Indeed, electronics (in particular, semiconductors and integrated circuits) did eventually explode, making very substantial changes in all parts of our lives. However, there was a lengthy period when it was controlled by that small number of large companies and such innovation did not occur. I would not like to see a 20-year period where software is controlled by a handful of companies before breaking free again.
A thousand or so of these half million dollar lawsuits are all that is needed to take down such a large corporation.
A thousand or so of these lawsuits are all that is needed to block everyone except large corporations out of the software development business. At some point, there will be enough software patents that it will be, literally, impossible to write any reasonably complicated piece of code without infringing on one or more of them. If this case holds up on appeal, no one can use plugins in this fashion without paying royalties until the patent expires.
The situation may well be like the electronics industry in the 1950s and 1960s -- a few large corporations with extensive circuit patent portfolios built all the electronic devices, and avoided patent lawsuits by cross-licensing the portfolios back and forth. But little guys without a portfolio were effectively locked out. They couldn't afford to license the patented circuits they needed individually.
If this becomes the established practice, Microsoft and IBM and Sun and a few other companies will be able to write software "legally", but no one else will. I believe that RMS has written repeatedly that software patents have the potential to destroy the open-source software community.
Well, my point might be that there's a big difference (to me) between hacking Linux, the OS, and hacking Apache, a Web server. Or hacking some other Web server app running on Linux. Since I pay attention to what software gets run on my Linux boxes, and most of them don't run a Web server, this number may tell me nothing about how insecure my boxes might be.
Similarly, of course, a Windows box that is not running IIS or any other Web server app is not susceptible to hacks that exploit those apps.
An insecure app does not make the underlying OS insecure.
Having recently replaced the antique Linux box that functioned as our household router/NAT box for the past six years, I feel obligated to explain why it got replaced: for $40 plus tax, the new router/NAT consumer appliance uses 7.5W versus the 150W power supply, is completely silent versus the power supply in the old box that has always been fairly noisy, and occupies a few cubic inches next to the cable modem versus the cubic foot or better of the old desktop enclosure. On the down side, the appliance is not as flexible as the Linux box was, and certain network services (like printing) had to be moved to other computers.
As our children start to move out of the house, we are trying to simplify our lives and the household network. $40 seemed like a reasonable price to pay to replace the old box with something small and silent.
If Judge Jackson's ruling in the MS antitrust case to break the company in two had held up, my best guess at the time was that MS (the applications company) would have quickly introduced middleware that would run on multiple OSs to which their own apps were written and to which others could write apps. IMO, there was a good chance that
Over a very few years, said middleware API would have been able to displace the Windows API as the API of choice for commercial developers, particularly if MS allowed everyone to distribute it at no cost with their apps. Remember that the applications company was going to retain control of the development tools, and could introduce tools to aid in porting from Windows to the new middleware.
The middleware would run better on Linux and other UNIX-like OSs than it did on Windows due to better underlying kernel behavior -- scheduling, memory management, etc.
As a result, Windows (the operating system company) would have been marginalized quickly. Most of the benefits that the applications side of the house derived from control of the OS -- undocumented APIs, for example, that are not available to the rest of the world, or embedded functions like an HTML rendering engine -- could be realized in the middleware.
At the time, I believe my official prediction was that Bill Gates would stay with the applications company, with an unstated goal of putting the OS company out of business within five years.
Not a clue. Featureless molded gray plastic case just says "CompUSA". Similarly for all of the packaging materials. The manual, written in clear American English, is also just marked "Copyright CompUSA". I haven't opened the case to look at the circuit board for hints.
Just as a data point, I bought a CompUSA cable router last week (because it was $40 and all the other brands they carried were $50) and it insists on using 192.168.8/24 as the local subnet. No idea why they chose 8...
The author shows pie charts representing the fraction of total income going to the richest 20% of households. However, the US is a richer country in 2000 than it was in 1960. It seems reasonable to ask the question of whether the real income received by the lower 80% of households is greater in 2000 than it was in 1960, which it would be if the pie for 2000 is enough bigger -- 50% of a million dollars is more than 60% of $500,000. That is, the lower 80% of households may have lost ground to the upper 20%, but might still be better off than they were in 1960.
