And, NASA was mostly all engineers -- good ones. Now it's mostly PhDs. This is a big difference when it comes to actually accomplishing something. An engineer solves several problems a week, and writes reports about them -- all in the same week. A PhD has solved one problem, took a few years, then took another few years to write the report. And oh yeah, his solution doesn't have to work outside the lab. As a result of working with ex-NASA employees (the good engineers who got chased out by the academic snobbery) I found the corporate culture to be pretty sick in recent (some years ago) days. Gosh, this IS rocket science, and some of it is dangerous (work out how many horsepower hours it takes to put a car into orbit, with 100% efficiency -- it's one heck of a bomb those guys ride), but they are too timid to admit that surely some folks will die playing with it. It seems China has a more healthy outlook here, and might go somewhere with it.
Of course, if the academics weren't eating every last dime of the appropriations to "study stuff that can't be checked or proved", there might be money to get the job done, as there was last time.
It's profitable to remember that these super smart academics missed Mars by failing to know the difference between metric and English units. Of course they are scared to attempt something most perceive as "simple". They'll want to study it for the rest of their careers and pass the problem to the next guys.
Office simply loads most of the necessary dll's at boot time, which slows down boot, but makes office snappy -- only a relatively small exe must load at that point. This is probably a little harder to do in a cross platform piece of code, and the moral implications of this I leave to the reader. Most serious MS developer types have the tools to confirm this, I did when I was one of them. Gave them up at win2k and devstudio 6. Everything after that is intended to make us run in place and have to learn new frameworks while they continue to use the old stuff themselves to "innovate". Viva Linux!
Yes, we are losing our culture to IP vampires. But there IS a possible solution. It is easy to state:
If you have a copyright, but don't publish the material you have copyrighted, then you lose it. Now.
Sadly, probably never happen.
I'm fretting right now about a huge collection of records, tape, and wire recordings of old stuff I'm not legally allowed to copy for the use of people who lived during that time and who collected it (bought it) in the first place. I cannot buy a new copy in any media from the copyright holders, for any price (and I HAVE asked). In some cases I have better copies or restorations than they do, as I care about history, rather than simply bucks/sales.
This is highly stupid. We have the best laws money can buy. It's getting to be too obvious. It's not like it would cost the copyright owners anything to permit new copies of stuff they won't even sell anymore at all.
Last time everyone said "go look at mars, it's as close and bright as it is going to get in your lifetime", we did. We used a 10" Meade telescope, bought an adapter for a Nikon Coolpix 990 camera, and got...pictures of an orange/brown sphere. This was pretty disapointing until all the astro mags came in with an apology for getting us out there during a planet wide dust storm. We accurately saw that there was nothing to see. Fun star party, though.
As a published author myself, I have pretty much the same take on this. No grammar checker is good enough to be worth turning it on. Sadly, I've had the experience of having a book "edited" by some inexperienced person who thought that some grammar checker was god. Each change they made I had to justify changing back to something like what I'd written, and in many cases they had changed "true" to false, because they used the "tool" to avoid having to actually understand what had been written. Just try to get a good pun past one! They are just too simplistic at this time to be worth any serious condsideration by anyone who loves the language.
Lots of folks have gone off about using nanotubes and such to replace the fiber in "fiberglass" and all tests of bulk material properties (so far) that I'm aware of have shown essentially no net gain. So far, it seems, you can't make the nanotubes and such long enough to effectively be gripped by the binding component to do any good whatsoever. Or cross link the molecules well enough to do away with the need for the binder's strength. Else we'd already have things like fighter planes and cars made of them. So far, no good. This is not to say it won't happen. And yes, steel is neat stuff, but way out of it's league on this job.
