I completely agree: getting Windows XP "Professional" for a mere $1.80 would be enough to sabotage the intellectual development of any nation. Trash like that should be made so expensive that it is out of the reach of most people.
Let's hope the Eastern European legislatures crack down on this kind of problem. When Windows XP costs $200, as God intended it to, then open source software will look a lot more attractive to those people.
This is such a modest update that it won't cause a lot of people to upgrade. For under $2000, you already get PC laptops with 1440x1050, and some of the higher end PC laptops have 1600x1200 screens for less money than the PowerBook. I think Apple really needs to come out with a PowerBook that has a 1600x1024 screen and at least a 1GHz processor.
Exxon Valdez: The environmental cost was enormous. The world would have been better off if the spill hadn't happened.
And that's my point: yes, the Exxon Valdez spill generated lots of jobs and economic activities, but generating economic activity isn't useful in and of itself if there are externalities or opportunity costs involved. As another example, East Germany used to have lots of jobs just to keep people employed, like people who would stand at escalators to warn people about reaching the end of the escalator.
By analogy, deploying Microsoft operating systems for a nation-wide infrastructure may generate lots of MS sysadmin jobs, but thousands of Mexicans clicking away mindlessly at Microsoft dialog boxes is not productive work when the same tasks can be automated on Linux. The same people would be better off learning real computer science and programming skills and developing the next generation of killer apps and operating systems.
Windows machines are pretty easy to administer, which is a big plus.
Windows machines are not easier to administer if you have to administer a lot of them. (It is debatable whether they are easier to administer individually; tools like Webmin actually present a more unified and streamlined view of Linux administration than anything Microsoft offers, IMO.)
IE6 offers a state of the art web browsing experience. I recall the first time I browsed with IE after a few years of Netscape 4.x on linux -- I realized where all the hype about the internet came from -- IE was a richer experience. Mozilla has come a long way, but IE still takes the cake in my opinion... don't get me wrong, I really want to start liking Mozilla best... the lizzard is inching closer and closer.
Mozilla 1.0 is as good, if not better than, IE6, and there are several other good web browsers for Linux out there. In the Mexican case, much of the content is going to be authored for the chosen browser anyway.
But, what's more important, Mozilla on Linux is much easier to transform into a reliable and robust public, multiuser web client platform than IE on Windows.
Have you actually tried using public Windows web access terminals? They are full of security holes and privacy problems, in large part because Windows simply has no good infrastructure for supporting multiple users. I have yet to see a single public Windows-based web access terminal that I would trust with important information.
Are you saying that Mexico will be better off if the deal with Microsoft doesn't happen?
Absolutely. Microsoft contributes $6 million, but actually likely generates many times that in business and revenue, money that will ultimately be paid by the Mexican tax payer. Overall, the deal is very costly for Mexico in the long run. And for what? A platform that is less secure, requires more costly hardware, and is harder to administer (at least in bulk) than Linux.
Remember, we are not talking here about someone with a home machine running zillions of games, oddball multimedia plug-ins, or requiring nitty-gritty office suite compatibility. We are talking about a robust, widely-deployed, multi-user, secure Internet access infrastructure. Something that doesn't require upgrades every few months. Something that you can put down and that works for years to come. Microsoft simply has nothing to offer in that space. UNIX, and by extension Linux, have served exactly that need for nearly two decades, and UNIX and Linux have been hardened in generations of attacks on college campuses, in financial institutions, and on the Internet.
This kind of OSS elitism is really absurd in this case b/c Microsoft products generally have better UI standardization and they are generally easier to learn/integrate, plus they are embraced by more businesses, which makes them more valuable for those seeking (mostly non-technical) jobs.
No, what is really absurd is your brand of bean-counting cost-benefit analysis. If you look at its overall contribution to the economy, the Exxon Valdez disaster was a boon for the economy: it created jobs and stimulated economic activity. You need to look at the details of some endeavor in order to determine its effect, not just at money.