As someone once noted, the typical "poor" person in the US or Europe today enjoys better nutrition, health care, clothing, shelter, and education than the typical member of the nobility in the 1200s did. Maybe we are dividing into "elite" and "peasants" and the difference is getting wider. If the standard of living for the "peasants" is high enough, does that matter?
Follow-on thought. If Disney (for example) has just six stockholders (the ultra-wealthy), a CEO who insists that he/she get $200M in annual compensation instead of passing most of that on to the stockholders will find himself/herself out of a job very quickly. The current looting of corporations by their officers at the expense of the stockholders is, IMO, happening because so many of the shares are held by funds (mutual, pension and otherwise) that are run by individuals who are managing "other people's money."
One of the counter-arguments is the devastating effect this has on the demand side of the economy. If only capitalists (those who are rich enough to live off the income of their existing wealth, and their agents such as CEOs) can afford to go to the movies, it follows that either (a) ticket prices go through the roof (as in thousands of dollars) or (b) the studios go out of business -- a few hundred people paying $10 to see the movie will NOT cover the costs of the theatre, let alone movie production and distribution costs. In some fashion, supply and demand WILL balance.
If you're a pessimist, you think they will balance after an ugly downwards spiral into a world with a small number of the extremely rich and huge numbers of peasants. Well, and a certain number (not large) of people in the middle -- if you're one of the ultra-rich, you keep some doctors on staff to take care of you and your family. The doctors do better than peasants, but are probably true "wage slaves." And there has to be a small number of medical schools in order to insure the supply of doctors, etc.
If you're an optimist, you think that something less ugly will happen.
do you really want to be a technical writer for the next 30 years?
That's a good question -- do you really want to be any one thing for the next 30 years?
Last year I left (involuntarily) the cable/telecom industry after 25 years.
The job was fun for most of that time, in part because I was working for big enough companies that I could "change" jobs every few years.
Interestingly (or perhaps sadly),
none of the companies where I did that work still do any of the interesting parts -- everything has become very focused on cutting costs, and pretty much to hell with things like service improvement. As for me, I started work on a Ph.D. in economics this month -- something completely different.
My kids are almost adults and will be out on their own in a very few years.
Being a dad is fun and challenging, but I'll be happy to see them grown up and responsible for themselves.
Fortunately, that job also changes as the kids grow -- no one wants to be the parent of three-year olds for 30 years!
Still happily married after 22 years, and have no interest in changing that arrangement. Maybe that's the exception that proves the rule?
DeBeers doesn't import diamonds into the United States. American companies go to Europe or other locations where DeBeers can operate, buy the diamonds, and then import them. Or as I understand it for jewelry, in most cases buy finished diamonds from people who bought raw diamonds from DeBeers. I suppose you could stop the trade by totally banning the import of diamonds, but then you have to deal with putting all those jewelry companies out of business as well as killing various industrial applications that are dependent on diamond as a material.
A domestic diamond industry that can produce a sufficient quantity, especially for the industrial apps, might be a first step towards seeing such a ban. But I would be surprised if it ever happens.
NetGear has been VERY cooperative throughout this whole thing
I am NOT surprised that they have been cooperative. This appears to be a very clear case of negligence on NetGear's part doing significant injury to UW. IANAL, but have dealt with too many in the past, and this looks like a case where UW would have a very good chance of getting an immediate injunction against NetGear to stop current sales, and a good chance of forcing NetGear to track down and repair ALL of the defective product as an outcome of a trial.
I know of a case where a vendor of defective cable television boxes picked up the cost of the visits to customers' houses to replace 150,000 boxes rather than go to court at a cost of about $40 per visit. The cost to find and upgrade 700,000 deployed devices makes paying UW for bandwidth, some router capacity, and a few servers look pretty cheap.
Despite Bruce's statements about SCO asserting that they "own" the code, an earlier analysis of the IBM contract suggested that what SCO are actually claiming is that any code developed by IBM as part of a UNIX-derived system must not be revealed to third parties without SCO's permission. IBM can use it in any way they see fit in any product they want to develop, but can't reveal the source or describe the methods. If I understand properly, such developments acquire the status of trade secrets which IBM agreed to protect, but has not.