Hey, most plant companies make plants -- but not ones that reduce pollution. As an organic gardener amongst other things, I like it that I can get pretty much what I want in certain areas -- and I only mildly complain that some of it is hybrid and won't breed true so I can save seeds. That's a lot of bother that's rarely worth it. But this is a new thing, and a good direction, assuming it's truly an improvement. As a sometimes "farmer" I never thought oxides of nitrogen raining down on my garden were a bad thing, since otherwise I'd have to pay for them as fertilizer in some form, whether compost or chemical. But I live in the sticks, too, where pollution isn't yet a problem. We are in fact already paying to reduce nitrogen oxides, as our auto engines are mandated to be low compression, which means lower thermodynamic efficiency (poorer gas milage) to reduce nitrogen oxides in the first place. Although I'm dreaming here, it would seem a good thing for the planet to solve this in some way that didn't mandate greater use of fossil fuel. Hope this is the first of many. After all, plants can make more plants without our help, they have a lot of gain in effort over machines that don't self replicate.
Here's a post on/. that pretty much mirrors what I think --
the reason I read through all the tripe is that there are some real folks there too. On this issue, the "conservative" talk show guys are just totally off base -- and it discredits them to the point I have trouble listening to some of the rest of stuff they say. They don't respond, of course, to emails with links documenting the good science that has been done by just about every country in the world, just repeat the formulaic junk "there is no proof that mankind is causing this" -- which last year or two was "there is no proof it is happening". I think it might be instructive to follow the money on that. Sort of like those "independent studies" paid for by Microsoft.
You may quote me.
Here's the post:/*
Next to "abortion", saying "global warming" is the quickest way to fire up a troll fest. What's sad is everyone's so busy arguing they've missed the point. On the left, you've got chicken little screaming the sky is falling. On the right, they've stuck their head in the sand.
Hey lefties: calm down, New York isn't going to be washed away tomorrow and screaming your head off about "catastrophic changes" just makes you come off as wack jobs.
Oh and you righties: tell me how you can have 6 billion people on a single planet without affecting it somehow? You haven't noticed we've cut down a couple trees, paved some highways, and shot some bunnies?
Sadly, instead of managing our impact on the planet, we've let the extremists sink us into a troll fest.
*/
To which I (Doug) would add:
You lefties:
You went off about this long before there was any good science. You got lucky. That's all. Doesn't make you right about everything else -- even you disowned the fringe that first brought it up. And it's climate *change* that matters in a world where transportation is getting expensive, everything that isn't a farm is a city (there's almost no wilderness anymore), and whatcha gonna do about it when it's your house/backyard getting torn down now because it's the only place left to farm with decently doable adaptations of the varieties you can get/plant? NIMBY as usual? (puns intended) If this is not what you want, you'd better get enough of the rest of your act together to have an effect. It's obvious the righties are deliberately ignoring something everyone can observe for themselves.
To you righties:
Have you not noticed that the weather for the cities is ALWAYS 5-10 deg warmer then the outlying areas in the forecasts (and the actualities)?
Do you suppose that cities automatically make the outlying areas cooler by some magic? Does it matter what causes it when it's going to kill you, or should it matter more what can be done about it? It's everyone's ox that's gonna get gored on this at some point. Not just your precious energy co. holdings (And look at what they DO, rather than what they say. My solar system is built almost entirely out of parts from companies owned by the bigs in petroleum -- they know the music's gonna stop, and are investing in extra chairs, and I thank them for that). Those of us paying attention are not trying to "destroy capitalism", but to save it. I kind of like it when it works.
To all:
What shortsightedness/vanity made everyone assume it would all be the same forever -- regardless of man's effects? And what about outside influences? Or even continuously improve due to our efforts (or despite them)? We've built an entire world on that assumption-set, which the most beginner scientist or historian knows is totally bogus -- man makes this mistake over and over, and civilizations pay horrible consequences over and over, too. We should be on the edge of an ice age. Does it look like that to anyone? (according to the most recent periodicity of ice ages ~26,000 yrs for the last 4-5 cycles, we should be at the top of slowing warming now, and going down soon -- the temps should not be rising fast now, but leveling off, at the top of a sine curve). Well, it's going to save me some heatin
Nice thing about a pulse of light is that it can be made to reach lots of places at the same time, or nearly so. Just a normal burst of light from a point source has a spherical wavefront, but this can be modified by optics in various ways. Having designed plenty of really fast stuff and having had to deal with skew problems, I can see the advantadge, if real use can be made of it. I think it might even be possible on silicon, which would be required for quick adoption -- after all, the LSI only has to receive, the clock light source can be made of anything. Making a hybrid of course drives costs way up, though. but at current profit margins for fast cpu's this may not be much of a real issue.
