Current inorganic solar cells have around 30% efficiency, and current organic solar cells have around 6% effeciency. However, the price per kilowatt is about the same.
A device with the kilowatt/sq-ft of a silicon panel and the cost/sq-ft of an organic panel would go a long way.
I calculated that in my area, with current technology and current electricity rates, a solar panel investment would break even in ~ 30 years (and this factors in tax incentives). Maybe break-even would occur in only 15-20 years when you factor in future increases in the cost of electricity.
If this new invention would really improve the cost/kilowatt by 5x, then a solar investment could break even in only around 5 years. That would cause a turning point in adoption.
If we're talking about courting over the mass market of non-geek users, what we need is decent free software office apps for Windows. For most users, even if they wanted to use Linux, the time and trouble to swap out their OS would be prohibitive. Windows is already dominant on the desktop, and people tend to stick with what they already know unless something motivates them to change. For this reason, having an exact clone of Windows features and applications will end up doing exactly nothing. This was not true in the server market, where frankly, Windows has never had a dominant market share and *NIX has always exceeded in technology.
What is really needed is decent free software office apps for Windows. Even though their availability isn't by itself a compelling reason for users to change to free software, it doesn't have to be. MS Office is so expensive, and full of so many annoyances, that the open market will force the switch. Consider you are buying a new low/mid end PC for your home. The cost is about $1500 right now, or $2000 with MS office. $500 or 1/3 the cost of the system is a big difference, so much so that major computer manufacturers do not pre-install Office by default on these types of PC's. Before long PC manufacturers will be able to offer users a choice of OpenOffice for Windows at no extra charge. The only advantage of paying the extra $500 for Office would be to share files with other users, but home users for the most part don't share files anyway.
There is always the issue that the availability of free software for Windows could be a reason for users not to switch to Linux. Yet, GNU software was available for commercial UNIX systems long before a free kernel was available. During this time many users and system administrators learned to use and rely upon GNU software. When Linux became available, switching from GNU on a commercial UNIX kernel to GNU on a Linux kernel was not that big a leap. Likewise, once Windows users become accustomed to using inexpensive and powerful free software, the transition to using those same applications on Linux will not seem as difficult.
Bringing up the history of GNU, there was not only the issue of cost at stake, but the fact that GNU software often performed much better or had more compelling features than commercial software from the UNIX vendors. This is equally true with desktop software today. Microsoft's most compelling reason for users to pay to upgrade to their new Office XP software is removing the annoying paper clip character. In open source software, we wouldn't need to pay for upgrades for that, we'd just comment out those lines from the code. There are plenty of annoyance in Office to improve upon in free software.
Once users become familiar with free office software, sooner or later some of them will start installing it at work, at first just to share files with their home systems. As this becomes more widespread, there is pressure on the IT department to offer some sort of support. At first, most IT departments will say they need Office because everyone else has it. But a few innovative ones (and maybe a couple strapped for cash) will switch. And the more that do, the less valid the argument about needing Office to share files with other companies.
And, bang, the #1 reason for not deploying Linux on the desktop will be eliminated.
This will take some time to happen. Just consider how many companies run Apache on commercial UNIX, in spite of the availability of Linux. The period during which both Linux and Windows flourish will be a long one. However, this is how it should be. It allows all the users to enjoy a much smoother transition. And what about Microsoft? As long as they can offer their customers more for their upgrade dollars than just removing Clippy, Microsoft has the potential to do just fine. But for the users out there, it will be a homerun for sure.
I was so accused after submitting a final paper for a liberal arts class I was taking. The professor thought it was "too good" for me to have written it, and said that I must have copied from some other source.
This reminds me of the movie Finding Forrester. Did you, too, get your Pullitzer Prize-winning mentor to come to your aid?:-)
This type of situation happened to me while in college (CS major). After working very hard on an assignment, I received a zero grade. When I went into the grad student who had done the grading, he accused me of cheating. This was quite a shock to me, because I knew I hadn't!