Microsoft's envisioned role is in contributing software to run Internet servers and web access terminals. But Windows machines are not the most important Internet server platform. Furthermore, Windows machines make lousy web access terminals: they are hard to secure and unreliable. On top of that, maintenance of Windows servers and Windows clients is very costly and labor-intensive, and while that creates jobs, the labor and money that is wasted on maintaining those machines could be more usefully redirected to other purposes.
Linux handles both tasks, Internet servers and public web access terminals, beautifully and at a much better TCO than Windows.
That's complete nonsense. Using the GPL doesn't make your software belong to the FSF. In fact, many companies (foremost, Troll Tech) dual-license their software under some commercial license and the GPL.
You do need to sign over your copyright to the FSF if you want the FSF to distribute the software and assume maintenance for it. That has nothing to do with the GPL.
When people are offering you something for free, it's pretty rude to complain that they're not offering you even more.
It is decidedly not rude, however, to explain to others what the problems with a self-proclaimed "open source" license are and why they shouldn't use the code either. It is also not rude to explain to the authors, politely, why one can't use the license the way it is; that may help the authors figure out how they might be able to grow their user community.
You can't compare sending people to the gulags with forcing monopoly choice of operating systems on them.
I didn't make that comparison, you did. At issue isn't how many people Stalin killed, but his style of economic organization. Stalin built the Soviet economy by single-mindedly focussing on political unity, productivity, and centralization, which is quite analogous to the style of non-competition and centralization Gates apparently wants in the computer industry. And that's a quite valid comparison because, in the end, whether pushed by Gates or Stalin, this economic strategy is doomed to failure, even if it looks attractive at first. What's particularly obnoxious about Gates is that he proclaims himself a free-market champion when what he is arguing for is Soviet-style centralization in the computer software industry.
Plan 9 has some interesting ideas, and the code is probably clearnly written. It would be nice if it were an open source operating system so that people could use it, built commercial systems on top of it, contribute to it, and enhance it.
Unfortunately, the Plan 9 license is unacceptable, as Stallman and Myers point out. And it doesn't look like that's going to change either.
The Soviet Union was Good because it kept consumers from getting confused by all those competing offers. If you were lucky, you got the care you were assigned, the apartment you were assigned, and the health care you were assigned. You didn't have much, but at least you weren't confused.
Microsoft is the same way: they don't give you much, but they are going to fight tooth and nail to keep you from getting confused by too much choice. Come to papa Gates, he'll take care of you, just like papa Stalin did before.
I think a lot of today's claims to something like Star Wars are based on trademarks. So, preserving legal protection for their trademark characters may be important to them.
Ever try to find a open source tax preperation program? Doesn't exist.
You can do taxes over the web from Linux and other free operating systems just fine. But taxes are a special case anyway (highly legalistic, highly time constrained, of no independent interest to scientists or programmers). Scientific and educational software is about as different as you can get.
True, you can get lots of programs from the open source world, but the more specialized the programs get, the less likely you will find a free alternative.
There are plenty of very specialized programs that you can only get for free. In fact, most research software starts out that way before some company picks it up, makes it closed source, and generally ends up making it much less useful.
These programs normally take a higher expertise level (ie, you need to be a chemistry expert to design a feasible chemistry app), and the open source need just isn't there.
Scientists who develop software as part of publically funded grants, or who want to publish results related to their software, should be required to make the software available for free: it's necessary for experimental reproducibility, and why should the tax payer fund private software companies anyway?
Many scientists appreciate those reasons. And many scientists don't want to become software entrepreneurs anyway and publish their software even if they could commercialize it.
And your average unversity isn't going to spend tens of thousands of dollars in salary to develop a complex app and then give it away for free to their competitors (ie, other universities).
Universities generally don't spend money on developing science-related software; funding agencies do. Universities are trying to get into the act by asserting rights to software they didn't pay for, but we shouldn't let them get away with that. In fact, these days, it's often the universities that try to close source against the wishes of researchers and funding agencies.
If you have an MS site license, most of your worries with the BSA go away, since you then can easily produce a license for the OS and application software on most machines. The few additional software packages that the BSA cares about are much easier to keep track of than thousands of individual Windows and Office licenses.
Besides, the issues are unrelated. Even if you have paid for a Windows site license, you might still want to switch over as many machines to Linux as possible: it saves you adminstrative costs and gives you better performance. And maybe, eventually, you can drop the Windows site license.
