"Why would they want a dual-core laptop? I thought the idea was to not burn the juice?"
A good design might be a multi-core notebook that can power cores on and off as required to save power. "Dual" is just the start in a few years we may see 4, 8 or 16 core machines and a good way to manage power use would be to power the cores up and down as required. Even on desktop machines people should care about power. Running my dual CPU Xeon 24/7 likely costs $25/month.
Now that we are hitting a Ghz limit we may see Moore's law aplied to cores with a doubling every couple years. I mean, 12Ghz is impractical and 100Ghz is likely to never happen but 16-core chips will be
available in time. And Yes, there ARE uses for 16 CPUs, just look at how many 16-cpu servers are sold today and people are wiling to pay $250,000 for them so they must be pretty usfull
Don't worry. The TPM chip does NOT cripple the hardware. I'm typing this on an HP "xw8200" that has a TPM chip on the motherboard. The hardware acts no different TPM is just something that an operating system can choose to use or chose to ignore. I have Solaris 10 install in this box and it ignores the TPM stuff, Linuix would ignore it too as would Darwin or any current version of Windows.
What TPM does is this: The OS asks the chip to "measure" the hardware. It does this and returns a cryptographic hash. THe OS can compare this to a list of good hashes and then decide what to do.
TPM can also do things like "measure" the software on a disk and the disk itself. This could even be a Good Thing on an Open Source operatring system For example Linux could use TPM as a way to make certain the system is not been compromised. It would be a powerfull security measure. Apple will use the TPM to insure that yu did not swap out the whole machine for a Dell, Gateway or whatever.
ANy operating system that wants to be secure has to depend on having a secure and trusted "security kernel" that is "tamper proof" and known to work and be well tested. By "tamper proof" I mean you can __prove__ that it is not been messed with.
Right now can you __prove__ that your Linux machine with it's one year "up time" has not had one byte of code changed in it's running kernel. You can't, OK you could write a checksun routine that runs periodically, but how do you know the checksum routine was not modified? You can't, not without locking the checksum routine into the hardware and that is what TPM does.
If Apple does this right I'll be happy to know that Mac OS X uses TPM.
You are only half right. Those missions would not have been saved by the crew but they would have been saved by the design and flight procedures used for crewed missions. With a crewed mission the entire spacecraft is filled with more redundancy and there is a larger full time suport staff on the ground and many more reviews and testing. It is actually cost efective to loose unmanned spacecraft. Putting people aboard rises the cost more than ten times so it is cheaper to loose half the unmanned misions then to build and fly them as if they were manned where a only a 2% loss is acceptable.
You said "its slower than the 2.2Ghz dual-core Athlon X2 that's sitting next to it"
Apple is not moving to AMD they are moving to Intel.
Now compare your 2+ Ghz G5 to a Pentium. The G5 may win. I think the move to Intel
was done because of notebooks. G4s are not fast enouh and the G5 would requre
the user to cary a backpack full of batteries
I really wish Apple had gone to AMD. I want an Apple version of this
http://www.sun.com/desktop/workstation/w2100z/spec s.jsp
or maybe even this
http://www.sun.com/servers/entry/v40z/specs.jsp
Note that Sun calls an eight-core Opteron box an "entry level" system.
I'd love to see Apple softwar on AMD based Sun hardware. Sun has actualy
gotten it's prices in-line. thier dual Operon box starts at $2,200 not
far from Apple's Dual G5.
(I'm wrtin this on a dual Xeon box made by HP that runs Solaris 10. Interestingly
the motherbord has the TM chip on it although Solaris ignores it. The chip actualy
_measures_ the system and then expresses the result with a crytpographic hash.
Part of that measuremet is the software. It would be darn hard to defeat
"people used to see no way that a plane could possibly go faster than sound" No, __some__ people thought that many didn't. but the big difference is that the "sound barier" was jst an enginerring problem. Even the people who thought it could not be broken know it was possable in theory. For example a rifle bulet fired in the early 1900's was supersonic and propeller tips of 1930's vintage airplanes could go super sonic. (ever hear that loud buzzing soulnd in old diver bombings, it's the sonic booms from the prop
tips) So the sonic barier doubters in the 1940's where saying that you just could not build a powerfull enough engine.