When I asked him to explain, he said that he had a very clever way of detecting cheating. He compiled the programs and ran cmp, and my program had come out the same as someone else's! I had set my umask to allow world-read of my entire set of files, naively thinking that I could trust my classmates. The other student had copied my assignment straight off the disk drive.
Fortunately, I was able to prove I was innocent because - get this - when the student copied my program he didn't bother to remove my name from the comments!
The grader changed my grade but told me in the future he would hold me accountable for allowing others to cheat off of me, even if that wasn't my intention! Ever since then I was much more careful in how I set my file permissions. The sad part was, the person who cheated off of me had been a good friend of mine. I would have gladly offered to help him out (in the honest sense) if he had let me know he was having trouble in the class.:-(
I'll second these words of praise for LISP and its libraries. I remember a college assignment to do RSA encyrption. We weren't given many restrictions: it could be in any language, it had to encrypt/decrypt a short text string, it had to use 128 bit keys, and we could work in groups but didn't have to. We were given about 2 weeks or so.
Most of the class decided to write it in C, but I chose Common LISP. I was the only person in the class to complete the assignment on time, even though I didn't work in a group. LISP helped me for two reasons:
Most of the class spent the entire time writing C functions to handle large integers, but I was able to use the BIGInt library provided as part of the CL distribution.
LISP is so closely tied with the underlying mathematics that I was able to code the algorithm in unbelievably fast time. Furthermore, I was able to spend my free time finding a better algorithm instead of debugging code, which led to my program being much faster than any of the others.
Now if only I had had the opporunity to sell it to Yahoo for $50 Million!...:-)
BLUG, a bold new startup, has demonstrated the first implementation of a revoutionary and unique wireless technology called PigeonNet (tm).
A key advantage of PigeonNet over competing wireless standards, such as Bluetooth and 802.11b, is that it is free from interference with the wide variety of devices operating in the unlicensed 2.4 GHz spectrum. In addition, PigeonNet is environmentally sensitive. It is the only network technology to use only renewable energy sources in the transmission of data.
I had a similar experience a few years ago. A couple of the minor details may have changed in my memory, but it's mostly accurate.
The problem was that my MS Access database got corrupted. Whenever I tried to run a certain report, Access crashed. Unfortunately, I had spent over a week creating this report, and I wanted to recover as much of my work as possible.
I called MS Support, spent a few minutes wading through their phone tree, and then waited on hold about 20 minutes. Me:   My Access database is corrupt. Access crashes whenever I try to run a certain report. I want to get back as much of my report as possible. MS Rep:  Ok, try doing file/open, and pick the name of your database. Me:  Do you want me to kill and restart Access first, because it is crashed. MS Rep:  Huh? Me:  I'll just kill and restart it first. Ok, the database is opened. MS Rep:  And it works, I see. Can you press the "Reports" button? Me:  Ok, I get a list of my reports. Now what? MS Rep:  Can you tell me which reports are on the list? Me:  Sure, here are the reports... MS Rep:  And which one is having problems? Me:  This one... MS Rep:  Ok, can you try to double-click it? Me:  Sure. Ok, it's starting to run. Now I get the problem. Access has crashed. MS Rep:  Can you describe your problem? Me:  There are funny colors all over the computer screen and nothing I do in Access has any effect. MS Rep:  Oh, I figured it out. That means your database is corrupt. Goodbye. Me:  WAIT WAIT! I know that it's corrupt. That's why I called. But what can I do about it? Ms Rep:  You need to create a new database. Goodbye. Me:  But what I was asking was how to... CLICK... end of call
This type of information gathering isn't necessarily dangerous. Using the database to find the address of a convicted felon serves society by helping the police to catch a criminal. It's tempting to watch the movies and get romantic notions about how the powerful government will hunt down the wrongfully accused, but in reality most convicted criminals really are criminals, and it's better for all of us to get them behind bars.
