Based on the limited experience of atheists in power (mostly 20th century communists who have the highest 'fellow citizens killed' totals and percentages ever) I'd say that there's ample real world reason to exhibit a bit of concern.
I can smell the snake oil from here. Low life expectency patients (read desperate), techniques that have not, apparently been peer reviewed and demonstrated efficacious, this sounds suspiciously like shamanism which, if they're just injecting saline and calling them fetal cells, I don't have much of a problem but if they're actually using the cells to gain advantage of the placebo effect...
Naziism, formally called national socialism, was very much not christian which any serious examination of their belief system would bear out. It's an anti-christian libel to view Nazi denial of jewish humanity as different only in degree with christian fury over the christ-killer libel. Christians were angry with jews over what their ancesters did, Nazis believed that inherently the jews were subhuman. And christians, unlike nazis, have doctrines of love and forgiveness that tended to ameliorate anger.
It was the denial of humanity that allowed all those medical experiments to be done as if jews, homosexuals, gypsies, and the other inmates of the death camps were just animals that could be used as means to nazi ends.
As for religious restraint doing more good than harm, how do you determine the good and harm of an experiment that was not run? Medical ethics boards don't tend to trumpet to the public the immoral ideas their staff come up with that they shoot down. As an alternative, I'd look at the history of immoral scientific experiments that could have used a bit more moral supervision. I'd suggest a little more restraint on the part of they doctors who refrained from treating those black syphilis patients with more than a placebo just so they could record 'what would happen' would have been a good thing.
The problem of a Mengele appearing is not theoretical but historical. The example I gave was all too real and the research data was under embargo for decades after WW II. From a purely scientific point of view this was and is bad science.
Ok, 200 million limits you to about 10-20 of the richest people on earth, or millions of small investors who put in a small share of the risk. I'm astonished that for both projects there isn't some small investor interface (perhaps an ASP?) that would allow low transaction cost investments of $1 and up. I know that every time somebody ran a story they'd get a boost in investment with a steady stream from the committed (perhaps via small payroll deductions that go into the investment fund).
You don't need high-cost salesman making big pitches to get large checks anymore but nobody seems to have figured that lesson out. If I could punch in a credit card number and auto-invest $25 a month I would do it. But nobody's giving me that option now are they?
On the space elevator side, I was projecting the discovery after an orbital mining capability was developed. Orbital mining means that you don't have to have 3 years of building *after* the original cable is rocket launched and fed out via a spool but rather you can get the high strength cable built in orbit from the first production run.
Sorry, even a libertarian (as opposed to anarchist) society will have to deal with the question of who or what is a rights bearing being. Artificial intelligences, embryos, the retarded, catatonic, and other border conditions have to be addressed in any society that's as advanced as we are.
Actually, I am not aware of any actual embryonic stem cell treatments that are in clinical use. The hope has been high but hasn't panned out yet. The supposedly less capable adult stem cell therapies *are* out there and curing people without all the moral difficulties embryo therapies present.
Could you give a ref. to an actual embryonic treatment that is in actual use?
Watch out what you ask for. If you get those religious/moral types too far out of science you end up with Mengele reruns throwing jews into freezing water just to measure how quickly they die. It's good science but morally impermissible.
The current situation is that there are a large number of scientists around the world, each playing with carbon nanotubes, trying to solve the problem of limitlessly extending them or splicing them like rope with no loss of strength. The risk for the mining project is that the investors will spend billions based on cost estimates of $10,000/kg to lift to orbit and by the time they're ready to profit, some joker shouts Eureka! in a lab in Boise and suddenly their sunk costs @ $10,000/kg have to be recovered in an economic arena of $100/kg. Take a look at the short duration of the pony express to get an idea of what I mean.
You state that 'a few hundred' kilos of launched mass could bring back 'tens of thousands' of kilos of water to high earth orbit. Let's say you're right and 'a few hundred' kilos = 500. Let's further posit that the actual cost of the stuff is $1M and launch costs are $10,000/kg. That means that for the venture capital cost of a less than one dot bomb ($6M) you could provide a vastly profitable resource extraction opportunity with profit achieved in let's say 2 years.
There are space enthusiasts like John Carmack who can personally write checks for $6M. None of them appear to be stepping up to the plate on this so perhaps things are not so cut and dried as your boosterism implies?