So, a good test of does this lead to new insights
would be him showing us how to do digital signal processing or the like with the new system, and see if anything improves, right? Pretty hard to compute spectra without trig functions, although there are other orthagonal basis that do other interesting things. I don't see how this improves on the other and more useful usages, other than making it a bit easier to pass some test in high school.
Hope he does rat them out. These low level guys aren't the problem -- if you have money, you can always find someone with low morals and some skills (I run a business and look for BOTH to be good in a potential hire. This means it's hard to find people I want to work with.) It might be more interesting to find out where the real problems come from. Ok, I'm cynical and think I know already...It would still be good to have the sources of this publicly confirmed.
Interesting that up until recently, what I call the Kung Fu effect has more or less protected this fragile network infrastructure. I know plenty of people who could bring the whole thing down, but they love it being up, and like a real martial arts master, have nothing to prove, so they don't go around beating people up gratuitously.
Bruce Schneier (counterpane.com) has published on and linked to a lot of other publications on the implications of biometrics, and how easy they are in general to steal. Can't just change your password, you've only got 10 fingers (I hope!) and so on. The whole thing is a very bad idea, and most extant schemes are trivially cracked no matter how "secure" the backend. Pictures of retinas/faces have worked, lifted fingerprints translated to gummy silicone have worked, and so forth. No fancy skillz needed to get past any existing system.
Yes, and when little Susie searches for pictures from her digital camera, and finds Dad's pron, then what? I predict this idea will die _very_ quickly.
No way they thought it through. I rather prefer the linux model, where if I set it up that way, other users can't even find out I am there as another user, much less look through my stuff. Much more secure by design.
I have a few pIII 1.13 tualitins here, and love them. For approximately 10W you get 80% of the performance of the fastest pIV, as measured on our own math/memory benchmarks. On solar power, this really makes a difference in what it costs to run our lan. And, oh yeah, with the right heatsink, no fan, or a fan with a 30 ohm resistor in series is enough, and they are nice and quiet. We DO have a beowulf...of these, and it rocks. Running Fedora Core 2 at the moment.
http://clab.mystarband.net
Would be nice if it were that simple, but a nice trick used throughout GOOD c++ programming is to allocate, say, an MFC dialog on the stack, as this automatically solves many memory leak problems and is tons faster than new. Much good software would be broken by disallowing execution from the stack. Of course, stack overflow exploits are the fault of the programmer, not the language, as it's trivial to use the length checking versions of strxxx and so on, and they've been around for quite awhile. And bad code can be trivially found by a text search...
We finally started a full switch to Linux because of service packs to win2k that broke our old DOS CAD tools in the name of security -- and started "phoning home" like XP. OK, allowing full access to the hardware by old DOS programs is a potential security problem, but it would seem that in these days it would be easy to do an emulated sandbox for such that would not be allowed to do serious damage, rather than just making it impossible to run old programs. Win 3 to 9x allowed this via VXD's but this is lost in the more "advanced" versions. For example, TraxEdit (the old version of the now prohibitively expensive Protel toolset) is broken on recent Windows versions that have been patched. But even expense aside, we like these lean, mean tools written to run well on a 286 far better than the new stuff. Imagine how nice they are on a pentium!
I've done hard real time in NT and various other Windows versions. Whether you can do it depends somewhat on the peripheral hardware and of course the needs. One very HiFi soundcard I designed and built had plenty of ram (256kx32) to have plenty of input and output buffering, for example. This card didn't need or use interrupts or dma at all, and even the opsys going to sleep for awhile (the pentium pause we called it) didn't affect the correct flow of full duplex audio through the card. You just have to design things correctly. Smart peripherals rock if done right. The host CPU didn't have to respond audio sample by sample, but in blocks of 4k stereo samples, which took about 2% of it's time to transfer on the ISA bus using rep outsd
or rep insd. The on card cpu took care of keeping all the realtime sample by sample timing straight.