But this raises serious issues about the potential for digital dragnets. That's where, for example, the IRS does an SQL query on all people in the lower tax brackets possessing Ferrari's, and uses the results to build a list of suspects. Being in a lower tax bracket or possessing a Ferrari aren't crimes, but put them together and you'll have a high stastical correlation with tax evaders. At first it might sound innocent, but it amounts to a general search and seisure. It's kind of like searching the home of everyone on the block, just to see who might have drugs in their house. They invade the privacy of a lot of innocent people on the offchance of finding a guilty person. Before you know it, "they" know everything about everyone, people develop a fear of any information released through computers, commerce, or the government, and what we end up with is essentially a police state.
And then compound in the fact that the reporting agencies are prone to these little screwups, like assigning prison records to innocent people, and all of a sudden we have a bug-ridden police state, where anybody can be investigated without cause.
Personally, I'd rather not live in the kind of paranoia of this type of thing could create. Keep the info away from the government, change the law so they need a warrant to use it (and only for looking up individual suspects), or whatever. There's always a reasonable compromise.
So what this means is that the next time Microsoft gets hacked and some poor shlemazle's credit card numbers get e-mailed to everyone in Bill Gate's address book, Microsoft can't be held legally liable because "all your bits are belong to us".
Actually, the definition of 20/20 was purely arbitrary. The guy who invented the eye chart in 1864 (Hermann Snellen) picked somebody who seemed to him to have normal eyesight, and determined that he could read a 3/8" letter at twenty feet, so that's the definition of 20/20.
The 3/8" size of the 20/20 line on the eye chart is arbitrary, but the 20' is not. At ~20' and beyond the eye is "unfocused", meaning the muscle that controls the accomodative lens is completely relaxed. So if you have "normal eyesight" at 20', you also have "normal eyesight" at any distance over 20'.
However, you could debate that the original subject had "better than normal" eyesight. I remember hearing that about 50% of the American population needs glasses, and that in some populations it's as high as 90%!
I believe that society has adjusted to Snellen's definition of 20/20. Driving, schools, and other modernisms require people to have 20/20 (either natural or corrected) in order to function. If Snellen had picked a subject with worse eyesight, probably all the street signs would be printed in a little bit bigger letters. If we were to give superior vision surgery to everybody, society would probably adjust again and 20/10 vision would become "normal".
It's possible that a very small number of people can see better than 20/10 naturally, because the bottom line on the Snellen eye chart is 20/5.
Some of those children are going to be born with birth defects. Just like all surgeries can have complications and all drugs can have side effects, there will be the occasonal problem with this too.
And then what happens? Will they dispose of the unwanted children? Will the poor things grow up unloved because the parents had expected a better result?
Imagine how many genetically mutated mice and sheep must have been produced and disposed of before science finally succeeded in producing Dolly.
We really need to think long and hard before we allow these sorts of experiments to be conducted on human beings.
I've noticed a lot of posts on this posting and others that Trolltech is greedy to charge commercial developers for their library. Nothing could be further from the truth. Commercial developers would much rather use a commercial library for commercial development any day.
First of all, if you aren't making enough money to pay your suppliers, you shouldn't be doing closed-source development anyway. Commercial companies are in busniess to make a profit and will understand when their suppliers do too. A rule of business is that you only want to work with partners who are profitable - it guarantees they won't go out of business and leave you with no support.
Second, if you need a feature implemented in the next version, you can come at them with the "we're your customer" perspective. With open source tools, the developers are often busy adding in their own bells and whistles, and if they don't want to implement your feature, than you have to spend your own time and money to do it yourself.
Third, Open source development is usually not done to a schedule. I.e. "it's ready when it's ready". But commercial companies have to ship to a schedule they've given to their customers, and they take the heat whether or not a slip was due to a supplier. Much better to work with a supplier that can meet a deadline.
Finally, the QPL is much better suited for commerical development than the LGPL. The LGPL states that to statically link with the library, your application can't even be commercial. It could even be interpreted that in order to distribute the library together with your app, your whole app has to be GPL'd. Not good if you want a non-Linux port and the library doesn't exist on all your target platforms. Without an established case history to solidify the legal meaning of the LGPL, what commercial developer would want to risk their precious intellectual-property on an untested license? At least with the QPL, you know that your closed-source application can always stay closed-source unless/until you decide to open-source it.