One final observation, current elevator plans have a multiyear building process for the first elevator because it would be too expensive to launch the full elevator cable in one shot. If all you need to launch is a carbon nanotube construction machine, use locally mined carbon (by the elevator group itself) and lower the thing, the time to build could be reduced significantly along with the building cost. Thus a mining infrastructure in orbit would be compelled to slit its own throat. Waste rock byproducts already in earth orbit would be the elevator's raw material.
The entire puzzle is hardware and software. If you have to be locked in on hardware but not on software, pick that solution. At some point, that board is going to fail and if you're locked in via software, the temptation is going to be to evaluate one OS hardware for the next round. If the hardware is all that's keeping you on Windows, you can make Linux compatibility part of the spec for the replacement board and easily switch then. Vendor lock-in is always least secure when it is only one lock. You might not be able to break free now but you should make it as easy as possible to break free later.
Some very expensive equipment is running with MS-DOS drivers even today. Other stuff is being controlled by Windows 3.1. So where is the academic exception to the "we don't support stuff that's over 5 years old" rule that MS has? There is no such exception? Ooops.
Beyond the problem of equipment with a 30 year lifespan running on an OS with a 5 year lifespan there is the problem of MS legal. Have you seen MS license 6.0? MS is clearly pushing for forced updates every couple of years. Then there's the lovely BSA who have the habit of coming in the door with sheriffs and forcing everybody to stop working while they check the net for license compliance. Do you think you're immune because you are in an educational institution? The 6 largest school districts in Oregon got audit notices for the busiest season of the year (around finals). If you think it can't happen to you, you're wrong.
It doesn't take more than 10 minutes to create a project on Source Forge and it doesn't cost anything. Why not chalk it up to advertising for the institution and dump the current version out to the community. You might just find that the benefits will be more than just a little feel good institution promotion.
The problem is that MS software comes with MS lawyers attached and with MS being a monopoly company whose business plan is predicated on hiring the best programmers that ever increasing stock options can buy, those lawyers have been tasked with restricting your rights and increasing MS revenues using any means necessary. That's a train wreck waiting to happen and you don't want to be in the middle of a time sensitive delicate experiment when the sheriff's deputies burst into the lab saying step away from the computers, we have a software compliance warrant and you can't touch anything for the next 2 days while all machines on the net are checked for pirated software.
The problem with MS is not just technical but legal as well and as the technical problems are solved, the legal ones just seem to be getting worse.
That would give you a text file listing every rpm package you have. Anybody out there have a simple way of doing the same thing for Windows? Sure, you can get a list from add/remove software but that list isn't always accurate. I remember one example regarding Internet Explorer where if you upgraded and then downgraded, only some of the files were downgraded though the add/remove software window didn't reflect that reality.
I'm sure that somebody out there can come up with an accurate method but is it as simple or as easy as rpm -qa ?
If you end up needing to use custom kernels or custom packages, converting them to rpm and making them available on the net is not so hard. I'm sure that adherents of other package management systems can come up with similarly easy methods (maybe even easier) for their favorite package managers. Windows would have similar problems if the experimenter used a custom coded dll to obtain results.
Now you bring up responsibility towards your code. As far as I understand it, Microsoft will tell you up front that if it's older than 5 years, MS doesn't stand behind their code. No security or any other bug fixes whatsoever.
With OSS software there are still people maintaining old kernel lines that are older than 5 years. And if the old maintainers decide to pack it in, there's nothing stopping new maintainers stepping up to the needs of the community. it happens all the time. Who can maintain Windows 95 or NT 3.51/NT 4 now that they're end of life and unsupported?
As for suing Microsoft in case of a bad software experience, good luck. They are well protected against such suits. You give away your right to sue when you agree to use the software.
As for commercial software working better than open source, I'd suggest you look at Apache v. IIS usage rates for a clear counterexample. The truth is that OSS software is put out by a variety of people, some committed and excellent, some poorly qualified and lazy. Commercial software is also widely varying in quality and a big name software house is no guarantee that a product they put out will be useful or even well put together. Does the IBM engineer who is working on high availability Linux cease to be a highly qualified professional because his work product is not turned over to a sales arm to be hawked to the company's customers? I don't think so.
I've never seen a $1B estimate for asteroid mining. It would be nice to have references. There's an entire set of infrastructure that you need to maintain to mine and make materials. The starter tools have to come up the gravity well and you haven't even begun to address the comparative cost to mine.
You're wrong on the catch-22 of lift costs. Lift costs will go down by a factor of 100 when they figure out how to make very long carbon nanotubes. That progress is independent of lift demand at the current $10k/kg price tag.
