Even the worst government planner can reasonably predict future demand for certain professions given the necessary data.
I'm just suspicious of central planners. Would a bureaucrat have allowed my case? I spent 25 years doing systems analysis and applied research in the telecom and cable field, essentially within one company. When my position was eliminated by industry consolidation (along with 1700 of my associates), I decided that I had accumulated enough wealth that I could go back to school and study something completely different. So I'm an economics graduate student, and twice as old as the rest of the students. Or would the bureaucrat have said, "Graduate school is for 23 year-olds -- you must continue to work, go to this place and push buttons."
I'm interested in the economics of aging populations -- not just because it's my own personal situation, but because so many countries will be faced with that problem in the future. Even China, although it will take them longer to get there. Their population growth has been checked (by rather draconian laws), they are becoming rich enough to afford much better health care, so 40 years from now they will have to deal with the problems that come with a large fraction of the population having reached "retirement" age.
And China doing it - this is just a poor starving commie dictatorship trying to make everyone forget it's nothing more than a poor, starving commie dictatorship.
As several people pointed out to me in a different thread, China has the world's 6th largest economy by GDP, 2nd largest by purchasing power, and far and away the fastest growing of the top ten economies. They are a net exporter of food. Certainly they are farther behind on a per capita basis, but on a total basis, of the top ten economies in the world, they may be better able to afford a space program than anyone.
It seems a little hypocritical to criticize China for not putting all its money into poverty relief. Check out the economic status in areas like Appalachia at the time NASA started the moon program in the 60's. Nor are we immune to the same criticism today -- the world's richest country and 20M people live in poverty and 40M people have no health insurance against catestrophic illness.
Yep, they're still a commie dictatorship. I'm going to class at the University of Colorado with someone from China who would much rather be studying dance or art, but the central planners decided that they needed economists, and that she should be one of them.
I'm no expert here, but doesn't this qualify as vaporware?
Too bad the DOJ isn't taking their job of overseeing the convicted monopolist seriously.
One of the things that can get monopolists in further trouble is called "overhanging the market," that is,
announcing products long before they can be ready in order to keep lesser players from gaining any market share.
When I worked for the R&D arm of the local telcos after the 1984 breakup of the Bell System,
we had to be extremely careful about market overhang.
Of course, in that case, the judge had kept oversight for himself instead of leaving it in the hands of the DOJ.
Not completely. There are bandwidth limits to color NTSC that are well defined (horizontal bandwidth has to be limited so there is a place to put the color information). To digitize an analogue signal at 720 pixels per line is sensible with good starting material. If you want easy to compute square squares and circular circles, 640 isn't that bad a figure. Strange values in between have been seen too.
IIRC, NTSC bandwidth translates into a nominal 704 pixels per scan line -- non-rectangular pixels, with which I dislike dealing. The MPEG profiles allow a maximum of 720 pixels per scan line for standard-definition video. And, as you say, 640 pixels per line by 480 lines gives nice square pixels and is, in practice, quite close to the actual resolution of the frame. Of course, there's that other ugliness that you're actually dealing with two 640x240-pixel fields, which have to be interlaced to provide a full frame. For many applications, I have found that grabbing 320x240-pixel images from odd frames only (or even frames, take your choice) provides adequate image quality without the interlace problem.
It's been a while since I was doing anything with video on Linux, but I recall that frame grabbers based on the BT8x8 family do down-sampling of the image in hardware, so it's easy to choose a frame size that fits the application. Also easy to grab grayscale images rather than color, which involve less data and are faster to process, and may be entirely adequate for motion-detection work. The last time I used v4l one part of the project involved simple motion detection, and grayscale worked well for me.
There were people who proposed flat tax but was shut down becaue aruged that it is "pro rich".
Okay, for purposes of discussion, let's have a simple-but-not-flat tax.
First, admit that Social Security accounting is a fraud, it's a pay-as-you-go system, and we're going to unify all of the federal taxes on income.
Second, no more social engineering, everyone pays taxes on all their income, regardless of source and regardless of what they spend it on.
I'm not really in favor of that, since I would like to encourage people to save and to get an education, but it's a slippery slope.
Third, we'll have three tax brackets, indexed for inflation.
I'll suggest that the first $20,000 is tax free.
No income tax on the money required for living at a subsistence level.