Closed-source/commercial developers know all of this already. The pro-Gtk comments seem to come mostly from people who have only done open source development.
Accurately counting anything that large is impossible. If you count the number of new PC shipments it misses the installed base. If you count the number of boxed sets, it's impossible to distinguish between a new install or an upgrade.
Who would have predicted 6 years ago that Linux and free software would become the phenomenon that they have? Certainly not IDC.
We might want to write up Linux, Apache, etc. each into it's own patent application and file them as statutory patents. Such actions would form the ultamate shield that _proves_ prior art!
This is an excellent idea! How about extending that and making a "patent-left". Patent technologies as an ordinary patent, and don't charge royalties, but make the legal condition that in order to use the technology the code has to be open-sourced.
BTW, who's Microsoft gonna sue? Richard Stallman? Tux? Everyone who uses Linux? This isn't the typical company-suing-company case. I can't see how M$ could use patent-law to stop a group of volunteers.
Slashdot Mirror
Current inorganic solar cells have around 30% efficiency, and current organic solar cells have around 6% effeciency. However, the price per kilowatt is about the same.
A device with the kilowatt/sq-ft of a silicon panel and the cost/sq-ft of an organic panel would go a long way.
I calculated that in my area, with current technology and current electricity rates, a solar panel investment would break even in ~ 30 years (and this factors in tax incentives). Maybe break-even would occur in only 15-20 years when you factor in future increases in the cost of electricity.
If this new invention would really improve the cost/kilowatt by 5x, then a solar investment could break even in only around 5 years. That would cause a turning point in adoption.
Or we could get started now, and in 30 years have the whole planet wired with fiberoptics to the home. Helluvalot faster than DSL or Cable.
What is really needed is decent free software office apps for Windows. Even though their availability isn't by itself a compelling reason for users to change to free software, it doesn't have to be. MS Office is so expensive, and full of so many annoyances, that the open market will force the switch. Consider you are buying a new low/mid end PC for your home. The cost is about $1500 right now, or $2000 with MS office. $500 or 1/3 the cost of the system is a big difference, so much so that major computer manufacturers do not pre-install Office by default on these types of PC's. Before long PC manufacturers will be able to offer users a choice of OpenOffice for Windows at no extra charge. The only advantage of paying the extra $500 for Office would be to share files with other users, but home users for the most part don't share files anyway.
There is always the issue that the availability of free software for Windows could be a reason for users not to switch to Linux. Yet, GNU software was available for commercial UNIX systems long before a free kernel was available. During this time many users and system administrators learned to use and rely upon GNU software. When Linux became available, switching from GNU on a commercial UNIX kernel to GNU on a Linux kernel was not that big a leap. Likewise, once Windows users become accustomed to using inexpensive and powerful free software, the transition to using those same applications on Linux will not seem as difficult.
Bringing up the history of GNU, there was not only the issue of cost at stake, but the fact that GNU software often performed much better or had more compelling features than commercial software from the UNIX vendors. This is equally true with desktop software today. Microsoft's most compelling reason for users to pay to upgrade to their new Office XP software is removing the annoying paper clip character. In open source software, we wouldn't need to pay for upgrades for that, we'd just comment out those lines from the code. There are plenty of annoyance in Office to improve upon in free software.
Once users become familiar with free office software, sooner or later some of them will start installing it at work, at first just to share files with their home systems. As this becomes more widespread, there is pressure on the IT department to offer some sort of support. At first, most IT departments will say they need Office because everyone else has it. But a few innovative ones (and maybe a couple strapped for cash) will switch. And the more that do, the less valid the argument about needing Office to share files with other companies.
And, bang, the #1 reason for not deploying Linux on the desktop will be eliminated.
This will take some time to happen. Just consider how many companies run Apache on commercial UNIX, in spite of the availability of Linux. The period during which both Linux and Windows flourish will be a long one. However, this is how it should be. It allows all the users to enjoy a much smoother transition. And what about Microsoft? As long as they can offer their customers more for their upgrade dollars than just removing Clippy, Microsoft has the potential to do just fine. But for the users out there, it will be a homerun for sure.