With several seconds of buffering possible, we never had problems, even on 486 machines.
Must have been cast by the very few congress critters who actually had a chance to read the bill before voting. At the time, they were in their own panic over the anthrax thing, and had no copies to read. The fact that they held the vote anyway is so irresponsible it's hard to believe, but it seems that some of them now are going to use that as an excuse for voting for it. Don't let them get away with it!
And break other stuff, like some of my good old DOS cad apps -- say TraxEdit, for laying out PC boards, broken by win2k sp2. In the name of security, they get it wrong again, taking the easy way out, just disable it all. Security is hard to get right, and they just don't get it. Sure, I could trade in my copy of EasyTrax or whatever for the now $10k+ Protel toolset, which does nothing new I need. Oh boy, what a deal. Luckily, as we convert and help several large (500+ machines) networks convert to Linux, it seems most of our major interests (compilers, cad tools and the like) are suddenly getting Linux support. Hmmm, things might be going better than some of the numbers show, at present. And oh yeah, we've found win2k machines doing unauthorized "phone home" after some of the more recent service packs in our firewall logs. Gotta get rid of those last few on our network. We do have a machine here running NT4, which came with our old DevStudio subscription, about SP3 or so, and it's been totally reliable (behind a really serious firewall, no browsing or email on this one) for many years. Later service packs caused trouble.
I was there then, working for the gov't at a _very_ high level. Reagan's play on this was truly inspired, even though he was no particular friend of mine (get all these hippie hackers off the budget and buy more bullets, was his line, to paraphrase). But it worked. The wall fell. Only an actor could have pulled this off, and there's more to tell than I can, using my real name, but it was really good. Besides, SDI kept some very important scientific/engineering teams together with something to do when there was basically nothing to do. But as we've found in the past, just disbanding them has bad results later when you suddenly realize you need more science or engineering. It was a good play. It worked. And, like most successes of this nature, the full truth won't come out soon, that only happens with the big failures. You might get a clue by reading Aviation Week for the period in question...
See http://clab.mystarband.net This one powers my *electronix business*, a smaller one does the house. Been there, done that. You just can't be wasteful. However, the business has a complete machine shop, chemistry that does electroplating/anodizing, welding, the usual stuff. See for yourself. It works, and has been since we went off the grid in about 1982. Zero power failures are a bonus if you're running a 12 machine network as we are. You just put in the money up front, instead of at ruinous rates to ensure someone else's profit. This one's already paid for itself. A couple times over.
The problem isn't global warming, per se -- some like it hot. The problem could be better described as climate change. Sure the Earth's been through many cycles, but none where we were trying to have a technology-based civilization at the time, with food production concentrated in small areas, and the rest as cities/suburbs. All it would take to create major problems would be a major change in the pattern of rainfall. No one's going to want to tear down, say, New York, just because the climate there is suddenly good for growing crops, while California's went too dry and hot for that. And oak trees take a long time to migrate. Sure, the race will survive, but it might not be with as much fun as it could have been.
I own a solar powered business and a solar powered house, and I think this thing is retarded and overpriced. The numbers quoted seemed like they had to be just for the solar part, not the whole thing. 42 panels? We use 16 for the business, and 10 for the house, and either system can back the other up. Why waste energy converting to and from hydrogen (it's nowhere near 100%) when you can just use the electricity as it comes in, saving only a little for nightime use in whatever sort of batteries you favor? PV panels are EXPENSIVE, but worth it if you don't waste the power. This design was obviously motivated by where the designer works. He's got a hammer, and now everything looks like a nail. I wouldn't want to be around when that hydrogen-embrittled storage tank goes up. A better choice of battery for lots of reasons will be the redox Vanadium Pentoxide cells. These store energy in the electolyte, which can be stored in tanks for "infinte" capacity, and they cost a lot less than fuel cells, because they don't need a fancy precious metal catylist. These are already being used as factory-wide UPS systems in Japan.
