Our planet already has 'nano-scale' machines which self replicate. Bacteria have been breaking down complex molecules in order to exponentially self replicate for, well, about as long as life has existed on this planet. What has stopped a single celled organism turning everything into 'grey goo' already?
I expect it something to do with the amount of energy required to do the job. Although there's a lot of energy around, it's distribution is fairly sparse. Evolution has already made some pretty damn good systems for capturing, storing and using stored energy. Unless nanobots happen to be an order of magnitude more efficient than any possible thing evolution has ever produced, I doubt that it would be possible to achieve any high-impact 'grey goo' scenario.
One of the issues which this will raise (yet again) is how terrible, natural, unnatural, suicidal, dangerous, safe, humourous and so on it is that humans will adapt animals to meet their needs. I would argue that such exploitation is not necessarilly the way you see it: time to flip the coin.
How many wild chickens of original genus have you seen in the western world? Of all the chickens in the western world, how many have avoided a few thousand years of selective breeding? For both these questions, I would say it is virtually 0. But is this/our/ plan at all?
Humans eat chickens, amongst other things. In fact, humans eat an awful lot of chicken. World chicken population has been estimated to be/twice/ that of humans. That's around 13 BILLION chickens folks.
Why are there so many? Well, because we eat them and the eggs they produce. If chickens were poisonous, how many chickens would there be? Certainly not 13 BILLION of the little buggers!
Just think of how we have been exploited by the chickens: we let them breed to huge proportions, we feed them, keep them safe from predators (except one), try to keep them disease free using the tools we have developed... but/why/! Well, they're quite tasty when hacked to pieces and cooked... and their (unfertilised) eggs aren't bad either.
Just think of what the 6.4 billion of us and the 13 billion chickens could do together in the future - we could terraform planets and spread ourselves (and the chickens) across the stars... the first chicken nugget served on another planet will be a great day in the human-chicken partnership!
I for one are quite happy with this symbiotic relationship: the chickens help us achieve our (not so obvious) goals (food is important!) and in return we keep their species alive and, like we have done before all over this planet, we will take them wherever we settle.
Personally, I think the chickens get a great deal.
Ian Woods (Just don't get me started on how humans are being exploited by plants: do you think those arable crops growing over a lot of the better farming land all over the planet is because of just our desire to grow and eat them?)
One of my profs is interested in photochemistry and he gave our class a statistic: if you tally up all the energy humans use and produce in a year, you find that an equivalent amount of energy from sunlight strikes the surface of the earth in one hour.
Interesting. That's a hell of a lot of energy! Consider the amount of energy which doesn't strike the earth... staggering. Efficiency? What's that!
So if we have this much energy coming in, why bother converting to hydrogen at all? Inefficiency aside, you create the problem of storing a very large volume of highly explosive gas, when we could just be storing water.
You could easily make a distribution chain whereby electricity was produced by this process on a large scale either by cracking water onsite (the clever way) or getting it shipped in. Similarly, having cars running of hydrogen LPG is more feasible than running them from sunlight.
The benefit of having a distribution chain is that the water cracking plants can be much less efficient than a process to drive a machine in situ because it is being produced in bulk. This is mostly because we know how to efficiently use combustible gasses and don't know effective ways of using sunlight.
I don't see pressurised hydrogen gas as being any more or less dangerous than most hydrocarbon fuels swilling around the inside of billions of vehicles on the streets of the world.
There are accidents, and there are explosions but they aren't a major risk (unless you're driving a Ford...;)
I just hope this happens instead of oil being the predominant source of energy.
The only downside I can see is the number of Arabs and Texans who would find their market diminishing rapidly as companies and countries started making most of their energy from it's most abundant source - sunlight...
This really only amounts to someone in NASA trying to justify their job and trying to make policy that is pretty much silly and un-warrented.
Scenario 1:
American tourist billionaire who happens to be an alcoholic enters space station with contaband. Get's slaughtered. Accidentally destroys an experiment by throwing up over it(being drunk in a weightless environment I'd image is quite different to being drunk on Earth). Bits of vomit infest critical systems.
Cost of clean up... shit loads. New materials and shipping costs to relaunch the experiment, time spent restraining drunk passenger, time and material (and shipping) to fix/clean critical systems.
Who's at fault? Who would pay for the cost of cleanup... the people responsible for the unruly tourist or the people who have had their equipment damaged? Not to mention the physical risks involved.