If you want the ability for the VB guy to maintain your code but you don't want to get locked into windows, try RealBasic. The realsoftware folks are mac/windows dual platform and they are aggressively looking to expand their platforms. The last I checked they it was a tossup between Palm and Linux for the next expansion (the loser being platform #4). Sure, you end up running on a Windows box today but you're not going to get locked in forever and unlike the python/libglade stuff being proposed by others, your vb programmer will be able to maintain your code with a book and about a week of ramp up time. For those who do both VB and RB, the consensus seems to be that RB is a faser RAD environment and better all around.
Eventually your $1200 card will die out and you'll either get a new replacement or a newer hand-me-down. Trailing edge everything always has these problems. It isn't as shiny, functional, easy to use, or as cool as the new stuff that the better funded lab in the next building gets. But every purchase of new stuff today is not only affecting you but all those future frustrated scientists who get to deal with your hand-me-downs as well.
The key isn't to just demand more money (though it can't hurt) but to have a strategy for dealing with the fact that lab equipment can last decades but MS has an official company policy of obsoleting OS versions older than 5 years. That fact alone should make alternative OS support a priority because for a 20 yr lifespan instrument, at least three quarters of its useful life will be run without an underlying supported OS if you choose the MS solution route.
Well Apple seems to have the config file problem well in hand by converting everything over to XML. I wish that they would just spend a little money to hire some 2nd/3rd world computer people to go around the sourceforge archives and submit patches to do the same for the whole GNU/OSS collective. If it were little to no effort on the part of the core team, I can't see anybody turning that down.
I just wish that somebody would have stepped up and said to this arrogant fool that unless an NI instrument is the only producer available, NI will 'not be an option'. No doubt he gives that speech to lots of places so somebody else will have the chance to say it.
The vendor needs the customer to survive. The customer needs a solution to survive no matter what vendor provides it. If NI wants to limit its market by limiting its driver support it'll go down the road to irrelevance paved by countless other companies who refused to deliver what customers wanted.
Over the long haul, there are always new entrants to any market who will provide either direct competition to an arrogant incumbent or substitution competition for them. Microsoft is in this position (a huge chunk of their customers are looking for a commercially viable exit from their MS relationship) and it looks like NI is in the same boat.
News flash: If you have lift costs of 1% of current figures, you have the need of 1% as good a sales force to gather up your investment cash.
Maybe you're right and all that's needed is luck and skill and brains to assemble it all into a practical package that can get it rolling and profitable before something better comes along. you've got maybe 5 years. After that, the space elevator will grab the mindshare and everybody will be looking towards orbit.
The efficiency differences between the two methods of usage can be quite large. Here's a decent article on the subject.
The bottom line is for a practical, real world application, the crossover point is somewhere between 800C and 1000C, a level that doesn't generally get reached in a car or home generator. At a much more reasonable 200C, fuel cell efficiency is around 80% while Carnot limited internal combustion engines are only about 30% efficient.
As for electricity transport, you're right, that requires an inverter (which is pretty standard stuff) and there are some efficiency losses there. However, the home generation route is likely to be popular as you could reduce your electric bill by using a fuel cell stack as a natural gas powered water heater and coincidentally, home fuel cell generators would have the fewest demands put on them during the daytime (when industrial demand is high) and could run full blast during power price spikes to aleviate rolling blackouts or brownouts. GE is allied with a company called Plug Power to do exactly that sort of thing with their upcoming HomeGen line of fuel cells.
According to an ABC article I read, Criswell's system costs approximately 135B to get off the ground. That's a huge capital drain for an uncertain payoff as competition to lunar solar could eviscerate the expected profits during that initial decade. A Wired story on the elevator pegs its construction cost at 10B which is both a great deal less initial investment and would provide multiple benefits beyond enabling power satellites, like enabling orbital construction, space tourism, space mining, orbital waste disposal, etc.
It's too expensive to even launch the tools to build the tools to build the stuff. $10,000/kg is very expensive and you have to get the raw materials from somewhere. We don't have the ability to do asteroid mining so where does the raw silicon come from? Face it, launch costs are critical to the viability of doing just about anything in space until we can space mine and that technology is going to be a lot further off than 15 years.
Slashdot Mirror
Based on the limited experience of atheists in power (mostly 20th century communists who have the highest 'fellow citizens killed' totals and percentages ever) I'd say that there's ample real world reason to exhibit a bit of concern.