Slashdot Mirror
And, NASA was mostly all engineers -- good ones. Now it's mostly PhDs. This is a big difference when it comes to actually accomplishing something. An engineer solves several problems a week, and writes reports about them -- all in the same week. A PhD has solved one problem, took a few years, then took another few years to write the report. And oh yeah, his solution doesn't have to work outside the lab. As a result of working with ex-NASA employees (the good engineers who got chased out by the academic snobbery) I found the corporate culture to be pretty sick in recent (some years ago) days. Gosh, this IS rocket science, and some of it is dangerous (work out how many horsepower hours it takes to put a car into orbit, with 100% efficiency -- it's one heck of a bomb those guys ride), but they are too timid to admit that surely some folks will die playing with it. It seems China has a more healthy outlook here, and might go somewhere with it. Of course, if the academics weren't eating every last dime of the appropriations to "study stuff that can't be checked or proved", there might be money to get the job done, as there was last time. It's profitable to remember that these super smart academics missed Mars by failing to know the difference between metric and English units. Of course they are scared to attempt something most perceive as "simple". They'll want to study it for the rest of their careers and pass the problem to the next guys.
Office simply loads most of the necessary dll's at boot time, which slows down boot, but makes office snappy -- only a relatively small exe must load at that point. This is probably a little harder to do in a cross platform piece of code, and the moral implications of this I leave to the reader. Most serious MS developer types have the tools to confirm this, I did when I was one of them. Gave them up at win2k and devstudio 6. Everything after that is intended to make us run in place and have to learn new frameworks while they continue to use the old stuff themselves to "innovate". Viva Linux!
Yes, we are losing our culture to IP vampires. But there IS a possible solution. It is easy to state: If you have a copyright, but don't publish the material you have copyrighted, then you lose it. Now. Sadly, probably never happen. I'm fretting right now about a huge collection of records, tape, and wire recordings of old stuff I'm not legally allowed to copy for the use of people who lived during that time and who collected it (bought it) in the first place. I cannot buy a new copy in any media from the copyright holders, for any price (and I HAVE asked). In some cases I have better copies or restorations than they do, as I care about history, rather than simply bucks/sales. This is highly stupid. We have the best laws money can buy. It's getting to be too obvious. It's not like it would cost the copyright owners anything to permit new copies of stuff they won't even sell anymore at all.
Nuff said?
But that wouldn't occur to them, would it? Wouldn't take too long. Either to write or to run.
Last time everyone said "go look at mars, it's as close and bright as it is going to get in your lifetime", we did. We used a 10" Meade telescope, bought an adapter for a Nikon Coolpix 990 camera, and got...pictures of an orange/brown sphere. This was pretty disapointing until all the astro mags came in with an apology for getting us out there during a planet wide dust storm. We accurately saw that there was nothing to see. Fun star party, though.
As a published author myself, I have pretty much the same take on this. No grammar checker is good enough to be worth turning it on. Sadly, I've had the experience of having a book "edited" by some inexperienced person who thought that some grammar checker was god. Each change they made I had to justify changing back to something like what I'd written, and in many cases they had changed "true" to false, because they used the "tool" to avoid having to actually understand what had been written. Just try to get a good pun past one! They are just too simplistic at this time to be worth any serious condsideration by anyone who loves the language.
Lots of folks have gone off about using nanotubes and such to replace the fiber in "fiberglass" and all tests of bulk material properties (so far) that I'm aware of have shown essentially no net gain. So far, it seems, you can't make the nanotubes and such long enough to effectively be gripped by the binding component to do any good whatsoever. Or cross link the molecules well enough to do away with the need for the binder's strength. Else we'd already have things like fighter planes and cars made of them. So far, no good. This is not to say it won't happen. And yes, steel is neat stuff, but way out of it's league on this job.