Scenario 2:
Russion mobster accidentally pushes button and vents the propellant to be used to counteract orbital decay. Tells nobody. The next thing they know a piece of debris from a lower orbit which they have just entered and damages the station.
Again, who is responsible?
At all times you must remember that the station is an international effort with partners from all over the world. It is in the interests of all involved to exclude 'tourists' who may pose a physical or financial threat to themselves/and other partners/.
IMO such exclusions make sense: do you really want to put people who have psychological traits which may lead to unacceptable behaviour (alcoholism, drug abuse)? Do you really want people who's honesty has brought into question in the past to the point of criminal fraud?
In the future, when the environment is safer, perhaps. In reality, these measures are there to protect the staff who are on board the station, the station itself and the financial and scientific commitments which have been made by the partners.
Would you want to be on a space station, an environment where stupidity is easily fatal for yourself and others around you, with someone who you cannot completely trust?
Conversion between different forms of energy results in a low efficiency. Converting from electricity to hydrogen and back would have an efficiency of about 50% (at a guess), not including transport of the hydrogen. The electricity grid has an efficiency of about 75% (again an educated guess), and is very convenient.
This is indeed true: the processes which convert energy from one form to another are rarely efficient. There is one thing which has struck me about this: we don't need to be efficient. What we need it lots of energy which we can waste.
On a day to day basis we waste more energy than could be imagined. The great fiery fusion reactor in the sky emits it whether we use it or not. Lot's of it. More energy than even a Californian household could use! In a sense, sitting on a rock and collecting what comes out way isn't very efficient. I doubt we get more than a hundred billionth of the energy that we could get. It's just difficult to it!
Imagine a process of generating hydrogen using sunlight which was commercially viable but was only 1% efficient. Is this a problem? Not if you can collect 100 times more of the free energy which is being lost from the great fusion reactor in the sky. To put it in context, use a reflector 10 times the size.
This is entirely an engineering problem, and with current technology it's solvable but expensive. The end result though, is enough energy for rock and roll for as long as the fusion reactor in the sky is pumping it out. All we have to do is enough work to collect some of what's wasted!
(yes, I know. Ultimately, you want an elipse with a collector at one focus and the sun at the other so we don't waste any of it. Dyson sphere anyone?;)
You still need to recharge your laptop with Hydrogen, and it will polute just as much as batteries do today(very little)
There is a funamental difference between the use of hydrogen for power in a fuel cell and the use of a normal cell.
You 'make' a cell. It's made of some pretty nasty things. There are around 70 million mobile phones in the UK, each of which has around 100 grammes of heavy metal in their rechargable (but not infinitely so) cells. In a few years each of these cells will need to be replaced. By my calculation that's 7 million kilogrammes, or 7000 tonnes of heavy metals that needs to be processed and/or recycled (or, thrown away). That's a hell of a lot of material for a place the size of the UK.
A hydrogen fuel cell does not contain such nastiness and, baring wear and tear and failure, would last at least as long. Additionally, the oxidation of hydrogen (which is ultimately where the energy comes from) produces water. Not slightly nasty water, but pure water. The kind of thing which tastes awful;)
I think you'd find that a hydrogen fuel cell industry would be far better for the environment and cheaper to do (lower displosal, recycling, replacement cost) than any other chemical storage power supply.
The part of the equation which is missing is an efficient (read, cheap) way of producing hydrogen. Sunlight is clean and free - if we don't use it then it's wasted. The sun emits quite a lot of it continuously. Using a million times more of it than we necessarilly need is still a hell of a lot cheaper and better than drilling for crushed and fermented vegetation below rock strata. If it can be made commercially viable, I for one am all for the use of H2 for power. The advantages are clear: the only thing missing is a cheap and plentiful supply of H2.
Handheld game-playing devices have been in production since the days of the Atari Lynx. The gameboy (original) in particular shouldn't be too hard to get hold of, and certainly shouldn't be particularly expensive. They're based on a slightly cut down Z80 core (AFAIK the only real difference is the lack of the index registers (IX, IY) and their associated instructions).
The gameboy advance(d?) is another possibility: probably more expensive but much easier to get hold of. I've been told that they're based on an ARM7 core.