IBM apparently is claiming that they have a memo from AT&T releasing them from that restriction in their original contract. SCO seems to be claiming that Sequent and SGI had no such memos, so code developed by them and later acquired by IBM is not covered by the memo. If this mess actually gets to court, that part of it is going to be decided on the basis of obscure contract law, not IP stuff.
IMO, SCO threw in a bunch of this stuff on the general legal practice of accusing people that you sue of as many bad things as you can dream up. You don't have to win on more than one count in order to collect something, and sometimes you get lucky during discovery. Also note that unless you brought up a particular legal theory about damage during the original trial, you can't add it during an appeal of the decision -- so unless they bring up copyright infringement now, they can't bring it up in an appeals court later.
I know Boies has a big reputation, but I have to wonder about the quality of the legal advice that his firm is giving SCO. There seem to be a bunch of downside risks to their attempts to go after more than IBM. With IBM, it's just a contract dispute; with a bunch of the other threats they're making, there are risks of unfair trade practice investigations by the FTC, pump-and-dump investigations by the SEC, etc.
How far from the buildings are the tanks required to be located? When I lived in an area where propane was common, the tank had to be at least 100 feet away from any structures, the pipe had to be buried to some minimum depth, etc. In my current neighborhood, there are no locations that would meet the 100 foot requirement.
I certainly agree that in more rural settings, fuel cells and monthly propane delivery might be a much more efficient source of electricity than the current arrangements.
Lots of stories about home fuel cells powered by natural gas, like
this one.
No trucks, since most local codes would not allow you to store two months worth of liquified natural gas in your garage or backyard. Heavy dependence on the natural gas delivery pipes. Some potential problems (all amenable to solution, I believe, just be prepared to spend money):
In part because so many electric generating companies have decided to use natural gas for their newest plants, there are forecasts of shortages and substantial price increases starting this winter. Such shortages would be worse if there was a sudden large demand for gas to generate household electricity in areas currently using coal or petroleum.
Overseas transport of gas is much more difficult than petroleum. IIRC, Saudi Arabia produces enormous amounts of gas as a byproduct of their oil wells. Shipping it is so complex and expensive that they simply burn it off at the wells rather than trying to sell it.
Long-haul gas pipelines are subject to spectacular local failures. Recently saw one in action -- an 18" pipe ruptured and the gas ignited. Flames shooting several hundred feet into the sky. Impressive!
I do not believe that the national gas pipeline infrastructure has the same degree of interconnection that the power distribution grid has. This might be good -- no cascading failures. This might be bad -- lose one pipeline and large areas run out of gas/power as soon as local storage facilities are exhausted.
At one point when our Windows installations had gotten particularly unstable -- BSOD at least three times per day -- a colleague and I proposed a formal project for the time-tracking system for time spent on reboots after the first one each day. Whatever the problems were, rebooting was painfully slow and required manual responses to multiple dialog boxes. His estimate was that he was spending more than a half-hour per day on reboots, which was well above the normal threshold for a project number for an activity.
We didn't get a project number, but it did get IT to reformat the hard disks and do a clean install, something we were not allowed to do ourselves but had been asking for for weeks...
It has been reported that the US federal government alone spends several hundred million dollars each year on MS Office licenses. The 50 states undoubtedly spend considerable as well. Assuming those figures are true, I would assert that for a fraction of the current spending (let's say $100M per year) the governments could fund an organization that would produce and maintain comparable products. In addition to decreasing (rather than increasing) government spending, I suspect that there would be economic benefits if businesses stopped sending so much to Redmond for Word and Excel and were able to invest the money in other ways.
Do you think it's impractical to produce and maintain an office suite of software with a $100M annual budget?
It might be an interesting legal tussle. Red Hat claims that they are being damaged because SCO will not reveal which portions of the kernel code contain the potential "trade secrets" (my interpretation of the current status of IBM/SCO is that SCO essentially claims that IBM's additions to UNIX are such secrets because IBM was contractually forbidden from revealing them to third parties). SCO will no doubt claim that by publicly identifying the code containing the secrets, they are further damaged, perhaps in the sense that it makes their case against IBM more difficult. I could see a judge ruling either way.