From $20,000 to $200,000, you pay 20%.
Everything over $200,000, you pay 40%.
I actually have no idea if those figures would be revenue-neutral or not, but I'm sure that they could be with some adjustments.
This scheme is not pro-rich, since it's much less regressive than the current scheme is -- the working poor currently pay a 15% rate starting with the first dollar they earn due to SS and Medicare.
Half of that is hidden -- the employer pays half of your SS and Medicare "for you" and it doesn't show on the check stub.
I don't see where the rich have much to complain about -- when a CEO gets paid $100M for running a company into the ground, he/she still keeps 60%. How many yachts do you really need?
To the extent that the brackets have to move down or the percentages go up, at least it makes it transparent to everyone just what their total tax bill to the federal government is.
It also makes clear the SS fraud that has been perpetrated; when the future payments are estimated and it is obvious that the bracket rates will have to increase to 30% and 60%, everyone will know what's going on.
Oh, and I'd like to see borrowing against unrealized asset gains that is not reinvested (the basic tax avoidance scheme of the uber-wealthy) declared to be income and taxed.
It used to be skilled IT workers could get awesome pay, decent job perks and benefits.
There are a lot of interesting microeconomic issues tied up in this situation. Awesome pay is generally the result of one of two situations: the person earning the pay is really worth it, or there is a shortage of whatever skills the person is providing. At least on a worldwide basis, there no longer seems to be any shortage of IT skills. In the US, the current employment situation would suggest that there's no shortage of IT skills (relative to the demand) within the country, either. I believe an argument could be made that many IT workers are, indeed, nothing more than "skilled mechanics".
As a point for discussion, the "really worth it" is about impact on the bottom line; and big impact usually means that there's a lot of leverage, in the sense that a program that saves ten cents per customer transaction is worth much more to a company that does 20 million transactions per month than it is to a company that does 100,000. The type of technology analysis that I did until recently simply is not worth as much to a small firm as it is to a large one; if your annual costs are $10M, I can't figure out any way to cut them by $100M. So the awesome compensation would be more typically earned by someone working at a very large company.
To the extent that job growth is generally driven by smaller businesses, the results should not be unexpected. At some points in time, average salaries decrease, as the big companies cut their staffs and those people go to work at smaller firms. The nature of the business can have a big impact as well. When I worked for an international holding company, our revenue per employee at the holding company was very large. When we were purchased by a larger company and the overseas ventures sold for cash, the revenue per employee at headquarters was reduced dramatically (and pressure to lay people off began immediately). It is interesting to note that for a large firm in the US today, it is easier to cut staffing levels by 20% than it is to cut pay levels by the same amount.
38 years. Of course, making 40-year forecasts in economics is a pretty difficult job.
From the Economist's most recent World in Figures, the US GDP is a little more than five times the size of China's. If they outgrow the US by 4-5% per year (that is, if the US economy is growing at 3% per year, China has to grow at 7-8%) for the next 40 years, they would indeed catch up. IIRC, on a historical basis, 5% per year sustained growth is considered a spectacular accomplishment. 8% per year for 40 years may be possible, but seems unlikely. If they only outgrow the US by 2% per year over that 40 years (5% to 3%), their GDP would still be only half that of the US.
One of the big economic problems facing the US, Japan and western Europe is the rapid aging of their populations. China's current population is much younger, but by means of fairly draconian laws, they have stabilized its growth. As a result, in less than that 40 years, they will face the same problems of trying to support a large and rapidly growing population of elderly people. I happen to think this is one of the most fascinating economic problems around: medical and general sanitation improvements have greatly extended the lifespan; current cultural mores tend towards the position that the elderly should not have to work, and indeed many of the elderly are not capable of working due to physical or mental infirmities; can the remaining workers be productive enough to support the elderly, and how do you manage the redistribution of income?
I'm not saying that China can't do it, nor that China isn't going to be the world's economic powerhouse in 40 years. My original post was meant to point out that, at this point in time, the Chinese economy is much more dependent on continued exports to the US than the US is dependent on those imports.
What will the US do if China decides to stop exporting to the US?
What will China do if the US decides to stop importing from China? While China currently runs a large trade surplus with the US (OTOO $100B per year, I believe), they do not run a trade surplus with the world as a whole. IIW, they are spending those dollars on goods and services, they're just not buying them from the US. What happens to their growing economy if the US quits funding the expansion? Think in terms of serious economic crash...
