)...now what could that possibly mean????
OK I haven't thought this out too much so don't flame me too much if I missed something obvious but: ....does that make sense to anyone!?!?!
Geostionary orbit is the distance from the earth that an object with angular velocity that keeps in stationary orbit over the same point on the earth is at equilibrium from centrepetal aceleration and gravational attraction.
Assuming that statement is correct then to support the weight of the 'cable' and payload which are deeper into the gravity well of earth then the endpoint needs to be much farther out than a normal geostationary orbit so that the force generated by centrepetal acceleration minus gravitational attraction of the whole system still has a resulting force vector that tensions the cable.....
You will need new equipment, and please note that Galileo differs from GPS in that it offers a 'for money' commercial high accuracy signal also. But don't get upset and just blame the Europeans because the US is also about to change the GPS system, adding a new second civilan signal to the L2 band where there was formerly only only a secure (p-code) signal and also adding the new L5 frequency. In short if you want to work with all the new signals available by the end of the decade then you will be buying new hardware.
Amazingly all the porno spam I got on my yahoo account today actually had the SEUALLY-EXPLICT warning...spammers are taking this seriously it seems.
You know the one thing I feel I lack when I read (with great interest) peoples concerns about electronic polling is just how bad were previous low tech systems. We all remember the Florida presidential debicle, but I wonder if there is somewhere much more info collected across many elections and systems that gives us something to compare with.
Please do ......I would love to hear what any stray charge from it does to your Ipod electronics who's ESD protection circuitry would be utterly overwhelmed by this kind of voltage
Intestingly enough I recall when I bought my IPOD I chuckled at Apples custom laser engraving service thinking hell why not spend even more $$$ on this thing!! Suddenly I see that indelable name and address etc burned into the steel case in a whole different light, making these things at least a little trackable should they ever be recovered by police or offered for sale on EBay etc...
It will be interesting to see how the outages look geographicly. Just reading the description of the layout of the tunnel system and the limted egress points I would expect long distance trunks to be far more affected than local exchanges as it seems BT uses it as a ways of bringing in new backhauls to the city centre without civil works. In many ways I was expecting the affects on the city centre telephone service to be quite limited. It will also be interesting to see if the data outage is a reflection on peoples focus on logical redundancy at the planning stage rather than physical redundancy (all fibres in the same hole...)
Manchester has a population comfortably in excess of 1 Million people and a large buisness centre. 100K dead telephones represents only a small but significant amount of the city.
In there marketing movie the talk about about compressed air as there means of storing energy....then they say that after the aircraft has floated aloft as lighter than air vehicle they use the stored compressed air to drive turbines to compress air to increase the mass of the vehicle so it can become heavier than air and commence to glide (downwards) as a heavier than air vehicle....Now lets get this straight they can use X quantity of compressed air to compress Y quantity of air where the mass of Y is presumably much greater than X.....Err perpetual energy machine anyone???? Surely what they really do is compress the helium bags reducing there volume and hence allowing the air they would displace to reenter the craft??
ION
The right idea, but the wrong technology, GPS is virtually useless indoors, and the evolution A-GPS from Qualcomm works by supplying aiding information based on approximate knowledge of your whereabouts. Thats fine when a cell tower takes the call and can pinpoint your location to a couple of hundred metres but no use for VoIP because you litterally could be anywhere on the planet and your IP address gives no reliable consistant fix on your location. See www.rosum.com if want to a more viable solution.
Lets see 20 years ago this is what I remember being taught in undergrad CS (Though I'd been an assembly programmer for 7 years already then):
1st Year: Pascal, PDP-11 assembly
2nd Year: 68000 Assembly,
3rd Year: Ada, Eiffel, Modulo-2, Smalltalk, VHDL, EDIF
Interestingly despite using mostly UNIX hardware C was left to be self taught, though it could be used in place of Pascal.
The point really was use appropriate tools for appropriate jobs: Thus we learnt OS/Kernel basics in assembler, likewise IRQ concepts and HAL stuff. For algorithmic stuff we worked in a mainstream strongly typed high level language. For trends yet to propagate to the commerical world (some doomed ones!) we learned in exotic academic languages.
Assembly is as valid today as it was in the 1950's, its just that its most appropriate for only some tasks that a minority of programmers do: Deeply embeded microcontroller SW, OS kernel work (take a look in our beloved linux kernel code), hardware bootstrap, hard real time etc etc etc
Its not an issue of Intel having not "designed" for 10GHz operation, this idea that clock speeds can be dramticaly increased at drasticly reduced temeratures fails on two simple counts:
1) Material science. The transistor operating speed and signal propagations just don't scale so dramaticly due to temperature. If I run an analysis of a chip at say a temperature of 0C and again at 70C (typical operating temperature range for a commercial use chip) then I will see something like a difference of 20% in the max clock speed.
2) Non-Linearity of operation. Different signal paths will change there propagation delays differently for a given change in temperature. Eventually the circuit will just not work correctly, typically for low temperature due to a so called race or short path error. If the (die) junction temperature was sucessfully reduced to the kind of temperatures that spawned this discussion then the chances of correct operation would frankly be very slim. I would guess that even in this case there is still a very steep temperature gradient within the package and that the die is operating at a far higher temperature than people understand.
Its hard in such limited space to explain just to what lengths (we) chip designers go to model the operation of our designs, but suffice to say ultimately we model the propagation of signal transitions across a device to a degree of detail that would utterly amaze (and baffle) most people who consider them selves very tech savy. Hand on heart, I can honnestly admit that even though I have designed more chips than I have fingers and toes, that I would be fairly clueless at this point to explain the physics that the software tools that I use model. The crux of chip design has always been management of overwhelming complexity, and computation and design tools never get any closer to overtaking the designs we work on as ever year we hit the next set of atomic physical effects with a chip 4x more complex than last years. Put simply we don't set out to design chips that work under conditions we don't care about, we do every thing we can to simplfy our job and that includes(In Intels case) deciding a very precise operating voltage and temperature range for the correct operation of there microprocessors otherwise the design task would become intractably complex.
As some other posters have noted the true panacea of such low temperatures is the possibilty of the use of super conducting materials, however at least for IC's that still belongs in the realm of esoteric research.
It is a concern. NAND flash which is generally the technology that leads the density curve needs special algorithms called 'wear leveling' in the device driver/file system to try to prevent 'hot spots' that cause bits to fail prematurely. the lower density NOR flash devices don't have the problem and tend to be used in application where this is expected to be a problem. That being said NAND flash has been used for many years in this type of application so I would describe it as an already solved problem. Given that there is no order of magnitude storage advantage for these miniturized HDD's over flash I can't imagine that flash would not still dominate in both the existing and new applciations it is touted for.
I know I would think twice about have an HDD in a camera etc, I always treat my IPOD with extra care knowing what mechanical magic lurks within....
OK, I should have read that link first , so there is also some last mile distribution in the actual conductors also in UK it seems.