In both cases, you get enough peripherals for some interesting programming challenges. Seeing your program output to the LCD is a far more rewarding experience than examining the result in a register or a set of memory locations. Also, I'd expect their architectures to be very simple - their hardware is very well understood, and other than their compactness nowhere near cutting edge (and so no bizarre hacks required).
Ian Woods
There are telescopes and then there are telescopes
on
The Humble Space Telescope
·
· Score: 2, Informative
IMO comparing this and the HST is misleading. They may be both telescopes, they may both be on satellites and they may both be in orbit (or, eventually be) but they are very different devices.
The diffraction limit of a telescope is proportional to the size of it's aperture - the size of the 'end' where the light goes in. Basically, the more light you can capture the more you can see: generally, very faint things are hard to see. The resolving power of a telescope is believed to be limited by the diffraction limit. (I may of course be wrong, I'm not a qualified astronomer - I've only just began.)
Ground based telescopes are considerably larger and, if there wasn't kilometres of air between it and the near-vacuum of space, would have much greater resolving power than the HST.
At the time the HST was launched, there was a need to 'see' without the atmosphere being in the way. It has been useful. Modern ground based telescopes using things like adaptive optics can see things which the HST can't - don't underestimate the resolving power of a telescope on the ground compared to the HST. The HST is still very useful, even AO doesn't correct 'perfectly', but then the HST optics are not perfect either: those spikes that come off the center of the stars aren't supposed to be there. They're called diffraction spikes and I'm lead to believe they're caused by the telescopes structure.
The MOST will be a useful tool, but it's a tool for a purpose and as a general telescope I'm not confident that it would compare well to the HST or ground based telescopes.
At the rate ground based telescopes are improving, don't expect the additional cost of producing general telescopes like the HST to be met often - but I'd expect more devices with specific purposes (for instance, X-Ray astronomy which is quite scary - I've been told that a handful (5) of photons is a good detection!)
The embedded system designers are not techno-clumsy CEOs. They are highly experienced engineers, for the most part. They know that they need to take MS's words with a grain of salt, and they know that Linux systems have hidden costs associated with them.
I think it's fairly clear from the wording that MSs white paper isn't targetted at the people developing in any serious way - it's almost completely made out of the kind of information a managerial type might find useful.
Once Linux is 'demystified' it's clear that MSs white paper is a marketing tool and nothing more. Of course, if you have the hearts, minds and wallets of the managers at various levels then they can ensure than Windows XP Embedded is used in projects. Anyone who has enough knowledge of the project itself, Linux internals and what MS says in it's white paper would find it difficult to rate XP Embedded well/unless/ you're talking about a PDA rather than an 'embedded' system.
Embedded Windows and embedded Linux are both getting their asses kicked up and down the block by WindRiver's VxWorks, so this whole thing is like watching a couple sparrows fighting over a crumb that the crow dropped.
By itself this isn't much to be concerned with, and I agree with your statement. Unfortunately, MS seems to be targetting Linux at as many levels as possible to eliminate, before it's too late, any large-scale business interest it may have or be acquiring. The bickerings over the crumb spill over into the other arena's where comparisons between Linux and MS OSs are made.
In UK patent law there is an 'obviousness' clause - if a patent claim is obvious to a reasonable proportion of people trained in a suitable field then it doesn't stand as a patent claim.
I'd argue that the major patents these two organisations are trying to bandy about are obvious not only to softies, or people who work on digtal video or the TV industry, but obvious to pretty much everyone with an IQ over 110.
US patent law doesn't have such a clause that I am aware of (but, as stated before, IANAL and definately not an American one) and so it's possible to patent something blindingly obvious and throw lawyers at anyone who does the same, obvious thing.
Funny how with the windows instances it was only necessary to say "Windows crashed, it sucks" but where there was a linux problem the whole thing was grossly rationalized.
I've seen this as well, but then... there is another aspect to it. When windows BSODs... I almost never know why. It's always a random death. The times when I have seen kernel panics or faults under Linux systems they're generally because I have done something insanely stupid and I can tell what I've done wrong. I don't need to 'rationalise' - I can say that if I try and boot Linux with this thing wrong then it whines...
Shame you just can't do the same thing with windows...
Why would you want to run fibre? Let's just run through some of the reasons why...
Fibre doesn't suffer many of the EM interferance problems that electrons-in-wires do. Useful, in a high noise environment...
...then again, my computer doesn't crash when I turn a light on nearby when the case is off. Not running parallel to AC-power carrying wires is a must though - long parallel lengths will introduce crosstalk. Just don't do it, or when you do maximise the distance between the power cable and the data cable (or use data cable).