Easy enough to say, in a collective sense. Much harder to do when it's your mom, or your favorite aunt, that you are talking about unplugging. Can I be there when you tell your mom, "I know the cataract thing is inconvenient, but I'd rather have someone walk on Mars than have you see."?
The system is broken. The boomers' parents bought themselves a nice retirement package, getting the boomers to pay for it by promising that they could have the same deal when they got old. But for various reasons, the numbers don't work out and it doesn't look like that promise can be kept. But the boomers may well bankrupt the country trying to get it anyway.
IMO, US motivation won't count this time around. The US economy has piled up so much debt, and is facing the enormous future expenses of the retiring baby-boom generation. When the time comes that Congress has to decide between manned spaceflight and medical care for the boomers, the boomers are going to win. Regardless of what the rest of the world may do, the US can't afford a significant new manned space program. Most of western Europe and Japan are in the same position due to rapidly aging populations.
Space flight is a game for young, growing populations.
We did not make the incredibly stupid decision to design applications so that they execute programs that arrive in the mail.
Yes, but it is probable that for Linux to achieve significant desktop market share (relative to Windows), someone will have to write an e-mail client that will execute programs that arrive in the mail. The users will demand it. The broad population of computer users will not give up what they regard as conveniences in order to gain security. And once the author of a trojan can get the user to fork() a particular process on an Internet-connected machine, it becomes more difficult to stop that process from downloading other programs (eg, a rootkit) and running that.
My personal opinion is that the only answer is an execution environment which supports fine-grained control of access to system resources. That is, a process that is launched in such a fashion might have access to the screen, or keyboard, but not the disk, or only to sections of the disk, and not to the network, etc. Said access to be enforced by the operating system. I would not bet on anything except interpreted languages as being able to provide that degree of control.
The Parker ballpoints seem to be the only ones that are durable enough for me. Every other brand I seem to smash the ball into the socket and after a few days it starts to skip and blob. Never been able to do that to a Parker. I seem to recall reading somewhere once that their ball-and-socket materials are considerably more expensive than everyone else's.
Actually, my favorite writing utensil is a "00" drafting pen with black India ink, but keeping it clean is a hassle and they leak badly when you take them on airplanes.
IIRC, several of the big media companies testified before Congressional committees, under oath, that they would never use the law to do things like stifle competition or scientific research.
This seems to be fairly common practice these days; companies take a particular position in their testimony, and do exactly the opposite at the first opportunity.
I, for one, would like to see them called back to explain to the committees how they reconcile their words and their actions.
And do some jail time for perjury, although that's a bit of a fantasy.
Of course, embarrassing the companies in public would no doubt cost the Congress critters the donations that assure their reelection, so I'm not holding my breath.
As others note, speedy trial is a right in a criminal proceeding.
Some civil trials, and this will probably be one of them, are slow because of the volume of the necessary "evidence". Each side is entitled to access to the other side's internal records as part of the discovery process. Having seen this in action in civil cases where I worked, at some point SCO will demand access, and at IBM, everyone who has ever worked on a Linux project will probably have their office and computer contents placed under a "hold" order. At some point legal people will come around and make copies of everything -- in our case, they literally made an image of all the hard disks, as well as copies of paper stuff. It may take weeks or months for them to get to your office, and while you're waiting, don't be throwing stuff away! Then all of those paper and electronic copies have to be examined.
IIRC, when the federal government had IBM in court in the 1970's, IBM was required to produce copies of approximately 50 million pages as part of the discovery process. Ironically, the federal government had to buy a document management system from IBM to manage the "evidence" -- IBM was the only company that built a system that could handle that volume.
IANAL, but I think not, or at least only under specific conditions and for very limited purposes. Fair use allows for use for specific purposes; for example, quoting in order to facilitate a literary discussion.
I would put header code in a category of instantiating a specification -- as in, "the system call returns a structure with elements having the following names and types" -- and all instantiations of that spec will necessarily be quite similar. Whether your header file infringes would probably depend on the context provided by the specification. If the spec were provided for purposes of interoperability (eg, the POSIX specs) then you're almost certainly not infringing. If it were provided for the purpose of using the system call in a program written for a particular OS, then you might be infringing if you used it to implement a compatible OS.