Light is an entirely differet _kind_ of limit. There is very solid theoreticalreasons why it is a hard limit.
In Azamov's (SP??) book "I Robot" there is a story about the people who final did breakthe light speed limit. Turned out from the view point of our conentional physics objects breaking the limit appear to cease to exist. They simply leave the universe. That may be the only way as the rules of the observable universe don't alow it so you go someplace else.
I'm typing this on a system with two 3.6Ghz Xeons with Hyperthreading enabled. The system uses two 300GB ltra-320 scsi disks set up in a mirror and has 4GB RAM installed.
When I run Nescape the performance is about as good as on my other system which is a 2.4Ghy P4 with 1GB RAM and one SATA drive.
However the Dual Xeon system runs my DBMS querries blazingly faster, __much__ faster then the P4 based system. Many DBMSes work like Apache and "fork" a new server process for each client, so when 12 process each connect to the DBMS I see 12 copies of the DBMS server software running DBMSes tend to run in tight loops where the instructions stay cached in the CPU's L2 cache
Solaris is smart enough to "know" which CPUs share on-chip cache and that all CPUs are not
"symetric" and so can schedule process to stay within a group of "close" CPUs. At boot time
the system looks around and builds a hiarchical model of the machine's configuration, noting
which PCI busses, CPU and memory are connected to what. On the larger systems there are
cards wth four CPU and up to 4GB RAM and some PCI buses all on one card and then
the cards are connected by backplane So when you read about these new CPUs don't think
about a low-end PC runnig Windows These could be intended to go into a 20-way mutiprossor
runnig Solaris (or Mac OS X) You should figure that in 10 years the system on your desk will use techniques like today's high end systems. What Sun's Sunfire does at the board level (four CPUs, Busses and RAM) wil be done at the chip level and you will see four core chips and then 16 core chips andthen computersbuilt with multiple 16-core chips designed like today's "starfire".
What would yo use such a system for? How ablut controlling a robot that can walk up a flight of stairs and responce to voice commands and identify objects with a vision system all at once.
Another aplication is video rendering, lots of data t process but the _code_ stays cached ad rentering like DBMS and web servering to paralizable. In fact the video render problem is the prime example of what to use a room full of CPUs for. And just wait 'till you buy a High Def Video camera. You will want one of those "quad core" power macs
There is a test facility down the street from here built by "TRW" (now part of Northrop Grummun) The test facility has the ability to reproduce the acustic signature of a large space lift boster, like an Atlas or the Space Shuttle. "Loud is not quite the word. A better word is "Destructive" The sound of the launch can literally rip stuff apart if it is not built right and hence the reason for the test chamber. The chamber is large enough to hold a typical satilite payload, which can be about the size of a stadard shipping container.
So, my point is that there are good reasons to build powerfull subwoffers other thn reproduction of music. One other use was in a facility I vistited in Florida that housed several Abrams A1 tank simulaters. The sound was realistic (as judged by some US Army tank core peole) I would just say it was "way loud".
Aple does this now. "Core Image" is built into the OS and all "correctly written" applications that need to do graphic use Core Image. Core Image wil use the GPU if one is available. This is a very good idea but the hardest part of getting this to work on a non-Apple platform will be standarizing the API so that we can use any GPU. OK X11 did this fr displays on UNIX ad we have OpenGL for 3D graphic so we can hope something will happen an API for GPU based image transfomation. The biggest use for this wil not be just simple transcoding but editing and dispay programs for still and moving images Think "gimp" and "cinelerra".
Solaris has "mapmalloc" which is mmap based malloc and it has "watchmalloc" which uses something like a breakpoint that you can have watch for reads and writes to the "wrong" places. Basically two independent malloc replacements. With watchmalloc being the more sophisticated of the two. These will find errors in stuff that has worked for years. Solaris man pages for mapmalloc and watchmalloc are on-line for anyone who wants them (see google)
Solars now is Open Source too. From my experiance using it and linux daily both thier intial releases (I remember Linix 0.x and I went into a Sun office see Solaris before it hit the street) Opinion: The Solaris kernel and related utilities is the most adavnace there is on the X86 patform. It is Very stable and will servive things like pulling a CPU looseout of a socket and will scale to over 100 processors. However it is the worst *NIX desktop OS on the plannet. I should know it use it as my desktop every day all day
"In some social-study class or other where it can be taught along side of Astrology, Divination, tea-leaf reading and the theory of the Abominable Snowman. Just not in science class."