The transmission length of a fibre is much greater than that of a wire at high speeds...
...but then, how big is your house? 100BaseT ethernet over Cat5 gets you 100m typically... is your house so large that it won't reach? If you need more bandwidth, then Gigabit ethernet over balanced copper gets you 25m... but if you really need that then compression will have gone out of fashion.:)
Reasons why not...
Fibre is expensive! (Copper wire is cheap.)
One thing which is also worth noting is that 10BaseT and 100BaseT only use 2 pairs (4 wires), and CAT5 has 8. This means you can put one 10/100BaseT stream and still have 2 pairs left for 2 audio/phone lines, a second ethernet stream or whatever.
This is obviously a contentious issue, and isn't flamebait on purpose (though it is perhaps possible that a fibre-in-the-home advocate might see it as such). IMO, copper wire will be around for a very long time - it's actually not too bad as long as you don't want to move data over to the other side of town....
I bet the LCD manufacturers would be eager to supply them if so.
I doubt that. How many LEDs are bought on a day to day basis? How many do you have in your home, your work, the shop you use for supplies and so on and so forth. There's a lot of LEDs in the world - and more every day.
The increased revenue from making a few million LED 'bulbs' is hardly worth the effort of developing it - especially since they won't burn out for a long time.
Since that is not the case, something must be done either to dechipher individual brains, or as you suggest make different brains look alike. But a single interface would not solve the problem, the interfaces to the external interface would still be higly variable.
The 'interface' of the huge bundle of neurons that makes up the optic nerve is not the same for each individual, but almost all individuals can see. I expect this is similar for all of the interfaces which humans have (the development of motor-control is very interesting in young humans - 5 year olds can't write well not because they don't know what shapes to draw but because they don't have the fine control required for such delicate work).
I am not a neurologist and my understanding of the brain is very superficial (how many neurologists read./ anyway?:P ), but I'd expect that a Meeting-The-Problem-Halfway(TM) approach would be successful: don't expect the brain to work with the interface, or vice versa - let at least one of them adapt. The adaptability of even adult brains shouldn't be overrated (but perhaps not up to the level required): the good old 'walk around with vertically inverting glasses, and after a week or so you're brain with 'adapt' to see the world correctly' experiment has shown some significant adaptation.
Certainly, neurological interfaces are one step closer - but for now I'm more interesting in finding nerves which I can control, or learn to control which aren't necessarilly in my brain! That extra limb can come in handy:) (I can now hold the Big Mac with both hands,/and/ hold the carton underneath to catch the lettuce that falls out the other side!)
Speech recognition is a pain in the ass. They almost certainly just want to use telephone tones.
Very large dictionary VR is still a little unpredictable, even when they're trained to one specific person. For limitted dictionary (say, a few dozen words), speaker independent(*) VR is an awful lot more successful. One such project was on slashdot a long time ago and has been in development since - Sphinx.
As for picking out spoken numbers, back in the days when I was figuring out how to do simple VR I trained myself to the point where I could recognise a number (between 0 and 30 or so) by only looking at a graph of it's waveform - creating a markov model for that purpose would get results as good if not better.
(* This depends, in Sphinx's case, on the quality of the language model. A language model well suited for recognising mid-west American accents would work poorly on Icelandic... for obvious reasons.)
But surely we could agree on some standard - a set of benchmarks evaluating the machine's performance at representative tasks.
The problem is defining what a representative task is. The requirements of even simple computation varies so much that no matter what 'tasks' you perform there will always be some algorithm which doesn't have any parrallel task against which you've benchmarks.
I've already thrown out every single benchmark I've found for one task I'm doing - the requirements of the algorithm are nothing like sorting a spreadsheet, or searching in a database.
The problem is that the information on/real/ performance of a device is either difficult or impossible to get hold of easily. I could analyse a number of hardware platforms in great detail (say, 4 or 5 days research each) and get a reasonable idea of their suitability for this task, but it's an awful lot of effort. In some cases, the information is simply not available (or if it is, it's hard to find... anyone used Intel's website recently? I was told by one of their guys to use google to find what I wanted on their site because they didn't know where it was...)