And if you were copying 200 lines of code that implemented a particular algorithm particularly well into your program, you're almost certainly infringing. Whether you could quote it without permission in a design book could be subject to debate, in large part depending on how a court interprets the size of your "quote". In my experience, most authors writing such a book would take the effort to get permission to quote that much code. The copyright holder would have to take into account their perceived value of the code, since they know it's going to be used without permission after it appears, despite any restrictions that might accompany it.
In fact, many professors must publish books because they do not make enough money teaching and through research.
Many professors publish books because they are seeking tenure. Publications count. Published books count quite a bit, IIRC, although the metrics vary from school to school. Unless you are fortunate (and skilled) enough to write "a classic" -- a book so good that classes at many schools use it -- the direct income from the book is probably less important than the bump in pay and security that tenure carries.
Many of the tiered plans that cable companies have looked at involved lower tiers as well as higher ones. For example, a lower tier that competed with dial-up with always-on 128 kbps downstream and 32 kbps upstream for $19.95 or less. There was (and probably still is) a great deal of fear that this type of offering would "cannabilize" the existing service; that many current users would find the 128k service to be fine (most Slashdot readers clearly not in that camp) and would move to the lower-priced service, and high-speed data revenues would fall dramatically.
Many businesses have found that the willingness-to-pay at various prices by their customers is such that profits are maximized by selling to fewer customers at higher prices. Cable modem service at today's prices seem to be an example of that.
I know of no way other than being able to compile and run the resulting binary to verify that the source code provided is indeed the correct, working source code. Your point about the custom compiler may be valid -- that is, I'll give you my source code changes, but they are specific to a particular compiler you do not have. Given full source code, you can presumably port it to whatever compiler you do have. However, there would certainly be lines across which said custom compiler could not step (IMO) and have the whole thing remain GPL-compliant.
For example, one of the Bell Labs' UNIX gods (I forget which) demonstrated how a C compiler could (a) insert backdoor binary code into applications it was compiling and (b) recognize when it was compiling itself and insert the backdoor-inserting code. Thus none of the source files, for either the compiler or the application, showed that there was a backdoor. They were making the point that the system is not secure if you're initially dependent on some chunk of binary code (or at least you have to analyze that binary, which is much more difficult).
In this GPL example, if the custom compiler inserted binary code needed to build a working program, and no other compiler working strictly from the source could produce a working program, there's pretty clearly a violation.
Re:New light to shed on Bill Gates, Microsoft and
on
SCO's Plan Examined
·
· Score: 1
Even the Bush people aren't that dumb. When every big company in the country except MS screams, the Bushies will see their campaign contributions disappearing and listen. And the companies screaming start with IBM, Sun and Oracle, and go on from there.
Ten years from now, all my work will be in ascii text still, and wether or not LaTeX exists at that time, all I need is a text editor to view my work!
Certainly for a professional writer
(or anyone whose profession calls for writing),
longevity of the file format should be an important consideration.
I have a friend who started a book back around 1995, who wanted to go back and pick up his work on it.
It was written in whatever version of Word was current at the time.
2003 versions of Word would not successfully open the files.
I don't know if MS provides any sort of converters for files that old, but TTBOMK, my friend never was successful in fully recovering the contents.
There was quite a bit of math in the manuscript, and I believe that those parts had to be re-entered manually.
I can't speak for LaTex, but I have 20-year-old papers prepared with troff/mm that format just fine with groff/mm today.
I get the impression that the majority of the keyboards in the "*nix hardware" world were like this originally.
The majority of keyboards in the computing world were like this originally. In the days of alphanumeric terminals, a variety of special ASCII characters (backspace, escape, etc) were in routine use almost everywhere, and it was much easier to type control-h rather than hunt for the backspace key. IIRC, the caps-lock key didn't start showing up next to the "A" key until computers and terminals began to be used by more secretary-types. Those users wanted a caps-lock key in the same place it had been on their typewriters (you remember typewriters, don't you?), and were eventually a bigger market than programmers. My fingers still touch-type control-h instead of the backspace key; one of my pet peeves is PC applications that use control-h for something other than backspace (some, for example, use it to invoke a help window).