No, that's a bit harsh. ID could be taught in a class that covers comparitive religion. For example they could also cover "mainstream" Christian teachings (Roman Catholic Church and a couple major Protestent sects as well) then they could cover maybe Hindu and American Indian creation stories too. It could actually be an interesting and usfull class there they show how each society makes up a creation story to fit thier culture. In fact this could even be taught in science. Yes real science with statistics and math and all that stuff.
The root of this conflict is that for many peope science IS faith. They haven't a clue. They think science is just a collection of always changing facts that one most memorize and accept on faith.
I always like those arguments that basically say "THEY could ot have done X because I can't figure out how to do X" The fact that some guy could not figure out how to burn wood with a mirror does not mean that one of the smartest guys in all of history could not have. In later centurys it was common to build solar furnaces Here is a link to the largest modern solor reflector
http://www.imp.cnrs.fr/foursol/1000_en.shtml
FOr non-triveal software it would be _very_ hard to identify who was responcible fo a bug. Most software is a group effort. You have someone who specifies what it is to do, some who design it but don't write the code and others who do write code but don't get to do the overall design and then you have the people who test and document it. When something fails you have to look at the process that created it as having failed.
A perfect example of this is the MS Windows OS. The root causes of it's problem are conceptual. No one person is to blame the system just eveloved from the non-networked, single user, single task OS, into
a networked, one user at a time, multitaking OS. When you try to carry backward compatabilty through such an evolution you have a mess.
I read a lot of things about sodftware managment from people who have not written a line of code in 20+ year or even ever. they sem to loose touch with reality
It should be possable to destroy an RFID device with a very powerfull _pulse_ of RF energy. People who like to use shreaders could buy a device that frys the RFID chips. I think these "RFID shreders" could be low priced too as the _average_ power emitted would be low
A two inch diagonal screen just won't cut it with most people So how to make it work?
They make "ear buds" so why not "eye buds"? Ok bad idea. But "iGlasses" would work. A little projector makes a head up display. It appears to the viewer as a large screen about a meter away.
It's bad enough now, but just wait 'till you start seeing people driving on the freeways with these "iGlasses". I imagine the porn industry woud find a use for this too.
Sun's "ZFS" applies error detection and correction to all data. ZFS is a file system that also includes a disk manager that provide RAID and volume concatination. It does it all in a single layer.
I don't know if standard PC's are fast enough yet that checksumming all the data as it goes to and from the disks would work. But on Sun server hardware the extra checksume is triveal ZFS puts 64 bit checksums on all the data and the metadata
ZFS is also Open Source so someday we might, maybe see a port to some other OS
When you install the machine and periodically thereafter you would "null" it. That means you adjust the needle to read zero when there is nothing in the detection chamber.
I used an oxigen sensor resently. The first step is to expose it to just plain air and adjust it to read 21% (air is about 21% oxigen, 79% nitrogen) After doing this it can detect very small amounts
In both cases you have to tell the machine "This is the normal background." After this the machine detects changes.
If these really will be given away free to anyone who needs one they will be _quite_ hard to sell. Kind of like if I tried to sell you a bag full of air. At first these might be sold but after millins of these things are dumpped into all the schools thier street price will be nearly zero.
This is the only way to eliminate drug dealers too. If you could get all the pot or speed or you wanted for free at any government building (Jst walk into a post office and help yourself) the people seling the stuff on the street would go broke. Same with PCs, if the schools had stacks of the things there would be no black market.
That said, if these Really do cost $100 to make why doesn't he ellthen here in the developed worl for $200 each. I'd be happy to buy two at that price which would pay for two kids in Africa.