Slashdot Mirror
Our planet already has 'nano-scale' machines which self replicate. Bacteria have been breaking down complex molecules in order to exponentially self replicate for, well, about as long as life has existed on this planet. What has stopped a single celled organism turning everything into 'grey goo' already?
I expect it something to do with the amount of energy required to do the job. Although there's a lot of energy around, it's distribution is fairly sparse. Evolution has already made some pretty damn good systems for capturing, storing and using stored energy. Unless nanobots happen to be an order of magnitude more efficient than any possible thing evolution has ever produced, I doubt that it would be possible to achieve any high-impact 'grey goo' scenario.
One of the issues which this will raise (yet again) is how terrible, natural, unnatural, suicidal, dangerous, safe, humourous and so on it is that humans will adapt animals to meet their needs. I would argue that such exploitation is not necessarilly the way you see it: time to flip the coin.
/our/ plan at all?
/twice/ that of humans. That's around 13 BILLION chickens folks.
/why/! Well, they're quite tasty when hacked to pieces and cooked... and their (unfertilised) eggs aren't bad either.
How many wild chickens of original genus have you seen in the western world? Of all the chickens in the western world, how many have avoided a few thousand years of selective breeding? For both these questions, I would say it is virtually 0. But is this
Humans eat chickens, amongst other things. In fact, humans eat an awful lot of chicken. World chicken population has been estimated to be
Why are there so many? Well, because we eat them and the eggs they produce. If chickens were poisonous, how many chickens would there be? Certainly not 13 BILLION of the little buggers!
Just think of how we have been exploited by the chickens: we let them breed to huge proportions, we feed them, keep them safe from predators (except one), try to keep them disease free using the tools we have developed... but
Just think of what the 6.4 billion of us and the 13 billion chickens could do together in the future - we could terraform planets and spread ourselves (and the chickens) across the stars... the first chicken nugget served on another planet will be a great day in the human-chicken partnership!
I for one are quite happy with this symbiotic relationship: the chickens help us achieve our (not so obvious) goals (food is important!) and in return we keep their species alive and, like we have done before all over this planet, we will take them wherever we settle.
Personally, I think the chickens get a great deal.
Ian Woods
(Just don't get me started on how humans are being exploited by plants: do you think those arable crops growing over a lot of the better farming land all over the planet is because of just our desire to grow and eat them?)
One of my profs is interested in photochemistry and he gave our class a statistic: if you tally up all the energy humans use and produce in a year, you find that an equivalent amount of energy from sunlight strikes the surface of the earth in one hour.
;)
Interesting. That's a hell of a lot of energy! Consider the amount of energy which doesn't strike the earth... staggering. Efficiency? What's that!
So if we have this much energy coming in, why bother converting to hydrogen at all? Inefficiency aside, you create the problem of storing a very large volume of highly explosive gas, when we could just be storing water.
You could easily make a distribution chain whereby electricity was produced by this process on a large scale either by cracking water onsite (the clever way) or getting it shipped in. Similarly, having cars running of hydrogen LPG is more feasible than running them from sunlight.
The benefit of having a distribution chain is that the water cracking plants can be much less efficient than a process to drive a machine in situ because it is being produced in bulk. This is mostly because we know how to efficiently use combustible gasses and don't know effective ways of using sunlight.
I don't see pressurised hydrogen gas as being any more or less dangerous than most hydrocarbon fuels swilling around the inside of billions of vehicles on the streets of the world.
There are accidents, and there are explosions but they aren't a major risk (unless you're driving a Ford...
I just hope this happens instead of oil being the predominant source of energy.
The only downside I can see is the number of Arabs and Texans who would find their market diminishing rapidly as companies and countries started making most of their energy from it's most abundant source - sunlight...
Ian Woods
This really only amounts to someone in NASA trying to justify their job and trying to make policy that is pretty much silly and un-warrented.
/and other partners/.
Scenario 1:
American tourist billionaire who happens to be an alcoholic enters space station with contaband. Get's slaughtered. Accidentally destroys an experiment by throwing up over it(being drunk in a weightless environment I'd image is quite different to being drunk on Earth). Bits of vomit infest critical systems.
Cost of clean up... shit loads. New materials and shipping costs to relaunch the experiment, time spent restraining drunk passenger, time and material (and shipping) to fix/clean critical systems.
Who's at fault? Who would pay for the cost of cleanup... the people responsible for the unruly tourist or the people who have had their equipment damaged? Not to mention the physical risks involved.