The best keyboard I ever used was the one that came with the third (I believe) version of the BLIT terminal at Bell Labs. The designers listened to the programmers that used those terminals. That version of the keyboard had a wonderful touch, all the keys in the right places, all the keys the right sizes, was heavy enough to stay put on a work surface and light enough to hold on your lap without cutting off the blood flow to your legs.
Just out of curiousity, is this position paid on an hourly basis, or is it salaried? If it's a salaried position, docking people's pay over a few minutes of tardiness seems, with few exceptions, just plain silly. After all, salaried people are typically expected to put in all the extra hours that it takes to "get the job done" without overtime pay when there's a crisis. One area of exceptions might be when the salaried employee is responsible for some portion of a 24/7 coverage scheme, and the person "on duty" can't leave until their replacement shows.
The big cable companies are certainly hoping that it's a passing fad. The last figures that I saw indicated that on the order of one-third of the downstream bandwidth used on the cable modem networks was generated by a quite small percentage of the users doing P2P downloads. The fraction of upstream bandwidth being consumed by those same users with P2P was much higher. And of course, since much of the usage involved transfers to/from machines not on the local network, they consume a lot of expensive bandwidth into the backbone networks.
Cable modem service would be much more profitable if the companies could shed the right 2% of their users. We used to love it when those users would threaten us with "I'll take my business somewhere else." Couldn't say it, but the company would have loved to help them move to DSL...
Not just the business model, but the existing widely-deployed technology.
Suppose I am a cable company with 25% digital service penetration (that is, 25% of the households I serve have one or more digital boxes). If I'm a big cable company, I have a couple billion dollars tied up in proprietary encryption and decryption equipment -- and don't tell me that I should have used an "open standard" for that equipment, there wasn't one suitable for deployment in the US when I started my digital business. Different companies have different proprietary solutions, so CableLabs was in the tough position of having to create a standard that could work with all of the deployed encryption schemes.
Also from a technology perspective, keep in mind the conditions under which the system has to work. In particular, the system must provide adequate security even when operating on one-way plant, and under the assumption that all end users are "hostile". The single one-way feed has to provide the encrypted content, the real-time decryption keys (which must also be protected), and all of the authorization information. And it operates in a true broadcast mode. TTBOMK, all schemes that work under such conditions require that each receiver include a unique piece of "secret" information that is also known by the encryption system. The degree to which that information is accessible by the end user is inversely related to the overall security -- greater access equals less security. The information for satellite TV receivers is fairly easily accessible, and the piracy rates for satellite are much higher than for digital cable.
SCO's case is a simple contract violation lawsuit against IBM. Thats it. Its a case that they might actually have a legitimate claim with.
Exactly.
While it is traditional when suing a company that you claim has damaged you to include as many different legal theories about how they damaged you as possible (because you can't introduce new theories on appeal), this case appears to come down to a fairly simple matter of contract law.
Can SCO convince a judge that a contract between IBM and AT&T signed back in the 80s in which IBM agreed that any additions they made to UNIX would be treated as trade secrets is still relevant and binding?
Can IBM convince the judge that AT&T handled UNIX secrecy so poorly themselves (the BSD case) that any trade secret status had disappeared before AT&T sold the rights?
IANAL, but have dealt with an unfortunate number of them over the years, and feel that SCO's contract case is pretty weak. I think the BSD case provides a significant argument that AT&T did not adequately protect the "secret" status of the UNIX code -- you can't just declare something to be secret, you have to act as though it is, and the BSD case seemed to establish pretty well that AT&T had not acted in an appropriate fashion. And if UNIX was no longer secret, agreements to treat it as such were probably no longer binding. In the same vein, SCO will have to address their own actions in distributing Linux source code WRT preserving their own trade secrets. Ignorance, as in "we didn't know our secret code was in the Linux kernel" is generally a poor defense. The courts tend to expect you to know and understand what it is you are selling (or giving away).
Slashdot Mirror
I'm interested in the economics of aging populations -- not just because it's my own personal situation, but because so many countries will be faced with that problem in the future. Even China, although it will take them longer to get there. Their population growth has been checked (by rather draconian laws), they are becoming rich enough to afford much better health care, so 40 years from now they will have to deal with the problems that come with a large fraction of the population having reached "retirement" age.