Hey, most plant companies make plants -- but not ones that reduce pollution. As an organic gardener amongst other things, I like it that I can get pretty much what I want in certain areas -- and I only mildly complain that some of it is hybrid and won't breed true so I can save seeds. That's a lot of bother that's rarely worth it. But this is a new thing, and a good direction, assuming it's truly an improvement. As a sometimes "farmer" I never thought oxides of nitrogen raining down on my garden were a bad thing, since otherwise I'd have to pay for them as fertilizer in some form, whether compost or chemical. But I live in the sticks, too, where pollution isn't yet a problem. We are in fact already paying to reduce nitrogen oxides, as our auto engines are mandated to be low compression, which means lower thermodynamic efficiency (poorer gas milage) to reduce nitrogen oxides in the first place. Although I'm dreaming here, it would seem a good thing for the planet to solve this in some way that didn't mandate greater use of fossil fuel. Hope this is the first of many. After all, plants can make more plants without our help, they have a lot of gain in effort over machines that don't self replicate.
Here's a post on /. that pretty much mirrors what I think --
the reason I read through all the tripe is that there are some real folks there too. On this issue, the "conservative" talk show guys are just totally off base -- and it discredits them to the point I have trouble listening to some of the rest of stuff they say. They don't respond, of course, to emails with links documenting the good science that has been done by just about every country in the world, just repeat the formulaic junk "there is no proof that mankind is causing this" -- which last year or two was "there is no proof it is happening". I think it might be instructive to follow the money on that. Sort of like those "independent studies" paid for by Microsoft.
You may quote me.
Here's the post: /*
Next to "abortion", saying "global warming" is the quickest way to fire up a troll fest. What's sad is everyone's so busy arguing they've missed the point. On the left, you've got chicken little screaming the sky is falling. On the right, they've stuck their head in the sand.
Hey lefties: calm down, New York isn't going to be washed away tomorrow and screaming your head off about "catastrophic changes" just makes you come off as wack jobs.
Oh and you righties: tell me how you can have 6 billion people on a single planet without affecting it somehow? You haven't noticed we've cut down a couple trees, paved some highways, and shot some bunnies?
Sadly, instead of managing our impact on the planet, we've let the extremists sink us into a troll fest.
*/
To which I (Doug) would add:
You lefties:
You went off about this long before there was any good science. You got lucky. That's all. Doesn't make you right about everything else -- even you disowned the fringe that first brought it up. And it's climate *change* that matters in a world where transportation is getting expensive, everything that isn't a farm is a city (there's almost no wilderness anymore), and whatcha gonna do about it when it's your house/backyard getting torn down now because it's the only place left to farm with decently doable adaptations of the varieties you can get/plant? NIMBY as usual? (puns intended) If this is not what you want, you'd better get enough of the rest of your act together to have an effect. It's obvious the righties are deliberately ignoring something everyone can observe for themselves.
To you righties:
Have you not noticed that the weather for the cities is ALWAYS 5-10 deg warmer then the outlying areas in the forecasts (and the actualities)?
Do you suppose that cities automatically make the outlying areas cooler by some magic? Does it matter what causes it when it's going to kill you, or should it matter more what can be done about it? It's everyone's ox that's gonna get gored on this at some point. Not just your precious energy co. holdings (And look at what they DO, rather than what they say. My solar system is built almost entirely out of parts from companies owned by the bigs in petroleum -- they know the music's gonna stop, and are investing in extra chairs, and I thank them for that). Those of us paying attention are not trying to "destroy capitalism", but to save it. I kind of like it when it works.
To all:
What shortsightedness/vanity made everyone assume it would all be the same forever -- regardless of man's effects? And what about outside influences? Or even continuously improve due to our efforts (or despite them)? We've built an entire world on that assumption-set, which the most beginner scientist or historian knows is totally bogus -- man makes this mistake over and over, and civilizations pay horrible consequences over and over, too. We should be on the edge of an ice age. Does it look like that to anyone? (according to the most recent periodicity of ice ages ~26,000 yrs for the last 4-5 cycles, we should be at the top of slowing warming now, and going down soon -- the temps should not be rising fast now, but leveling off, at the top of a sine curve). Well, it's going to save me some heatin
Nice thing about a pulse of light is that it can be made to reach lots of places at the same time, or nearly so. Just a normal burst of light from a point source has a spherical wavefront, but this can be modified by optics in various ways. Having designed plenty of really fast stuff and having had to deal with skew problems, I can see the advantadge, if real use can be made of it. I think it might even be possible on silicon, which would be required for quick adoption -- after all, the LSI only has to receive, the clock light source can be made of anything. Making a hybrid of course drives costs way up, though. but at current profit margins for fast cpu's this may not be much of a real issue.