Scenario 2:
Russion mobster accidentally pushes button and vents the propellant to be used to counteract orbital decay. Tells nobody. The next thing they know a piece of debris from a lower orbit which they have just entered and damages the station.
Again, who is responsible?
At all times you must remember that the station is an international effort with partners from all over the world. It is in the interests of all involved to exclude 'tourists' who may pose a physical or financial threat to themselves
IMO such exclusions make sense: do you really want to put people who have psychological traits which may lead to unacceptable behaviour (alcoholism, drug abuse)? Do you really want people who's honesty has brought into question in the past to the point of criminal fraud?
In the future, when the environment is safer, perhaps. In reality, these measures are there to protect the staff who are on board the station, the station itself and the financial and scientific commitments which have been made by the partners.
Would you want to be on a space station, an environment where stupidity is easily fatal for yourself and others around you, with someone who you cannot completely trust?
Ian Woods
Conversion between different forms of energy results in a low efficiency. Converting from electricity to hydrogen and back would have an efficiency of about 50% (at a guess), not including transport of the hydrogen. The electricity grid has an efficiency of about 75% (again an educated guess), and is very convenient.
;)
This is indeed true: the processes which convert energy from one form to another are rarely efficient. There is one thing which has struck me about this: we don't need to be efficient. What we need it lots of energy which we can waste.
On a day to day basis we waste more energy than could be imagined. The great fiery fusion reactor in the sky emits it whether we use it or not. Lot's of it. More energy than even a Californian household could use! In a sense, sitting on a rock and collecting what comes out way isn't very efficient. I doubt we get more than a hundred billionth of the energy that we could get. It's just difficult to it!
Imagine a process of generating hydrogen using sunlight which was commercially viable but was only 1% efficient. Is this a problem? Not if you can collect 100 times more of the free energy which is being lost from the great fusion reactor in the sky. To put it in context, use a reflector 10 times the size.
This is entirely an engineering problem, and with current technology it's solvable but expensive. The end result though, is enough energy for rock and roll for as long as the fusion reactor in the sky is pumping it out. All we have to do is enough work to collect some of what's wasted!
(yes, I know. Ultimately, you want an elipse with a collector at one focus and the sun at the other so we don't waste any of it. Dyson sphere anyone?
Ian Woods
You still need to recharge your laptop with Hydrogen, and it will polute just as much as batteries do today(very little)
;)
There is a funamental difference between the use of hydrogen for power in a fuel cell and the use of a normal cell.
You 'make' a cell. It's made of some pretty nasty things. There are around 70 million mobile phones in the UK, each of which has around 100 grammes of heavy metal in their rechargable (but not infinitely so) cells. In a few years each of these cells will need to be replaced. By my calculation that's 7 million kilogrammes, or 7000 tonnes of heavy metals that needs to be processed and/or recycled (or, thrown away). That's a hell of a lot of material for a place the size of the UK.
A hydrogen fuel cell does not contain such nastiness and, baring wear and tear and failure, would last at least as long. Additionally, the oxidation of hydrogen (which is ultimately where the energy comes from) produces water. Not slightly nasty water, but pure water. The kind of thing which tastes awful
I think you'd find that a hydrogen fuel cell industry would be far better for the environment and cheaper to do (lower displosal, recycling, replacement cost) than any other chemical storage power supply.
The part of the equation which is missing is an efficient (read, cheap) way of producing hydrogen. Sunlight is clean and free - if we don't use it then it's wasted. The sun emits quite a lot of it continuously. Using a million times more of it than we necessarilly need is still a hell of a lot cheaper and better than drilling for crushed and fermented vegetation below rock strata. If it can be made commercially viable, I for one am all for the use of H2 for power. The advantages are clear: the only thing missing is a cheap and plentiful supply of H2.
Ian Woods
Handheld game-playing devices have been in production since the days of the Atari Lynx. The gameboy (original) in particular shouldn't be too hard to get hold of, and certainly shouldn't be particularly expensive. They're based on a slightly cut down Z80 core (AFAIK the only real difference is the lack of the index registers (IX, IY) and their associated instructions).
The gameboy advance(d?) is another possibility: probably more expensive but much easier to get hold of. I've been told that they're based on an ARM7 core.
In both cases, you get enough peripherals for some interesting programming challenges. Seeing your program output to the LCD is a far more rewarding experience than examining the result in a register or a set of memory locations. Also, I'd expect their architectures to be very simple - their hardware is very well understood, and other than their compactness nowhere near cutting edge (and so no bizarre hacks required).