It seems a little hypocritical to criticize China for not putting all its money into poverty relief. Check out the economic status in areas like Appalachia at the time NASA started the moon program in the 60's. Nor are we immune to the same criticism today -- the world's richest country and 20M people live in poverty and 40M people have no health insurance against catestrophic illness.
Yep, they're still a commie dictatorship. I'm going to class at the University of Colorado with someone from China who would much rather be studying dance or art, but the central planners decided that they needed economists, and that she should be one of them.
It's been a while since I was doing anything with video on Linux, but I recall that frame grabbers based on the BT8x8 family do down-sampling of the image in hardware, so it's easy to choose a frame size that fits the application. Also easy to grab grayscale images rather than color, which involve less data and are faster to process, and may be entirely adequate for motion-detection work. The last time I used v4l one part of the project involved simple motion detection, and grayscale worked well for me.
I actually have no idea if those figures would be revenue-neutral or not, but I'm sure that they could be with some adjustments. This scheme is not pro-rich, since it's much less regressive than the current scheme is -- the working poor currently pay a 15% rate starting with the first dollar they earn due to SS and Medicare. Half of that is hidden -- the employer pays half of your SS and Medicare "for you" and it doesn't show on the check stub. I don't see where the rich have much to complain about -- when a CEO gets paid $100M for running a company into the ground, he/she still keeps 60%. How many yachts do you really need?
To the extent that the brackets have to move down or the percentages go up, at least it makes it transparent to everyone just what their total tax bill to the federal government is. It also makes clear the SS fraud that has been perpetrated; when the future payments are estimated and it is obvious that the bracket rates will have to increase to 30% and 60%, everyone will know what's going on.
Oh, and I'd like to see borrowing against unrealized asset gains that is not reinvested (the basic tax avoidance scheme of the uber-wealthy) declared to be income and taxed.
As a point for discussion, the "really worth it" is about impact on the bottom line; and big impact usually means that there's a lot of leverage, in the sense that a program that saves ten cents per customer transaction is worth much more to a company that does 20 million transactions per month than it is to a company that does 100,000. The type of technology analysis that I did until recently simply is not worth as much to a small firm as it is to a large one; if your annual costs are $10M, I can't figure out any way to cut them by $100M. So the awesome compensation would be more typically earned by someone working at a very large company.
To the extent that job growth is generally driven by smaller businesses, the results should not be unexpected. At some points in time, average salaries decrease, as the big companies cut their staffs and those people go to work at smaller firms. The nature of the business can have a big impact as well. When I worked for an international holding company, our revenue per employee at the holding company was very large. When we were purchased by a larger company and the overseas ventures sold for cash, the revenue per employee at headquarters was reduced dramatically (and pressure to lay people off began immediately). It is interesting to note that for a large firm in the US today, it is easier to cut staffing levels by 20% than it is to cut pay levels by the same amount.
38 years. Of course, making 40-year forecasts in economics is a pretty difficult job.
From the Economist's most recent World in Figures, the US GDP is a little more than five times the size of China's. If they outgrow the US by 4-5% per year (that is, if the US economy is growing at 3% per year, China has to grow at 7-8%) for the next 40 years, they would indeed catch up. IIRC, on a historical basis, 5% per year sustained growth is considered a spectacular accomplishment. 8% per year for 40 years may be possible, but seems unlikely. If they only outgrow the US by 2% per year over that 40 years (5% to 3%), their GDP would still be only half that of the US.
One of the big economic problems facing the US, Japan and western Europe is the rapid aging of their populations. China's current population is much younger, but by means of fairly draconian laws, they have stabilized its growth. As a result, in less than that 40 years, they will face the same problems of trying to support a large and rapidly growing population of elderly people. I happen to think this is one of the most fascinating economic problems around: medical and general sanitation improvements have greatly extended the lifespan; current cultural mores tend towards the position that the elderly should not have to work, and indeed many of the elderly are not capable of working due to physical or mental infirmities; can the remaining workers be productive enough to support the elderly, and how do you manage the redistribution of income?
I'm not saying that China can't do it, nor that China isn't going to be the world's economic powerhouse in 40 years. My original post was meant to point out that, at this point in time, the Chinese economy is much more dependent on continued exports to the US than the US is dependent on those imports.
Easy enough to say, in a collective sense. Much harder to do when it's your mom, or your favorite aunt, that you are talking about unplugging. Can I be there when you tell your mom, "I know the cataract thing is inconvenient, but I'd rather have someone walk on Mars than have you see."?