So, a good test of does this lead to new insights would be him showing us how to do digital signal processing or the like with the new system, and see if anything improves, right? Pretty hard to compute spectra without trig functions, although there are other orthagonal basis that do other interesting things. I don't see how this improves on the other and more useful usages, other than making it a bit easier to pass some test in high school.
Hope he does rat them out. These low level guys aren't the problem -- if you have money, you can always find someone with low morals and some skills (I run a business and look for BOTH to be good in a potential hire. This means it's hard to find people I want to work with.) It might be more interesting to find out where the real problems come from. Ok, I'm cynical and think I know already...It would still be good to have the sources of this publicly confirmed. Interesting that up until recently, what I call the Kung Fu effect has more or less protected this fragile network infrastructure. I know plenty of people who could bring the whole thing down, but they love it being up, and like a real martial arts master, have nothing to prove, so they don't go around beating people up gratuitously.
Bruce Schneier (counterpane.com) has published on and linked to a lot of other publications on the implications of biometrics, and how easy they are in general to steal. Can't just change your password, you've only got 10 fingers (I hope!) and so on. The whole thing is a very bad idea, and most extant schemes are trivially cracked no matter how "secure" the backend. Pictures of retinas/faces have worked, lifted fingerprints translated to gummy silicone have worked, and so forth. No fancy skillz needed to get past any existing system.
Yes, and when little Susie searches for pictures from her digital camera, and finds Dad's pron, then what? I predict this idea will die _very_ quickly. No way they thought it through. I rather prefer the linux model, where if I set it up that way, other users can't even find out I am there as another user, much less look through my stuff. Much more secure by design.
I have a few pIII 1.13 tualitins here, and love them. For approximately 10W you get 80% of the performance of the fastest pIV, as measured on our own math/memory benchmarks. On solar power, this really makes a difference in what it costs to run our lan. And, oh yeah, with the right heatsink, no fan, or a fan with a 30 ohm resistor in series is enough, and they are nice and quiet. We DO have a beowulf...of these, and it rocks. Running Fedora Core 2 at the moment. http://clab.mystarband.net
Would be nice if it were that simple, but a nice trick used throughout GOOD c++ programming is to allocate, say, an MFC dialog on the stack, as this automatically solves many memory leak problems and is tons faster than new. Much good software would be broken by disallowing execution from the stack. Of course, stack overflow exploits are the fault of the programmer, not the language, as it's trivial to use the length checking versions of strxxx and so on, and they've been around for quite awhile. And bad code can be trivially found by a text search...
We finally started a full switch to Linux because of service packs to win2k that broke our old DOS CAD tools in the name of security -- and started "phoning home" like XP. OK, allowing full access to the hardware by old DOS programs is a potential security problem, but it would seem that in these days it would be easy to do an emulated sandbox for such that would not be allowed to do serious damage, rather than just making it impossible to run old programs. Win 3 to 9x allowed this via VXD's but this is lost in the more "advanced" versions. For example, TraxEdit (the old version of the now prohibitively expensive Protel toolset) is broken on recent Windows versions that have been patched. But even expense aside, we like these lean, mean tools written to run well on a 286 far better than the new stuff. Imagine how nice they are on a pentium!
I've done hard real time in NT and various other Windows versions. Whether you can do it depends somewhat on the peripheral hardware and of course the needs. One very HiFi soundcard I designed and built had plenty of ram (256kx32) to have plenty of input and output buffering, for example. This card didn't need or use interrupts or dma at all, and even the opsys going to sleep for awhile (the pentium pause we called it) didn't affect the correct flow of full duplex audio through the card. You just have to design things correctly. Smart peripherals rock if done right. The host CPU didn't have to respond audio sample by sample, but in blocks of 4k stereo samples, which took about 2% of it's time to transfer on the ISA bus using rep outsd or rep insd. The on card cpu took care of keeping all the realtime sample by sample timing straight. With several seconds of buffering possible, we never had problems, even on 486 machines.