There is also plenty of uses for a $100 PC here in the US. For example my bother is a bit of a computer geek but lives in an "assisted living" facility (aka "old folks home") because he i9s disabled. He has been setting up and giving away computers to some of the people there who are stuck in bed. These guys couldn't afford a computer but it sure does expand thier "world" the web beats 24x7 TV. e-mail and decussion forums keep the brain from turnning to mush. At $100 a pop I could afford to give away a few computers an likley would.
Do you know the formula that relates transmit power, distance, and bandwidth?
Of course it is _possable_ to build such a network. My point was that it has not been done
Your write "assuming you can contact it, and it can contact you." It depends on the data rates. You can have a 10,000 watt tansmitter on Earth but the power available to a rover on mars may be only a few watts. The links tend to be hugly asymetric. A small Mars rover can send on order of a few hundred bits per second directly to Earth but can send 10X faster to an orbiter that is only a few hundred kilometers overhead.
A moveable microwave antenna is not the best setup for switching packets. The antennsa slew at very slow rates and of course when it is pointing at one location it can't pont in others. This is not to say general purposes packet switches can not be built. Look at this www.losangeles.af.mil/SMC/MC/Tsat.htm TSAT will be a switch but it will conect users on the ground not in space. Maybe years from now someone will built a TSAT that looks upward? but I doubt they will add TSAT-like capability to everythingthat is launched.
You've been reading to much science fiction. Yes there are some cases where spacecraft use inderect means of communication through a relay but this is not done ad-hoc using some general purpose capability built into every spacecraft. In every case wherwe relay is used the capabilty is plaanned from the beginning.
The idea of selecting some random spacecraft to use as a relay to soe other random spacecraft just can't work. The orbiters currently on mars were design specifically to relay.
Closer to Earth TDRSS acts as a relay between low Earth orbit and the ground. Notice (1) that TDRSS is the relay, thaey are NOT sending data between randon spacecraft and (2) the data are passed only between LEO and the ground, not through out the solar system or even to geosync. orbit.
http://msp.gsfc.nasa.gov/tdrss/oview.html
They have sold 21 million of these. Lets say Apple is totally out of control and went nuts and spent one hundred million on design. so it's like five bucks each on enginerring. But I'll bet much of the design work for the internal software andother parts is the same as in the other iPods and of course iTunes is not new for the nano. I'd bet between four anfeight bucks per unit now and it goes downwith each unit sold.
Other costs are for things like warenty and technical support
You said " The US is financially overstretched as it is." Let's do the math. NASA says it will cost about 100B over 13 years to go to the moon. there are 300M people in the US. If each person pays 50 cents per week then NASA gets enough to fund the program.
Yes I know, some peole will get hit up to $10 per week and other get off with paying zero bt it averages to 50 cents per person per week.
"why no filesystems seem to support copy-on-write semantics"
Look at Sun's "ZFS". It does this. It also supports pointin time recovery. And being transactional you can back it up whioe it is in use. Another great ZFS feature is that it includesthe "storage manager" layer in the FS. So mirrors and raid ad spanning volumes as well as access by a cluster of machines is al in the FS. You can do things lie add a new physical drive to the FS while the system is running. ANd being a 128 bit file system you are not likely to fill it up soon I've read about Sun's testing. Seems they have crashed system intentionally write doing heavy write manymany thousands of times and have yet tosee any data corrution. Now that Solaris is Open Source we may oneday be able to know who this works.
You are right. I was a CS major 20+ years ago and I'm sure happy I did not wase my time learnning the "technology of the day needed to get a job" COBOL on punched cards would do me no good today. But I did study the exact same subjects you will "theory of computation, machine leaning and so on. You should be studying the theory of language design and not any one specific programiing language. The theory of operating systems, compiles and databases was not changed from the 1970's when I first studied those things. Yes we did lots of programming but hte profs would not waste lature time explaining how to write a "foor loop" the TAs did that in the lab or we read up on our own.
If they start teaching basic programming to CS majors what is next? Classes on how to use MWS Word for English magers? No, programming is something you just have to know before you can study CS.
Open standards would mean the death of the world's largest software company and they know it. They will do everything they can to prevent the adoption of standards.