You left out some of the nightmare scenarios if SCO wins the IBM case -- that Red Hat and other distributors contributed to the damage suffered by SCO and should pay, or that because of the wide distribution of the SCO trade secrets, Red Hat and other distributors must cease distributing the product until they have a version that is demonstrably clean (and any developers who have read the IBM-derived code are tainted already), or that Red Hat should bear the costs of upgrading all users of tainted kernels.
This comment raises a good point -- different people learn things differently. Some do well by reading, some do better if they can listen. What situation fits you best? While I can learn and have learned math strictly from a textbook, I find that it is easier when I can listen to someone doing the explanation while I look at the figures and/or equations. If you're a person who needs to listen, definitely look into a local community college. Try to find out about the instructor first, though -- I've seen very good CC instructors, and some that are terrible.
Other small points. If you're doing it on your own with a book (or books), make sure you've got problem solutions available to you; that provides a quick check when you've misunderstood something. And if you have a friend who can give you an occasional tutoring session when you get stuck, take advantage of it. Many people are frustrated teachers deep down inside, and get a real kick when they can see the "lightbulb go on" for someone.
Jobexport is a sign that production is too expensive and/or people simply suck... it seems that the jobs that are given away are because there are no more quality developers!
You seem to be asserting at least the possibility of a shortage of qualified developers in the US. I don't recall reading any cases of a company that claimed they were moving development offshore because they could not find high-quality talent locally -- they all claim that they need to lower costs in order to be competitive. It seems unlikely that so many countries (since offshoring has appeared in Western Europe as well as the US) have suddenly and simultaneously developed a shortage of quality developers. This only makes sense when cost is factored in -- compared to India (as an example), there is a shortage of quality developers in the US (or Western Europe) who will accept $12,000 and no benefits as compensation for a year's work.
Large companies in the developed countries have always operated under some social obligations. In Western Europe, many of these were enforced by law -- it's hard to fire people, the mandated vacations and pensions are large, etc. In Japan there have been very strong traditions about lifetime employment, even if not codified. The US has probably the weakest set of such obligations, although until the last 25 years many people expected to be able to work at a single large company for an entire career. Companies appear to be dumping those obligations as quickly as they can -- more quickly in the US, but all of the developed countries are experiencing similar problems, even if the scale is not as great.
I guess I would come at it from a different direction -- is it surprising how few of their ventures have succeeded? The list of products would have to include:
DOS
Windows
Office
Visual Studio
IE
Windows Media Player
MSN
XBox
DOS and IE were initially products purchased from another company. Of the components that make up Office, I believe that PowerPoint was purchased from another company (and not sure, but want to say the same thing about Excel). WMP is given away for free (if you bought Windows). MSN takes in money, but I believe it is not profitable. Ditto for XBox. Visual Studio is probably profitable, but they don't sell a lot of copies relative to the market for Windows and Office.
If I were an investor, I would be concerned that the most successful software company in the world has so few successful products, and that even fewer of those were initially developed internally.
I believe my 20-year-old steel-framed road bike comes in at 27 pounds. IIRC, the frame makes up a surprisingly small portion of the total -- wheels, tires, gears, chain, handlebars, brakes, etc. add up to quite a bit (although the picture in the article shows bamboo handlebars).
For manufacturers in developing countries, one other potential advantage of steel over aluminum is the ease with which it can be worked. Bent steel can be straightened (bent aluminum tubes show a pronounced tendency to break when you try to straighten them). Welding steel is very easy, welding aluminum more difficult.
The article describes the frame as being glued together. Glue that reliably binds bamboo to metal (in the picture, all the frame joints appear to be metal fittings) may not be made locally, or be all that "green". Does anyone know what goes into the manufacture of epoxy glues?
As long as there have been fairly modern economies (the last few hundred years, at least), capital and labor have fought over how they would share the proceeds of their collective effort.
Over much of the last century, labor has done better than usual in the United States.