Must have been cast by the very few congress critters who actually had a chance to read the bill before voting. At the time, they were in their own panic over the anthrax thing, and had no copies to read. The fact that they held the vote anyway is so irresponsible it's hard to believe, but it seems that some of them now are going to use that as an excuse for voting for it. Don't let them get away with it!
And break other stuff, like some of my good old DOS cad apps -- say TraxEdit, for laying out PC boards, broken by win2k sp2. In the name of security, they get it wrong again, taking the easy way out, just disable it all. Security is hard to get right, and they just don't get it. Sure, I could trade in my copy of EasyTrax or whatever for the now $10k+ Protel toolset, which does nothing new I need. Oh boy, what a deal. Luckily, as we convert and help several large (500+ machines) networks convert to Linux, it seems most of our major interests (compilers, cad tools and the like) are suddenly getting Linux support. Hmmm, things might be going better than some of the numbers show, at present. And oh yeah, we've found win2k machines doing unauthorized "phone home" after some of the more recent service packs in our firewall logs. Gotta get rid of those last few on our network. We do have a machine here running NT4, which came with our old DevStudio subscription, about SP3 or so, and it's been totally reliable (behind a really serious firewall, no browsing or email on this one) for many years. Later service packs caused trouble.
I was there then, working for the gov't at a _very_ high level. Reagan's play on this was truly inspired, even though he was no particular friend of mine (get all these hippie hackers off the budget and buy more bullets, was his line, to paraphrase). But it worked. The wall fell. Only an actor could have pulled this off, and there's more to tell than I can, using my real name, but it was really good. Besides, SDI kept some very important scientific/engineering teams together with something to do when there was basically nothing to do. But as we've found in the past, just disbanding them has bad results later when you suddenly realize you need more science or engineering. It was a good play. It worked. And, like most successes of this nature, the full truth won't come out soon, that only happens with the big failures. You might get a clue by reading Aviation Week for the period in question...
See http://clab.mystarband.net This one powers my *electronix business*, a smaller one does the house. Been there, done that. You just can't be wasteful. However, the business has a complete machine shop, chemistry that does electroplating/anodizing, welding, the usual stuff. See for yourself. It works, and has been since we went off the grid in about 1982. Zero power failures are a bonus if you're running a 12 machine network as we are. You just put in the money up front, instead of at ruinous rates to ensure someone else's profit. This one's already paid for itself. A couple times over.
The problem isn't global warming, per se -- some like it hot. The problem could be better described as climate change. Sure the Earth's been through many cycles, but none where we were trying to have a technology-based civilization at the time, with food production concentrated in small areas, and the rest as cities/suburbs. All it would take to create major problems would be a major change in the pattern of rainfall. No one's going to want to tear down, say, New York, just because the climate there is suddenly good for growing crops, while California's went too dry and hot for that. And oak trees take a long time to migrate. Sure, the race will survive, but it might not be with as much fun as it could have been.
I own a solar powered business and a solar powered house, and I think this thing is retarded and overpriced. The numbers quoted seemed like they had to be just for the solar part, not the whole thing. 42 panels? We use 16 for the business, and 10 for the house, and either system can back the other up. Why waste energy converting to and from hydrogen (it's nowhere near 100%) when you can just use the electricity as it comes in, saving only a little for nightime use in whatever sort of batteries you favor? PV panels are EXPENSIVE, but worth it if you don't waste the power. This design was obviously motivated by where the designer works. He's got a hammer, and now everything looks like a nail. I wouldn't want to be around when that hydrogen-embrittled storage tank goes up. A better choice of battery for lots of reasons will be the redox Vanadium Pentoxide cells. These store energy in the electolyte, which can be stored in tanks for "infinte" capacity, and they cost a lot less than fuel cells, because they don't need a fancy precious metal catylist. These are already being used as factory-wide UPS systems in Japan.