The system is broken. The boomers' parents bought themselves a nice retirement package, getting the boomers to pay for it by promising that they could have the same deal when they got old. But for various reasons, the numbers don't work out and it doesn't look like that promise can be kept. But the boomers may well bankrupt the country trying to get it anyway.
IMO, US motivation won't count this time around. The US economy has piled up so much debt, and is facing the enormous future expenses of the retiring baby-boom generation. When the time comes that Congress has to decide between manned spaceflight and medical care for the boomers, the boomers are going to win. Regardless of what the rest of the world may do, the US can't afford a significant new manned space program. Most of western Europe and Japan are in the same position due to rapidly aging populations.
Space flight is a game for young, growing populations.
My personal opinion is that the only answer is an execution environment which supports fine-grained control of access to system resources. That is, a process that is launched in such a fashion might have access to the screen, or keyboard, but not the disk, or only to sections of the disk, and not to the network, etc. Said access to be enforced by the operating system. I would not bet on anything except interpreted languages as being able to provide that degree of control.
The Parker ballpoints seem to be the only ones that are durable enough for me. Every other brand I seem to smash the ball into the socket and after a few days it starts to skip and blob. Never been able to do that to a Parker. I seem to recall reading somewhere once that their ball-and-socket materials are considerably more expensive than everyone else's.
Actually, my favorite writing utensil is a "00" drafting pen with black India ink, but keeping it clean is a hassle and they leak badly when you take them on airplanes.
IIRC, several of the big media companies testified before Congressional committees, under oath, that they would never use the law to do things like stifle competition or scientific research. This seems to be fairly common practice these days; companies take a particular position in their testimony, and do exactly the opposite at the first opportunity. I, for one, would like to see them called back to explain to the committees how they reconcile their words and their actions. And do some jail time for perjury, although that's a bit of a fantasy. Of course, embarrassing the companies in public would no doubt cost the Congress critters the donations that assure their reelection, so I'm not holding my breath.
As others note, speedy trial is a right in a criminal proceeding.
Some civil trials, and this will probably be one of them, are slow because of the volume of the necessary "evidence". Each side is entitled to access to the other side's internal records as part of the discovery process. Having seen this in action in civil cases where I worked, at some point SCO will demand access, and at IBM, everyone who has ever worked on a Linux project will probably have their office and computer contents placed under a "hold" order. At some point legal people will come around and make copies of everything -- in our case, they literally made an image of all the hard disks, as well as copies of paper stuff. It may take weeks or months for them to get to your office, and while you're waiting, don't be throwing stuff away! Then all of those paper and electronic copies have to be examined.
IIRC, when the federal government had IBM in court in the 1970's, IBM was required to produce copies of approximately 50 million pages as part of the discovery process. Ironically, the federal government had to buy a document management system from IBM to manage the "evidence" -- IBM was the only company that built a system that could handle that volume.
IANAL, but I think not, or at least only under specific conditions and for very limited purposes. Fair use allows for use for specific purposes; for example, quoting in order to facilitate a literary discussion.
I would put header code in a category of instantiating a specification -- as in, "the system call returns a structure with elements having the following names and types" -- and all instantiations of that spec will necessarily be quite similar. Whether your header file infringes would probably depend on the context provided by the specification. If the spec were provided for purposes of interoperability (eg, the POSIX specs) then you're almost certainly not infringing. If it were provided for the purpose of using the system call in a program written for a particular OS, then you might be infringing if you used it to implement a compatible OS.
And if you were copying 200 lines of code that implemented a particular algorithm particularly well into your program, you're almost certainly infringing. Whether you could quote it without permission in a design book could be subject to debate, in large part depending on how a court interprets the size of your "quote". In my experience, most authors writing such a book would take the effort to get permission to quote that much code. The copyright holder would have to take into account their perceived value of the code, since they know it's going to be used without permission after it appears, despite any restrictions that might accompany it.
Many of the tiered plans that cable companies have looked at involved lower tiers as well as higher ones. For example, a lower tier that competed with dial-up with always-on 128 kbps downstream and 32 kbps upstream for $19.95 or less. There was (and probably still is) a great deal of fear that this type of offering would "cannabilize" the existing service; that many current users would find the 128k service to be fine (most Slashdot readers clearly not in that camp) and would move to the lower-priced service, and high-speed data revenues would fall dramatically.