Labor unions made a lot of progress in the early part of the century (eg, laws guaranteeing the right to organize).
FDR's New Deal emphasized putting people to work rather than corporate profits.
Consumer demand was enormous following WWII, the rest of the world was in no shape to produce consumer goods, so there was a labor shortage.
More recently, capital has taken advantage of its strengths.
First, capital is more mobile than labor is; it's much easier to move the factory from Michigan to Mississippi (or to Malaysia) than it is for the workers to pick up and move.
Second, capital knows what its goals are, and many parts of labor aren't sure: how many people have argued as long and consistently for reducing taxes on earnings as the capitalists have argued for reducing taxes on dividends?
Third,
capital has never been slow to spend money on equipment if they could eliminate labor -- fewer workers means less of the proceeds have to be shared.
Fourth, there's a worldwide labor glut -- guesses of the number of unemployed in China are about 150M, almost as many people as work in the U.S.
What remains to be seen is whether capital figures out that they need labor to get a big enough share to buy the goods and make the whole process work. You can be too greedy for your own long-term good.
I suspect that they say, "We represent the rapper Usher" and "We claim this MP3 is an infringement of that Usher's work." They are correct in the first statement (so there's no perjury), but wrong in the second (so there's no case). If they are blatent enough and frequent enough with the second error, there will eventually be changes in the law to attach penalties to failure to verify the content before sending the notices (which should have been in there from the beginning).
Slashdot Mirror
Indeed, electronics (in particular, semiconductors and integrated circuits) did eventually explode, making very substantial changes in all parts of our lives. However, there was a lengthy period when it was controlled by that small number of large companies and such innovation did not occur. I would not like to see a 20-year period where software is controlled by a handful of companies before breaking free again.
The situation may well be like the electronics industry in the 1950s and 1960s -- a few large corporations with extensive circuit patent portfolios built all the electronic devices, and avoided patent lawsuits by cross-licensing the portfolios back and forth. But little guys without a portfolio were effectively locked out. They couldn't afford to license the patented circuits they needed individually.
If this becomes the established practice, Microsoft and IBM and Sun and a few other companies will be able to write software "legally", but no one else will. I believe that RMS has written repeatedly that software patents have the potential to destroy the open-source software community.
Similarly, of course, a Windows box that is not running IIS or any other Web server app is not susceptible to hacks that exploit those apps.
An insecure app does not make the underlying OS insecure.
As our children start to move out of the house, we are trying to simplify our lives and the household network. $40 seemed like a reasonable price to pay to replace the old box with something small and silent.
As a result, Windows (the operating system company) would have been marginalized quickly. Most of the benefits that the applications side of the house derived from control of the OS -- undocumented APIs, for example, that are not available to the rest of the world, or embedded functions like an HTML rendering engine -- could be realized in the middleware.
At the time, I believe my official prediction was that Bill Gates would stay with the applications company, with an unstated goal of putting the OS company out of business within five years.
Not a clue. Featureless molded gray plastic case just says "CompUSA". Similarly for all of the packaging materials. The manual, written in clear American English, is also just marked "Copyright CompUSA". I haven't opened the case to look at the circuit board for hints.
Just as a data point, I bought a CompUSA cable router last week (because it was $40 and all the other brands they carried were $50) and it insists on using 192.168.8/24 as the local subnet. No idea why they chose 8...
As someone once noted, the typical "poor" person in the US or Europe today enjoys better nutrition, health care, clothing, shelter, and education than the typical member of the nobility in the 1200s did. Maybe we are dividing into "elite" and "peasants" and the difference is getting wider. If the standard of living for the "peasants" is high enough, does that matter?
Follow-on thought. If Disney (for example) has just six stockholders (the ultra-wealthy), a CEO who insists that he/she get $200M in annual compensation instead of passing most of that on to the stockholders will find himself/herself out of a job very quickly. The current looting of corporations by their officers at the expense of the stockholders is, IMO, happening because so many of the shares are held by funds (mutual, pension and otherwise) that are run by individuals who are managing "other people's money."