Slashdot Mirror
"Why would they want a dual-core laptop? I thought the idea was to not burn the juice?" A good design might be a multi-core notebook that can power cores on and off as required to save power. "Dual" is just the start in a few years we may see 4, 8 or 16 core machines and a good way to manage power use would be to power the cores up and down as required. Even on desktop machines people should care about power. Running my dual CPU Xeon 24/7 likely costs $25/month. Now that we are hitting a Ghz limit we may see Moore's law aplied to cores with a doubling every couple years. I mean, 12Ghz is impractical and 100Ghz is likely to never happen but 16-core chips will be available in time. And Yes, there ARE uses for 16 CPUs, just look at how many 16-cpu servers are sold today and people are wiling to pay $250,000 for them so they must be pretty usfull
Don't worry. The TPM chip does NOT cripple the hardware. I'm typing this on an HP "xw8200" that has a TPM chip on the motherboard. The hardware acts no different TPM is just something that an operating system can choose to use or chose to ignore. I have Solaris 10 install in this box and it ignores the TPM stuff, Linuix would ignore it too as would Darwin or any current version of Windows. What TPM does is this: The OS asks the chip to "measure" the hardware. It does this and returns a cryptographic hash. THe OS can compare this to a list of good hashes and then decide what to do. TPM can also do things like "measure" the software on a disk and the disk itself. This could even be a Good Thing on an Open Source operatring system For example Linux could use TPM as a way to make certain the system is not been compromised. It would be a powerfull security measure. Apple will use the TPM to insure that yu did not swap out the whole machine for a Dell, Gateway or whatever. ANy operating system that wants to be secure has to depend on having a secure and trusted "security kernel" that is "tamper proof" and known to work and be well tested. By "tamper proof" I mean you can __prove__ that it is not been messed with. Right now can you __prove__ that your Linux machine with it's one year "up time" has not had one byte of code changed in it's running kernel. You can't, OK you could write a checksun routine that runs periodically, but how do you know the checksum routine was not modified? You can't, not without locking the checksum routine into the hardware and that is what TPM does. If Apple does this right I'll be happy to know that Mac OS X uses TPM.
You are only half right. Those missions would not have been saved by the crew but they would have been saved by the design and flight procedures used for crewed missions. With a crewed mission the entire spacecraft is filled with more redundancy and there is a larger full time suport staff on the ground and many more reviews and testing. It is actually cost efective to loose unmanned spacecraft. Putting people aboard rises the cost more than ten times so it is cheaper to loose half the unmanned misions then to build and fly them as if they were manned where a only a 2% loss is acceptable.
You said "its slower than the 2.2Ghz dual-core Athlon X2 that's sitting next to it" Apple is not moving to AMD they are moving to Intel. Now compare your 2+ Ghz G5 to a Pentium. The G5 may win. I think the move to Intel was done because of notebooks. G4s are not fast enouh and the G5 would requre the user to cary a backpack full of batteries I really wish Apple had gone to AMD. I want an Apple version of this http://www.sun.com/desktop/workstation/w2100z/spec s.jsp
or maybe even this
http://www.sun.com/servers/entry/v40z/specs.jsp
Note that Sun calls an eight-core Opteron box an "entry level" system.
I'd love to see Apple softwar on AMD based Sun hardware. Sun has actualy
gotten it's prices in-line. thier dual Operon box starts at $2,200 not
far from Apple's Dual G5.
(I'm wrtin this on a dual Xeon box made by HP that runs Solaris 10. Interestingly
the motherbord has the TM chip on it although Solaris ignores it. The chip actualy
_measures_ the system and then expresses the result with a crytpographic hash.