Many businesses have found that the willingness-to-pay at various prices by their customers is such that profits are maximized by selling to fewer customers at higher prices. Cable modem service at today's prices seem to be an example of that.
For example, one of the Bell Labs' UNIX gods (I forget which) demonstrated how a C compiler could (a) insert backdoor binary code into applications it was compiling and (b) recognize when it was compiling itself and insert the backdoor-inserting code. Thus none of the source files, for either the compiler or the application, showed that there was a backdoor. They were making the point that the system is not secure if you're initially dependent on some chunk of binary code (or at least you have to analyze that binary, which is much more difficult).
In this GPL example, if the custom compiler inserted binary code needed to build a working program, and no other compiler working strictly from the source could produce a working program, there's pretty clearly a violation.
Even the Bush people aren't that dumb. When every big company in the country except MS screams, the Bushies will see their campaign contributions disappearing and listen. And the companies screaming start with IBM, Sun and Oracle, and go on from there.
The best keyboard I ever used was the one that came with the third (I believe) version of the BLIT terminal at Bell Labs. The designers listened to the programmers that used those terminals. That version of the keyboard had a wonderful touch, all the keys in the right places, all the keys the right sizes, was heavy enough to stay put on a work surface and light enough to hold on your lap without cutting off the blood flow to your legs.
Just out of curiousity, is this position paid on an hourly basis, or is it salaried? If it's a salaried position, docking people's pay over a few minutes of tardiness seems, with few exceptions, just plain silly. After all, salaried people are typically expected to put in all the extra hours that it takes to "get the job done" without overtime pay when there's a crisis. One area of exceptions might be when the salaried employee is responsible for some portion of a 24/7 coverage scheme, and the person "on duty" can't leave until their replacement shows.
The big cable companies are certainly hoping that it's a passing fad. The last figures that I saw indicated that on the order of one-third of the downstream bandwidth used on the cable modem networks was generated by a quite small percentage of the users doing P2P downloads. The fraction of upstream bandwidth being consumed by those same users with P2P was much higher. And of course, since much of the usage involved transfers to/from machines not on the local network, they consume a lot of expensive bandwidth into the backbone networks.
Cable modem service would be much more profitable if the companies could shed the right 2% of their users. We used to love it when those users would threaten us with "I'll take my business somewhere else." Couldn't say it, but the company would have loved to help them move to DSL...
Not just the business model, but the existing widely-deployed technology.
Suppose I am a cable company with 25% digital service penetration (that is, 25% of the households I serve have one or more digital boxes). If I'm a big cable company, I have a couple billion dollars tied up in proprietary encryption and decryption equipment -- and don't tell me that I should have used an "open standard" for that equipment, there wasn't one suitable for deployment in the US when I started my digital business. Different companies have different proprietary solutions, so CableLabs was in the tough position of having to create a standard that could work with all of the deployed encryption schemes.
Also from a technology perspective, keep in mind the conditions under which the system has to work. In particular, the system must provide adequate security even when operating on one-way plant, and under the assumption that all end users are "hostile". The single one-way feed has to provide the encrypted content, the real-time decryption keys (which must also be protected), and all of the authorization information. And it operates in a true broadcast mode. TTBOMK, all schemes that work under such conditions require that each receiver include a unique piece of "secret" information that is also known by the encryption system. The degree to which that information is accessible by the end user is inversely related to the overall security -- greater access equals less security. The information for satellite TV receivers is fairly easily accessible, and the piracy rates for satellite are much higher than for digital cable.
IANAL, but have dealt with an unfortunate number of them over the years, and feel that SCO's contract case is pretty weak. I think the BSD case provides a significant argument that AT&T did not adequately protect the "secret" status of the UNIX code -- you can't just declare something to be secret, you have to act as though it is, and the BSD case seemed to establish pretty well that AT&T had not acted in an appropriate fashion. And if UNIX was no longer secret, agreements to treat it as such were probably no longer binding. In the same vein, SCO will have to address their own actions in distributing Linux source code WRT preserving their own trade secrets. Ignorance, as in "we didn't know our secret code was in the Linux kernel" is generally a poor defense. The courts tend to expect you to know and understand what it is you are selling (or giving away).