If you're a pessimist, you think they will balance after an ugly downwards spiral into a world with a small number of the extremely rich and huge numbers of peasants. Well, and a certain number (not large) of people in the middle -- if you're one of the ultra-rich, you keep some doctors on staff to take care of you and your family. The doctors do better than peasants, but are probably true "wage slaves." And there has to be a small number of medical schools in order to insure the supply of doctors, etc.
If you're an optimist, you think that something less ugly will happen.
Last year I left (involuntarily) the cable/telecom industry after 25 years. The job was fun for most of that time, in part because I was working for big enough companies that I could "change" jobs every few years. Interestingly (or perhaps sadly), none of the companies where I did that work still do any of the interesting parts -- everything has become very focused on cutting costs, and pretty much to hell with things like service improvement. As for me, I started work on a Ph.D. in economics this month -- something completely different.
My kids are almost adults and will be out on their own in a very few years. Being a dad is fun and challenging, but I'll be happy to see them grown up and responsible for themselves. Fortunately, that job also changes as the kids grow -- no one wants to be the parent of three-year olds for 30 years!
Still happily married after 22 years, and have no interest in changing that arrangement. Maybe that's the exception that proves the rule?
A domestic diamond industry that can produce a sufficient quantity, especially for the industrial apps, might be a first step towards seeing such a ban. But I would be surprised if it ever happens.
I know of a case where a vendor of defective cable television boxes picked up the cost of the visits to customers' houses to replace 150,000 boxes rather than go to court at a cost of about $40 per visit. The cost to find and upgrade 700,000 deployed devices makes paying UW for bandwidth, some router capacity, and a few servers look pretty cheap.
Despite Bruce's statements about SCO asserting that they "own" the code, an earlier analysis of the IBM contract suggested that what SCO are actually claiming is that any code developed by IBM as part of a UNIX-derived system must not be revealed to third parties without SCO's permission. IBM can use it in any way they see fit in any product they want to develop, but can't reveal the source or describe the methods. If I understand properly, such developments acquire the status of trade secrets which IBM agreed to protect, but has not.
IBM apparently is claiming that they have a memo from AT&T releasing them from that restriction in their original contract. SCO seems to be claiming that Sequent and SGI had no such memos, so code developed by them and later acquired by IBM is not covered by the memo. If this mess actually gets to court, that part of it is going to be decided on the basis of obscure contract law, not IP stuff.
IMO, SCO threw in a bunch of this stuff on the general legal practice of accusing people that you sue of as many bad things as you can dream up. You don't have to win on more than one count in order to collect something, and sometimes you get lucky during discovery. Also note that unless you brought up a particular legal theory about damage during the original trial, you can't add it during an appeal of the decision -- so unless they bring up copyright infringement now, they can't bring it up in an appeals court later.
I know Boies has a big reputation, but I have to wonder about the quality of the legal advice that his firm is giving SCO. There seem to be a bunch of downside risks to their attempts to go after more than IBM. With IBM, it's just a contract dispute; with a bunch of the other threats they're making, there are risks of unfair trade practice investigations by the FTC, pump-and-dump investigations by the SEC, etc.
I certainly agree that in more rural settings, fuel cells and monthly propane delivery might be a much more efficient source of electricity than the current arrangements.
Lots of stories about home fuel cells powered by natural gas, like this one. No trucks, since most local codes would not allow you to store two months worth of liquified natural gas in your garage or backyard. Heavy dependence on the natural gas delivery pipes. Some potential problems (all amenable to solution, I believe, just be prepared to spend money):
At one point when our Windows installations had gotten particularly unstable -- BSOD at least three times per day -- a colleague and I proposed a formal project for the time-tracking system for time spent on reboots after the first one each day. Whatever the problems were, rebooting was painfully slow and required manual responses to multiple dialog boxes. His estimate was that he was spending more than a half-hour per day on reboots, which was well above the normal threshold for a project number for an activity.
We didn't get a project number, but it did get IT to reformat the hard disks and do a clean install, something we were not allowed to do ourselves but had been asking for for weeks...
Do you think it's impractical to produce and maintain an office suite of software with a $100M annual budget?