Part of that measuremet is the software. It would be darn hard to defeat
"people used to see no way that a plane could possibly go faster than sound" No, __some__ people thought that many didn't. but the big difference is that the "sound barier" was jst an enginerring problem. Even the people who thought it could not be broken know it was possable in theory. For example a rifle bulet fired in the early 1900's was supersonic and propeller tips of 1930's vintage airplanes could go super sonic. (ever hear that loud buzzing soulnd in old diver bombings, it's the sonic booms from the prop tips) So the sonic barier doubters in the 1940's where saying that you just could not build a powerfull enough engine. Light is an entirely differet _kind_ of limit. There is very solid theoreticalreasons why it is a hard limit. In Azamov's (SP??) book "I Robot" there is a story about the people who final did breakthe light speed limit. Turned out from the view point of our conentional physics objects breaking the limit appear to cease to exist. They simply leave the universe. That may be the only way as the rules of the observable universe don't alow it so you go someplace else.
I'm typing this on a system with two 3.6Ghz Xeons with Hyperthreading enabled. The system uses two 300GB ltra-320 scsi disks set up in a mirror and has 4GB RAM installed. When I run Nescape the performance is about as good as on my other system which is a 2.4Ghy P4 with 1GB RAM and one SATA drive. However the Dual Xeon system runs my DBMS querries blazingly faster, __much__ faster then the P4 based system. Many DBMSes work like Apache and "fork" a new server process for each client, so when 12 process each connect to the DBMS I see 12 copies of the DBMS server software running DBMSes tend to run in tight loops where the instructions stay cached in the CPU's L2 cache Solaris is smart enough to "know" which CPUs share on-chip cache and that all CPUs are not "symetric" and so can schedule process to stay within a group of "close" CPUs. At boot time the system looks around and builds a hiarchical model of the machine's configuration, noting which PCI busses, CPU and memory are connected to what. On the larger systems there are cards wth four CPU and up to 4GB RAM and some PCI buses all on one card and then the cards are connected by backplane So when you read about these new CPUs don't think about a low-end PC runnig Windows These could be intended to go into a 20-way mutiprossor runnig Solaris (or Mac OS X) You should figure that in 10 years the system on your desk will use techniques like today's high end systems. What Sun's Sunfire does at the board level (four CPUs, Busses and RAM) wil be done at the chip level and you will see four core chips and then 16 core chips andthen computersbuilt with multiple 16-core chips designed like today's "starfire". What would yo use such a system for? How ablut controlling a robot that can walk up a flight of stairs and responce to voice commands and identify objects with a vision system all at once. Another aplication is video rendering, lots of data t process but the _code_ stays cached ad rentering like DBMS and web servering to paralizable. In fact the video render problem is the prime example of what to use a room full of CPUs for. And just wait 'till you buy a High Def Video camera. You will want one of those "quad core" power macs
There is a test facility down the street from here built by "TRW" (now part of Northrop Grummun) The test facility has the ability to reproduce the acustic signature of a large space lift boster, like an Atlas or the Space Shuttle. "Loud is not quite the word. A better word is "Destructive" The sound of the launch can literally rip stuff apart if it is not built right and hence the reason for the test chamber. The chamber is large enough to hold a typical satilite payload, which can be about the size of a stadard shipping container. So, my point is that there are good reasons to build powerfull subwoffers other thn reproduction of music. One other use was in a facility I vistited in Florida that housed several Abrams A1 tank simulaters. The sound was realistic (as judged by some US Army tank core peole) I would just say it was "way loud".
Aple does this now. "Core Image" is built into the OS and all "correctly written" applications that need to do graphic use Core Image. Core Image wil use the GPU if one is available. This is a very good idea but the hardest part of getting this to work on a non-Apple platform will be standarizing the API so that we can use any GPU. OK X11 did this fr displays on UNIX ad we have OpenGL for 3D graphic so we can hope something will happen an API for GPU based image transfomation. The biggest use for this wil not be just simple transcoding but editing and dispay programs for still and moving images Think "gimp" and "cinelerra".