Ian Woods
IMO comparing this and the HST is misleading. They may be both telescopes, they may both be on satellites and they may both be in orbit (or, eventually be) but they are very different devices.
The diffraction limit of a telescope is proportional to the size of it's aperture - the size of the 'end' where the light goes in. Basically, the more light you can capture the more you can see: generally, very faint things are hard to see. The resolving power of a telescope is believed to be limited by the diffraction limit. (I may of course be wrong, I'm not a qualified astronomer - I've only just began.)
Ground based telescopes are considerably larger and, if there wasn't kilometres of air between it and the near-vacuum of space, would have much greater resolving power than the HST.
At the time the HST was launched, there was a need to 'see' without the atmosphere being in the way. It has been useful. Modern ground based telescopes using things like adaptive optics can see things which the HST can't - don't underestimate the resolving power of a telescope on the ground compared to the HST. The HST is still very useful, even AO doesn't correct 'perfectly', but then the HST optics are not perfect either: those spikes that come off the center of the stars aren't supposed to be there. They're called diffraction spikes and I'm lead to believe they're caused by the telescopes structure.
The MOST will be a useful tool, but it's a tool for a purpose and as a general telescope I'm not confident that it would compare well to the HST or ground based telescopes.
At the rate ground based telescopes are improving, don't expect the additional cost of producing general telescopes like the HST to be met often - but I'd expect more devices with specific purposes (for instance, X-Ray astronomy which is quite scary - I've been told that a handful (5) of photons is a good detection!)
Ian Woods
The embedded system designers are not techno-clumsy CEOs. They are highly experienced engineers, for the most part. They know that they need to take MS's words with a grain of salt, and they know that Linux systems have hidden costs associated with them.
/unless/ you're talking about a PDA rather than an 'embedded' system.
I think it's fairly clear from the wording that MSs white paper isn't targetted at the people developing in any serious way - it's almost completely made out of the kind of information a managerial type might find useful.
Once Linux is 'demystified' it's clear that MSs white paper is a marketing tool and nothing more. Of course, if you have the hearts, minds and wallets of the managers at various levels then they can ensure than Windows XP Embedded is used in projects. Anyone who has enough knowledge of the project itself, Linux internals and what MS says in it's white paper would find it difficult to rate XP Embedded well
Embedded Windows and embedded Linux are both getting their asses kicked up and down the block by WindRiver's VxWorks, so this whole thing is like watching a couple sparrows fighting over a crumb that the crow dropped.
By itself this isn't much to be concerned with, and I agree with your statement. Unfortunately, MS seems to be targetting Linux at as many levels as possible to eliminate, before it's too late, any large-scale business interest it may have or be acquiring. The bickerings over the crumb spill over into the other arena's where comparisons between Linux and MS OSs are made.
Ian Woods
IANAL and very far from being one, but...
In UK patent law there is an 'obviousness' clause - if a patent claim is obvious to a reasonable proportion of people trained in a suitable field then it doesn't stand as a patent claim.
I'd argue that the major patents these two organisations are trying to bandy about are obvious not only to softies, or people who work on digtal video or the TV industry, but obvious to pretty much everyone with an IQ over 110.
US patent law doesn't have such a clause that I am aware of (but, as stated before, IANAL and definately not an American one) and so it's possible to patent something blindingly obvious and throw lawyers at anyone who does the same, obvious thing.
It seems a tad on the silly side to me...
Ian Woods
Funny how with the windows instances it was only necessary to say "Windows crashed, it sucks" but where there was a linux problem the whole thing was grossly rationalized.
I've seen this as well, but then... there is another aspect to it. When windows BSODs... I almost never know why. It's always a random death. The times when I have seen kernel panics or faults under Linux systems they're generally because I have done something insanely stupid and I can tell what I've done wrong. I don't need to 'rationalise' - I can say that if I try and boot Linux with this thing wrong then it whines...
Shame you just can't do the same thing with windows...
Ian Woods
Why would you want to run fibre? Let's just run through some of the reasons why...
:)
Fibre doesn't suffer many of the EM interferance problems that electrons-in-wires do. Useful, in a high noise environment...