I can smell the snake oil from here. Low life expectency patients (read desperate), techniques that have not, apparently been peer reviewed and demonstrated efficacious, this sounds suspiciously like shamanism which, if they're just injecting saline and calling them fetal cells, I don't have much of a problem but if they're actually using the cells to gain advantage of the placebo effect...
Naziism, formally called national socialism, was very much not christian which any serious examination of their belief system would bear out. It's an anti-christian libel to view Nazi denial of jewish humanity as different only in degree with christian fury over the christ-killer libel. Christians were angry with jews over what their ancesters did, Nazis believed that inherently the jews were subhuman. And christians, unlike nazis, have doctrines of love and forgiveness that tended to ameliorate anger.
It was the denial of humanity that allowed all those medical experiments to be done as if jews, homosexuals, gypsies, and the other inmates of the death camps were just animals that could be used as means to nazi ends.
As for religious restraint doing more good than harm, how do you determine the good and harm of an experiment that was not run? Medical ethics boards don't tend to trumpet to the public the immoral ideas their staff come up with that they shoot down. As an alternative, I'd look at the history of immoral scientific experiments that could have used a bit more moral supervision. I'd suggest a little more restraint on the part of they doctors who refrained from treating those black syphilis patients with more than a placebo just so they could record 'what would happen' would have been a good thing.
The problem of a Mengele appearing is not theoretical but historical. The example I gave was all too real and the research data was under embargo for decades after WW II. From a purely scientific point of view this was and is bad science.
Ok, 200 million limits you to about 10-20 of the richest people on earth, or millions of small investors who put in a small share of the risk. I'm astonished that for both projects there isn't some small investor interface (perhaps an ASP?) that would allow low transaction cost investments of $1 and up. I know that every time somebody ran a story they'd get a boost in investment with a steady stream from the committed (perhaps via small payroll deductions that go into the investment fund).
You don't need high-cost salesman making big pitches to get large checks anymore but nobody seems to have figured that lesson out. If I could punch in a credit card number and auto-invest $25 a month I would do it. But nobody's giving me that option now are they?
On the space elevator side, I was projecting the discovery after an orbital mining capability was developed. Orbital mining means that you don't have to have 3 years of building *after* the original cable is rocket launched and fed out via a spool but rather you can get the high strength cable built in orbit from the first production run.
MSNBC says the story comes from the Washington Post. MSNBC is full of reprints from other sources.
Maybe you need to RTFA more carefully.
Sorry, even a libertarian (as opposed to anarchist) society will have to deal with the question of who or what is a rights bearing being. Artificial intelligences, embryos, the retarded, catatonic, and other border conditions have to be addressed in any society that's as advanced as we are.
Actually, I am not aware of any actual embryonic stem cell treatments that are in clinical use. The hope has been high but hasn't panned out yet. The supposedly less capable adult stem cell therapies *are* out there and curing people without all the moral difficulties embryo therapies present.
Could you give a ref. to an actual embryonic treatment that is in actual use?
Watch out what you ask for. If you get those religious/moral types too far out of science you end up with Mengele reruns throwing jews into freezing water just to measure how quickly they die. It's good science but morally impermissible.
The current situation is that there are a large number of scientists around the world, each playing with carbon nanotubes, trying to solve the problem of limitlessly extending them or splicing them like rope with no loss of strength. The risk for the mining project is that the investors will spend billions based on cost estimates of $10,000/kg to lift to orbit and by the time they're ready to profit, some joker shouts Eureka! in a lab in Boise and suddenly their sunk costs @ $10,000/kg have to be recovered in an economic arena of $100/kg. Take a look at the short duration of the pony express to get an idea of what I mean.
You state that 'a few hundred' kilos of launched mass could bring back 'tens of thousands' of kilos of water to high earth orbit. Let's say you're right and 'a few hundred' kilos = 500. Let's further posit that the actual cost of the stuff is $1M and launch costs are $10,000/kg. That means that for the venture capital cost of a less than one dot bomb ($6M) you could provide a vastly profitable resource extraction opportunity with profit achieved in let's say 2 years.
There are space enthusiasts like John Carmack who can personally write checks for $6M. None of them appear to be stepping up to the plate on this so perhaps things are not so cut and dried as your boosterism implies?