Solaris has "mapmalloc" which is mmap based malloc and it has "watchmalloc" which uses something like a breakpoint that you can have watch for reads and writes to the "wrong" places. Basically two independent malloc replacements. With watchmalloc being the more sophisticated of the two. These will find errors in stuff that has worked for years. Solaris man pages for mapmalloc and watchmalloc are on-line for anyone who wants them (see google) Solars now is Open Source too. From my experiance using it and linux daily both thier intial releases (I remember Linix 0.x and I went into a Sun office see Solaris before it hit the street) Opinion: The Solaris kernel and related utilities is the most adavnace there is on the X86 patform. It is Very stable and will servive things like pulling a CPU looseout of a socket and will scale to over 100 processors. However it is the worst *NIX desktop OS on the plannet. I should know it use it as my desktop every day all day
"In some social-study class or other where it can be taught along side of Astrology, Divination, tea-leaf reading and the theory of the Abominable Snowman. Just not in science class." No, that's a bit harsh. ID could be taught in a class that covers comparitive religion. For example they could also cover "mainstream" Christian teachings (Roman Catholic Church and a couple major Protestent sects as well) then they could cover maybe Hindu and American Indian creation stories too. It could actually be an interesting and usfull class there they show how each society makes up a creation story to fit thier culture. In fact this could even be taught in science. Yes real science with statistics and math and all that stuff. The root of this conflict is that for many peope science IS faith. They haven't a clue. They think science is just a collection of always changing facts that one most memorize and accept on faith.
I always like those arguments that basically say "THEY could ot have done X because I can't figure out how to do X" The fact that some guy could not figure out how to burn wood with a mirror does not mean that one of the smartest guys in all of history could not have. In later centurys it was common to build solar furnaces Here is a link to the largest modern solor reflector http://www.imp.cnrs.fr/foursol/1000_en.shtml
FOr non-triveal software it would be _very_ hard to identify who was responcible fo a bug. Most software is a group effort. You have someone who specifies what it is to do, some who design it but don't write the code and others who do write code but don't get to do the overall design and then you have the people who test and document it. When something fails you have to look at the process that created it as having failed. A perfect example of this is the MS Windows OS. The root causes of it's problem are conceptual. No one person is to blame the system just eveloved from the non-networked, single user, single task OS, into a networked, one user at a time, multitaking OS. When you try to carry backward compatabilty through such an evolution you have a mess. I read a lot of things about sodftware managment from people who have not written a line of code in 20+ year or even ever. they sem to loose touch with reality
It should be possable to destroy an RFID device with a very powerfull _pulse_ of RF energy. People who like to use shreaders could buy a device that frys the RFID chips. I think these "RFID shreders" could be low priced too as the _average_ power emitted would be low
A two inch diagonal screen just won't cut it with most people So how to make it work? They make "ear buds" so why not "eye buds"? Ok bad idea. But "iGlasses" would work. A little projector makes a head up display. It appears to the viewer as a large screen about a meter away. It's bad enough now, but just wait 'till you start seeing people driving on the freeways with these "iGlasses". I imagine the porn industry woud find a use for this too.
Sun's "ZFS" applies error detection and correction to all data. ZFS is a file system that also includes a disk manager that provide RAID and volume concatination. It does it all in a single layer. I don't know if standard PC's are fast enough yet that checksumming all the data as it goes to and from the disks would work. But on Sun server hardware the extra checksume is triveal ZFS puts 64 bit checksums on all the data and the metadata ZFS is also Open Source so someday we might, maybe see a port to some other OS
When you install the machine and periodically thereafter you would "null" it. That means you adjust the needle to read zero when there is nothing in the detection chamber. I used an oxigen sensor resently. The first step is to expose it to just plain air and adjust it to read 21% (air is about 21% oxigen, 79% nitrogen) After doing this it can detect very small amounts In both cases you have to tell the machine "This is the normal background." After this the machine detects changes.
If these really will be given away free to anyone who needs one they will be _quite_ hard to sell. Kind of like if I tried to sell you a bag full of air. At first these might be sold but after millins of these things are dumpped into all the schools thier street price will be nearly zero. This is the only way to eliminate drug dealers too. If you could get all the pot or speed or you wanted for free at any government building (Jst walk into a post office and help yourself) the people seling the stuff on the street would go broke. Same with PCs, if the schools had stacks of the things there would be no black market. That said, if these Really do cost $100 to make why doesn't he ellthen here in the developed worl for $200 each. I'd be happy to buy two at that price which would pay for two kids in Africa. There is also plenty of uses for a $100 PC here in the US. For example my bother is a bit of a computer geek but lives in an "assisted living" facility (aka "old folks home") because he i9s disabled. He has been setting up and giving away computers to some of the people there who are stuck in bed. These guys couldn't afford a computer but it sure does expand thier "world" the web beats 24x7 TV. e-mail and decussion forums keep the brain from turnning to mush. At $100 a pop I could afford to give away a few computers an likley would.