You left out some of the nightmare scenarios if SCO wins the IBM case -- that Red Hat and other distributors contributed to the damage suffered by SCO and should pay, or that because of the wide distribution of the SCO trade secrets, Red Hat and other distributors must cease distributing the product until they have a version that is demonstrably clean (and any developers who have read the IBM-derived code are tainted already), or that Red Hat should bear the costs of upgrading all users of tainted kernels.
This comment raises a good point -- different people learn things differently. Some do well by reading, some do better if they can listen. What situation fits you best? While I can learn and have learned math strictly from a textbook, I find that it is easier when I can listen to someone doing the explanation while I look at the figures and/or equations. If you're a person who needs to listen, definitely look into a local community college. Try to find out about the instructor first, though -- I've seen very good CC instructors, and some that are terrible.
Other small points. If you're doing it on your own with a book (or books), make sure you've got problem solutions available to you; that provides a quick check when you've misunderstood something. And if you have a friend who can give you an occasional tutoring session when you get stuck, take advantage of it. Many people are frustrated teachers deep down inside, and get a real kick when they can see the "lightbulb go on" for someone.
Large companies in the developed countries have always operated under some social obligations. In Western Europe, many of these were enforced by law -- it's hard to fire people, the mandated vacations and pensions are large, etc. In Japan there have been very strong traditions about lifetime employment, even if not codified. The US has probably the weakest set of such obligations, although until the last 25 years many people expected to be able to work at a single large company for an entire career. Companies appear to be dumping those obligations as quickly as they can -- more quickly in the US, but all of the developed countries are experiencing similar problems, even if the scale is not as great.
DOS and IE were initially products purchased from another company. Of the components that make up Office, I believe that PowerPoint was purchased from another company (and not sure, but want to say the same thing about Excel). WMP is given away for free (if you bought Windows). MSN takes in money, but I believe it is not profitable. Ditto for XBox. Visual Studio is probably profitable, but they don't sell a lot of copies relative to the market for Windows and Office.
If I were an investor, I would be concerned that the most successful software company in the world has so few successful products, and that even fewer of those were initially developed internally.
For manufacturers in developing countries, one other potential advantage of steel over aluminum is the ease with which it can be worked. Bent steel can be straightened (bent aluminum tubes show a pronounced tendency to break when you try to straighten them). Welding steel is very easy, welding aluminum more difficult.
The article describes the frame as being glued together. Glue that reliably binds bamboo to metal (in the picture, all the frame joints appear to be metal fittings) may not be made locally, or be all that "green". Does anyone know what goes into the manufacture of epoxy glues?
As long as there have been fairly modern economies (the last few hundred years, at least), capital and labor have fought over how they would share the proceeds of their collective effort. Over much of the last century, labor has done better than usual in the United States. Labor unions made a lot of progress in the early part of the century (eg, laws guaranteeing the right to organize). FDR's New Deal emphasized putting people to work rather than corporate profits. Consumer demand was enormous following WWII, the rest of the world was in no shape to produce consumer goods, so there was a labor shortage.
More recently, capital has taken advantage of its strengths. First, capital is more mobile than labor is; it's much easier to move the factory from Michigan to Mississippi (or to Malaysia) than it is for the workers to pick up and move. Second, capital knows what its goals are, and many parts of labor aren't sure: how many people have argued as long and consistently for reducing taxes on earnings as the capitalists have argued for reducing taxes on dividends? Third, capital has never been slow to spend money on equipment if they could eliminate labor -- fewer workers means less of the proceeds have to be shared. Fourth, there's a worldwide labor glut -- guesses of the number of unemployed in China are about 150M, almost as many people as work in the U.S.
What remains to be seen is whether capital figures out that they need labor to get a big enough share to buy the goods and make the whole process work. You can be too greedy for your own long-term good.
I suspect that they say, "We represent the rapper Usher" and "We claim this MP3 is an infringement of that Usher's work." They are correct in the first statement (so there's no perjury), but wrong in the second (so there's no case). If they are blatent enough and frequent enough with the second error, there will eventually be changes in the law to attach penalties to failure to verify the content before sending the notices (which should have been in there from the beginning).