One final observation, current elevator plans have a multiyear building process for the first elevator because it would be too expensive to launch the full elevator cable in one shot. If all you need to launch is a carbon nanotube construction machine, use locally mined carbon (by the elevator group itself) and lower the thing, the time to build could be reduced significantly along with the building cost. Thus a mining infrastructure in orbit would be compelled to slit its own throat. Waste rock byproducts already in earth orbit would be the elevator's raw material.
The entire puzzle is hardware and software. If you have to be locked in on hardware but not on software, pick that solution. At some point, that board is going to fail and if you're locked in via software, the temptation is going to be to evaluate one OS hardware for the next round. If the hardware is all that's keeping you on Windows, you can make Linux compatibility part of the spec for the replacement board and easily switch then. Vendor lock-in is always least secure when it is only one lock. You might not be able to break free now but you should make it as easy as possible to break free later.
It's not about the money necessarily.
Some very expensive equipment is running with MS-DOS drivers even today. Other stuff is being controlled by Windows 3.1. So where is the academic exception to the "we don't support stuff that's over 5 years old" rule that MS has? There is no such exception? Ooops.
Beyond the problem of equipment with a 30 year lifespan running on an OS with a 5 year lifespan there is the problem of MS legal. Have you seen MS license 6.0? MS is clearly pushing for forced updates every couple of years. Then there's the lovely BSA who have the habit of coming in the door with sheriffs and forcing everybody to stop working while they check the net for license compliance. Do you think you're immune because you are in an educational institution? The 6 largest school districts in Oregon got audit notices for the busiest season of the year (around finals). If you think it can't happen to you, you're wrong.
It doesn't take more than 10 minutes to create a project on Source Forge and it doesn't cost anything. Why not chalk it up to advertising for the institution and dump the current version out to the community. You might just find that the benefits will be more than just a little feel good institution promotion.
The problem is that MS software comes with MS lawyers attached and with MS being a monopoly company whose business plan is predicated on hiring the best programmers that ever increasing stock options can buy, those lawyers have been tasked with restricting your rights and increasing MS revenues using any means necessary. That's a train wreck waiting to happen and you don't want to be in the middle of a time sensitive delicate experiment when the sheriff's deputies burst into the lab saying step away from the computers, we have a software compliance warrant and you can't touch anything for the next 2 days while all machines on the net are checked for pirated software.
The problem with MS is not just technical but legal as well and as the technical problems are solved, the legal ones just seem to be getting worse.
Perhaps he was using it in the sense of free of MS' legal department? That's a definition of free that is significant for a lot of businesspeople.
Redhat Linux method of determining version
rpm -qa > whatImusing.txt &
That would give you a text file listing every rpm package you have. Anybody out there have a simple way of doing the same thing for Windows? Sure, you can get a list from add/remove software but that list isn't always accurate. I remember one example regarding Internet Explorer where if you upgraded and then downgraded, only some of the files were downgraded though the add/remove software window didn't reflect that reality.
I'm sure that somebody out there can come up with an accurate method but is it as simple or as easy as rpm -qa ?
If you end up needing to use custom kernels or custom packages, converting them to rpm and making them available on the net is not so hard. I'm sure that adherents of other package management systems can come up with similarly easy methods (maybe even easier) for their favorite package managers. Windows would have similar problems if the experimenter used a custom coded dll to obtain results.
Now you bring up responsibility towards your code. As far as I understand it, Microsoft will tell you up front that if it's older than 5 years, MS doesn't stand behind their code. No security or any other bug fixes whatsoever.
With OSS software there are still people maintaining old kernel lines that are older than 5 years. And if the old maintainers decide to pack it in, there's nothing stopping new maintainers stepping up to the needs of the community. it happens all the time. Who can maintain Windows 95 or NT 3.51/NT 4 now that they're end of life and unsupported?
As for suing Microsoft in case of a bad software experience, good luck. They are well protected against such suits. You give away your right to sue when you agree to use the software.
As for commercial software working better than open source, I'd suggest you look at Apache v. IIS usage rates for a clear counterexample. The truth is that OSS software is put out by a variety of people, some committed and excellent, some poorly qualified and lazy. Commercial software is also widely varying in quality and a big name software house is no guarantee that a product they put out will be useful or even well put together. Does the IBM engineer who is working on high availability Linux cease to be a highly qualified professional because his work product is not turned over to a sales arm to be hawked to the company's customers? I don't think so.
I've never seen a $1B estimate for asteroid mining. It would be nice to have references. There's an entire set of infrastructure that you need to maintain to mine and make materials. The starter tools have to come up the gravity well and you haven't even begun to address the comparative cost to mine.