Do you know the formula that relates transmit power, distance, and bandwidth? Of course it is _possable_ to build such a network. My point was that it has not been done Your write "assuming you can contact it, and it can contact you." It depends on the data rates. You can have a 10,000 watt tansmitter on Earth but the power available to a rover on mars may be only a few watts. The links tend to be hugly asymetric. A small Mars rover can send on order of a few hundred bits per second directly to Earth but can send 10X faster to an orbiter that is only a few hundred kilometers overhead. A moveable microwave antenna is not the best setup for switching packets. The antennsa slew at very slow rates and of course when it is pointing at one location it can't pont in others. This is not to say general purposes packet switches can not be built. Look at this www.losangeles.af.mil/SMC/MC/Tsat.htm TSAT will be a switch but it will conect users on the ground not in space. Maybe years from now someone will built a TSAT that looks upward? but I doubt they will add TSAT-like capability to everythingthat is launched.
You've been reading to much science fiction. Yes there are some cases where spacecraft use inderect means of communication through a relay but this is not done ad-hoc using some general purpose capability built into every spacecraft. In every case wherwe relay is used the capabilty is plaanned from the beginning. The idea of selecting some random spacecraft to use as a relay to soe other random spacecraft just can't work. The orbiters currently on mars were design specifically to relay. Closer to Earth TDRSS acts as a relay between low Earth orbit and the ground. Notice (1) that TDRSS is the relay, thaey are NOT sending data between randon spacecraft and (2) the data are passed only between LEO and the ground, not through out the solar system or even to geosync. orbit. http://msp.gsfc.nasa.gov/tdrss/oview.html
They have sold 21 million of these. Lets say Apple is totally out of control and went nuts and spent one hundred million on design. so it's like five bucks each on enginerring. But I'll bet much of the design work for the internal software andother parts is the same as in the other iPods and of course iTunes is not new for the nano. I'd bet between four anfeight bucks per unit now and it goes downwith each unit sold. Other costs are for things like warenty and technical support
You said " The US is financially overstretched as it is." Let's do the math. NASA says it will cost about 100B over 13 years to go to the moon. there are 300M people in the US. If each person pays 50 cents per week then NASA gets enough to fund the program. Yes I know, some peole will get hit up to $10 per week and other get off with paying zero bt it averages to 50 cents per person per week.
"why no filesystems seem to support copy-on-write semantics"
Look at Sun's "ZFS". It does this. It also supports pointin time recovery. And being transactional you can back it up whioe it is in use. Another great ZFS feature is that it includesthe "storage manager" layer in the FS. So mirrors and raid ad spanning volumes as well as access by a cluster of machines is al in the FS. You can do things lie add a new physical drive to the FS while the system is running. ANd being a 128 bit file system you are not likely to fill it up soon I've read about Sun's testing. Seems they have crashed system intentionally write doing heavy write manymany thousands of times and have yet tosee any data corrution. Now that Solaris is Open Source we may oneday be able to know who this works.
Re-reading what I just wrote. Maybe they should have made us take "typing 101".
You are right. I was a CS major 20+ years ago and I'm sure happy I did not wase my time learnning the "technology of the day needed to get a job" COBOL on punched cards would do me no good today. But I did study the exact same subjects you will "theory of computation, machine leaning and so on. You should be studying the theory of language design and not any one specific programiing language. The theory of operating systems, compiles and databases was not changed from the 1970's when I first studied those things. Yes we did lots of programming but hte profs would not waste lature time explaining how to write a "foor loop" the TAs did that in the lab or we read up on our own. If they start teaching basic programming to CS majors what is next? Classes on how to use MWS Word for English magers? No, programming is something you just have to know before you can study CS.
Open standards would mean the death of the world's largest software company and they know it. They will do everything they can to prevent the adoption of standards.