When I asked him to explain, he said that he had a very clever way of detecting cheating. He compiled the programs and ran cmp, and my program had come out the same as someone else's! I had set my umask to allow world-read of my entire set of files, naively thinking that I could trust my classmates. The other student had copied my assignment straight off the disk drive.
Fortunately, I was able to prove I was innocent because - get this - when the student copied my program he didn't bother to remove my name from the comments !
The grader changed my grade but told me in the future he would hold me accountable for allowing others to cheat off of me, even if that wasn't my intention! Ever since then I was much more careful in how I set my file permissions. The sad part was, the person who cheated off of me had been a good friend of mine. I would have gladly offered to help him out (in the honest sense) if he had let me know he was having trouble in the class. :-(
Most of the class decided to write it in C, but I chose Common LISP. I was the only person in the class to complete the assignment on time, even though I didn't work in a group. LISP helped me for two reasons:
Now if only I had had the opporunity to sell it to Yahoo for $50 Million! ... :-)
The problem was that my MS Access database got corrupted. Whenever I tried to run a certain report, Access crashed. Unfortunately, I had spent over a week creating this report, and I wanted to recover as much of my work as possible.
I called MS Support, spent a few minutes wading through their phone tree, and then waited on hold about 20 minutes. ... ... ... ...
Me:   My Access database is corrupt. Access crashes whenever I try to run a certain report. I want to get back as much of my report as possible.
MS Rep:  Ok, try doing file/open, and pick the name of your database.
Me:  Do you want me to kill and restart Access first, because it is crashed.
MS Rep:  Huh?
Me:  I'll just kill and restart it first. Ok, the database is opened.
MS Rep:  And it works, I see. Can you press the "Reports" button?
Me:  Ok, I get a list of my reports. Now what?
MS Rep:  Can you tell me which reports are on the list?
Me:  Sure, here are the reports
MS Rep:  And which one is having problems?
Me:  This one
MS Rep:  Ok, can you try to double-click it?
Me:  Sure. Ok, it's starting to run. Now I get the problem. Access has crashed.
MS Rep:  Can you describe your problem?
Me:  There are funny colors all over the computer screen and nothing I do in Access has any effect.
MS Rep:  Oh, I figured it out. That means your database is corrupt. Goodbye.
Me:  WAIT WAIT! I know that it's corrupt. That's why I called. But what can I do about it?
Ms Rep:  You need to create a new database. Goodbye.
Me:  But what I was asking was how to
CLICK
end of call
This type of information gathering isn't necessarily dangerous. Using the database to find the address of a convicted felon serves society by helping the police to catch a criminal. It's tempting to watch the movies and get romantic notions about how the powerful government will hunt down the wrongfully accused, but in reality most convicted criminals really are criminals, and it's better for all of us to get them behind bars.
But this raises serious issues about the potential for digital dragnets. That's where, for example, the IRS does an SQL query on all people in the lower tax brackets possessing Ferrari's, and uses the results to build a list of suspects. Being in a lower tax bracket or possessing a Ferrari aren't crimes, but put them together and you'll have a high stastical correlation with tax evaders. At first it might sound innocent, but it amounts to a general search and seisure. It's kind of like searching the home of everyone on the block, just to see who might have drugs in their house. They invade the privacy of a lot of innocent people on the offchance of finding a guilty person. Before you know it, "they" know everything about everyone, people develop a fear of any information released through computers, commerce, or the government, and what we end up with is essentially a police state.
And then compound in the fact that the reporting agencies are prone to these little screwups, like assigning prison records to innocent people, and all of a sudden we have a bug-ridden police state, where anybody can be investigated without cause.
Personally, I'd rather not live in the kind of paranoia of this type of thing could create. Keep the info away from the government, change the law so they need a warrant to use it (and only for looking up individual suspects), or whatever. There's always a reasonable compromise.