...then again, my computer doesn't crash when I turn a light on nearby when the case is off. Not running parallel to AC-power carrying wires is a must though - long parallel lengths will introduce crosstalk. Just don't do it, or when you do maximise the distance between the power cable and the data cable (or use data cable).
The transmission length of a fibre is much greater than that of a wire at high speeds...
...but then, how big is your house? 100BaseT ethernet over Cat5 gets you 100m typically... is your house so large that it won't reach? If you need more bandwidth, then Gigabit ethernet over balanced copper gets you 25m... but if you really need that then compression will have gone out of fashion.
Reasons why not...
Fibre is expensive! (Copper wire is cheap.)
One thing which is also worth noting is that 10BaseT and 100BaseT only use 2 pairs (4 wires), and CAT5 has 8. This means you can put one 10/100BaseT stream and still have 2 pairs left for 2 audio/phone lines, a second ethernet stream or whatever.
This is obviously a contentious issue, and isn't flamebait on purpose (though it is perhaps possible that a fibre-in-the-home advocate might see it as such). IMO, copper wire will be around for a very long time - it's actually not too bad as long as you don't want to move data over to the other side of town....
Ian Woods
I bet the LCD manufacturers would be eager to supply them if so.
I doubt that. How many LEDs are bought on a day to day basis? How many do you have in your home, your work, the shop you use for supplies and so on and so forth. There's a lot of LEDs in the world - and more every day.
The increased revenue from making a few million LED 'bulbs' is hardly worth the effort of developing it - especially since they won't burn out for a long time.
Ian Woods
Since that is not the case, something must be done either to dechipher individual brains, or as you suggest make different brains look alike. But a single interface would not solve the problem, the interfaces to the external interface would still be higly variable.
./ anyway? :P ), but I'd expect that a Meeting-The-Problem-Halfway(TM) approach would be successful: don't expect the brain to work with the interface, or vice versa - let at least one of them adapt. The adaptability of even adult brains shouldn't be overrated (but perhaps not up to the level required): the good old 'walk around with vertically inverting glasses, and after a week or so you're brain with 'adapt' to see the world correctly' experiment has shown some significant adaptation.
:) (I can now hold the Big Mac with both hands, /and/ hold the carton underneath to catch the lettuce that falls out the other side!)
The 'interface' of the huge bundle of neurons that makes up the optic nerve is not the same for each individual, but almost all individuals can see. I expect this is similar for all of the interfaces which humans have (the development of motor-control is very interesting in young humans - 5 year olds can't write well not because they don't know what shapes to draw but because they don't have the fine control required for such delicate work).
I am not a neurologist and my understanding of the brain is very superficial (how many neurologists read
Certainly, neurological interfaces are one step closer - but for now I'm more interesting in finding nerves which I can control, or learn to control which aren't necessarilly in my brain! That extra limb can come in handy
Ian Woods
Speech recognition is a pain in the ass. They almost certainly just want to use telephone tones.
Very large dictionary VR is still a little unpredictable, even when they're trained to one specific person. For limitted dictionary (say, a few dozen words), speaker independent(*) VR is an awful lot more successful. One such project was on slashdot a long time ago and has been in development since - Sphinx.
As for picking out spoken numbers, back in the days when I was figuring out how to do simple VR I trained myself to the point where I could recognise a number (between 0 and 30 or so) by only looking at a graph of it's waveform - creating a markov model for that purpose would get results as good if not better.
(* This depends, in Sphinx's case, on the quality of the language model. A language model well suited for recognising mid-west American accents would work poorly on Icelandic... for obvious reasons.)
Ian Woods
But surely we could agree on some standard - a set of benchmarks evaluating the machine's performance at representative tasks.
/real/ performance of a device is either difficult or impossible to get hold of easily. I could analyse a number of hardware platforms in great detail (say, 4 or 5 days research each) and get a reasonable idea of their suitability for this task, but it's an awful lot of effort. In some cases, the information is simply not available (or if it is, it's hard to find... anyone used Intel's website recently? I was told by one of their guys to use google to find what I wanted on their site because they didn't know where it was...)
The problem is defining what a representative task is. The requirements of even simple computation varies so much that no matter what 'tasks' you perform there will always be some algorithm which doesn't have any parrallel task against which you've benchmarks.
I've already thrown out every single benchmark I've found for one task I'm doing - the requirements of the algorithm are nothing like sorting a spreadsheet, or searching in a database.
The problem is that the information on
Ian Woods