You're wrong on the catch-22 of lift costs. Lift costs will go down by a factor of 100 when they figure out how to make very long carbon nanotubes. That progress is independent of lift demand at the current $10k/kg price tag.
If you want the ability for the VB guy to maintain your code but you don't want to get locked into windows, try RealBasic. The realsoftware folks are mac/windows dual platform and they are aggressively looking to expand their platforms. The last I checked they it was a tossup between Palm and Linux for the next expansion (the loser being platform #4). Sure, you end up running on a Windows box today but you're not going to get locked in forever and unlike the python/libglade stuff being proposed by others, your vb programmer will be able to maintain your code with a book and about a week of ramp up time. For those who do both VB and RB, the consensus seems to be that RB is a faser RAD environment and better all around.
Eventually your $1200 card will die out and you'll either get a new replacement or a newer hand-me-down. Trailing edge everything always has these problems. It isn't as shiny, functional, easy to use, or as cool as the new stuff that the better funded lab in the next building gets. But every purchase of new stuff today is not only affecting you but all those future frustrated scientists who get to deal with your hand-me-downs as well.
The key isn't to just demand more money (though it can't hurt) but to have a strategy for dealing with the fact that lab equipment can last decades but MS has an official company policy of obsoleting OS versions older than 5 years. That fact alone should make alternative OS support a priority because for a 20 yr lifespan instrument, at least three quarters of its useful life will be run without an underlying supported OS if you choose the MS solution route.
Well Apple seems to have the config file problem well in hand by converting everything over to XML. I wish that they would just spend a little money to hire some 2nd/3rd world computer people to go around the sourceforge archives and submit patches to do the same for the whole GNU/OSS collective. If it were little to no effort on the part of the core team, I can't see anybody turning that down.
I just wish that somebody would have stepped up and said to this arrogant fool that unless an NI instrument is the only producer available, NI will 'not be an option'. No doubt he gives that speech to lots of places so somebody else will have the chance to say it.
The vendor needs the customer to survive. The customer needs a solution to survive no matter what vendor provides it. If NI wants to limit its market by limiting its driver support it'll go down the road to irrelevance paved by countless other companies who refused to deliver what customers wanted.
Over the long haul, there are always new entrants to any market who will provide either direct competition to an arrogant incumbent or substitution competition for them. Microsoft is in this position (a huge chunk of their customers are looking for a commercially viable exit from their MS relationship) and it looks like NI is in the same boat.
News flash:
If you have lift costs of 1% of current figures, you have the need of 1% as good a sales force to gather up your investment cash.
Maybe you're right and all that's needed is luck and skill and brains to assemble it all into a practical package that can get it rolling and profitable before something better comes along. you've got maybe 5 years. After that, the space elevator will grab the mindshare and everybody will be looking towards orbit.
The efficiency differences between the two methods of usage can be quite large. Here's a decent article on the subject.
The bottom line is for a practical, real world application, the crossover point is somewhere between 800C and 1000C, a level that doesn't generally get reached in a car or home generator. At a much more reasonable 200C, fuel cell efficiency is around 80% while Carnot limited internal combustion engines are only about 30% efficient.
As for electricity transport, you're right, that requires an inverter (which is pretty standard stuff) and there are some efficiency losses there. However, the home generation route is likely to be popular as you could reduce your electric bill by using a fuel cell stack as a natural gas powered water heater and coincidentally, home fuel cell generators would have the fewest demands put on them during the daytime (when industrial demand is high) and could run full blast during power price spikes to aleviate rolling blackouts or brownouts. GE is allied with a company called Plug Power to do exactly that sort of thing with their upcoming HomeGen line of fuel cells.
According to an ABC article I read, Criswell's system costs approximately 135B to get off the ground. That's a huge capital drain for an uncertain payoff as competition to lunar solar could eviscerate the expected profits during that initial decade. A Wired story on the elevator pegs its construction cost at 10B which is both a great deal less initial investment and would provide multiple benefits beyond enabling power satellites, like enabling orbital construction, space tourism, space mining, orbital waste disposal, etc.
It's too expensive to even launch the tools to build the tools to build the stuff. $10,000/kg is very expensive and you have to get the raw materials from somewhere. We don't have the ability to do asteroid mining so where does the raw silicon come from? Face it, launch costs are critical to the viability of doing just about anything in space until we can space mine and that technology is going to be a lot further off than 15 years.