So what this means is that the next time Microsoft gets hacked and some poor shlemazle's credit card numbers get e-mailed to everyone in Bill Gate's address book, Microsoft can't be held legally liable because "all your bits are belong to us".
Actually, the definition of 20/20 was purely arbitrary. The guy who invented the eye chart in 1864 (Hermann Snellen) picked somebody who seemed to him to have normal eyesight, and determined that he could read a 3/8" letter at twenty feet, so that's the definition of 20/20.
The 3/8" size of the 20/20 line on the eye chart is arbitrary, but the 20' is not. At ~20' and beyond the eye is "unfocused", meaning the muscle that controls the accomodative lens is completely relaxed. So if you have "normal eyesight" at 20', you also have "normal eyesight" at any distance over 20'.
However, you could debate that the original subject had "better than normal" eyesight. I remember hearing that about 50% of the American population needs glasses, and that in some populations it's as high as 90%!
I believe that society has adjusted to Snellen's definition of 20/20. Driving, schools, and other modernisms require people to have 20/20 (either natural or corrected) in order to function. If Snellen had picked a subject with worse eyesight, probably all the street signs would be printed in a little bit bigger letters. If we were to give superior vision surgery to everybody, society would probably adjust again and 20/10 vision would become "normal".
It's possible that a very small number of people can see better than 20/10 naturally, because the bottom line on the Snellen eye chart is 20/5.
Some of those children are going to be born with birth defects. Just like all surgeries can have complications and all drugs can have side effects, there will be the occasonal problem with this too.
And then what happens? Will they dispose of the unwanted children? Will the poor things grow up unloved because the parents had expected a better result?
Imagine how many genetically mutated mice and sheep must have been produced and disposed of before science finally succeeded in producing Dolly.
We really need to think long and hard before we allow these sorts of experiments to be conducted on human beings.
I've noticed a lot of posts on this posting and others that Trolltech is greedy to charge commercial developers for their library. Nothing could be further from the truth. Commercial developers would much rather use a commercial library for commercial development any day.
First of all, if you aren't making enough money to pay your suppliers, you shouldn't be doing closed-source development anyway. Commercial companies are in busniess to make a profit and will understand when their suppliers do too. A rule of business is that you only want to work with partners who are profitable - it guarantees they won't go out of business and leave you with no support.
Second, if you need a feature implemented in the next version, you can come at them with the "we're your customer" perspective. With open source tools, the developers are often busy adding in their own bells and whistles, and if they don't want to implement your feature, than you have to spend your own time and money to do it yourself.
Third, Open source development is usually not done to a schedule. I.e. "it's ready when it's ready". But commercial companies have to ship to a schedule they've given to their customers, and they take the heat whether or not a slip was due to a supplier. Much better to work with a supplier that can meet a deadline.
Finally, the QPL is much better suited for commerical development than the LGPL. The LGPL states that to statically link with the library, your application can't even be commercial. It could even be interpreted that in order to distribute the library together with your app, your whole app has to be GPL'd. Not good if you want a non-Linux port and the library doesn't exist on all your target platforms. Without an established case history to solidify the legal meaning of the LGPL, what commercial developer would want to risk their precious intellectual-property on an untested license? At least with the QPL, you know that your closed-source application can always stay closed-source unless/until you decide to open-source it.
Closed-source/commercial developers know all of this already. The pro-Gtk comments seem to come mostly from people who have only done open source development.
Accurately counting anything that large is impossible. If you count the number of new PC shipments it misses the installed base. If you count the number of boxed sets, it's impossible to distinguish between a new install or an upgrade.
Who would have predicted 6 years ago that Linux and free software would become the phenomenon that they have? Certainly not IDC.
This is an excellent idea! How about extending that and making a "patent-left". Patent technologies as an ordinary patent, and don't charge royalties, but make the legal condition that in order to use the technology the code has to be open-sourced.
BTW, who's Microsoft gonna sue? Richard Stallman? Tux? Everyone who uses Linux? This isn't the typical company-suing-company case. I can't see how M$ could use patent-law to stop a group of volunteers.