While the idea of a campus that's totally open source is cute, the idea is totally unworkable and not a feasible solution
You don't have to go completely open source either. Keep a few Windows PCs and Macs with the proprietary stuff and let the BSA worry about those. You can fix the bulk of the problem by converting the bulk of the machines completely to open source software; the BSA can spend as much time as they want crawling around those machines.
We'd all like free software. However, with very rare exceptions, the best (or all) software in most domains is closed. Why? Because I can't find enough chemistry people and programmers who will cooperate to make me specialized software of superb quality unless I unload a big pile of cash.
That's complete BS. There is plenty of excellent, free software in almost all areas. There are some specialized areas where the only choice may be something proprietary and closed source, but most work in most departments at a university can run completely on open source software. That includes computer science, engineering, language tutoring, graphic arts, writing, chemistry, and other fields. Of course, it does take a little more thought by a faculty member to select a non-advertised open source solution over an equivalent heavily-advertised closed-source solution.
If I was still in high school, it would be a no-brainer to decide not to go to any school that didn't use any proprietary software.
Yes, and that demonstrates just about how much brains you have.
I was looking at doing some audio hacks with a compressed format. I was trying hard to find an Ogg Vorbis specification but couldn't find any. Eventually, I gave up and used MP3. Even if I had reverse-engineered the Ogg Vorbis en/decoder, without an official specification, there would have been no guarantee that things would have remained compatible or that I wouldn't have missed some important subtlety.
I find this pretty sad. The Segway is a neat gizmo, but it doesn't look like very attractive transportation. Bicycles and electrically assisted bicycles are faster, can be used on the road, and allow you to engage in some level of physical activity. Scooters (like the Vespa or Honda) have a much greater range and are about half the price, and they are mature and low-tech. Or you can get a variety of electric scooters, which are quiet and faster as well. The Segway, in contrast, is slow, can't be used on roads, provides no opportunity for exercise, has limited range, and is quite expensive. And, as the article points out, is probably quite dangerous on sidewalks.
Electrically assisted bicycles like this are in a legal limbo: you can't legally use them on bike paths in many places, you can't ride them as fast as a bicycle, and some variants require motorcycle licenses. Yet, they are suitable and highly efficient for travel and commuting.
Similarly, innovative scooters like the BMW C1 (a scooter with a secure passenger cell) fail to get approval in the US, even though they passed lots of safety and real-world tests in Europe.
Yet, a high-tech gizmo like the Segway just breezes through regulatory approvals, probably based on the excellent marketing and celebrity endorsements behind it. First, drivers have to engage in an arms-race with SUVs on the road; will pedestrians now all have to upgrade to Segways in order to use sidewalks safely?
Many H1B workers are US-educated: they have come here on student visas and then transition into the workforce. The ones that aren't US-educated usually come from excellent universities in their home countries. So, in my experience, these people tend to be highly qualified. Note that the H1B visa program requires employers to document worker qualifications and to make an effort to find similarly qualified US workers prior to hiring a foreign candidate. I won't pretend that the system is perfect, but the government does make an effort.
Nanotechnology has moved to the same secret labs in the government where you find partially dissected alien bodies, antigravity machines, eternal motion machines, and the fountain of youth.
In different words, nanotechnology is a lot of hot air. It has utterly failed to deliver on its promises: universal replicators, and the like, and there is no light at the end of the tunnel. The successes attributed to nanotechnology have instead come from traditional fields like materials science, physics, VLSI, micromachines, and molecular biology.
You can't usually return the software if you don't agree. If you can return it, it's usually going to cost you much more time than the amount you paid for it.
Often, the EULA that comes with the software is out of date anyway: companies reserve the right to change terms and conditions at any point, so what you read and agree to may already be out of date. Do you have the time and interest to check whether there is a newer version? Most people don't.
For most web sites, you are a fool if you give them any information you care about. For the few web sites where you need to give a correct address and a credit card number, reputation probably counts more than EULA, but there is one case where you might want to check about any unexpected charges and marketing tie-ins.
Now, can anybody hold you to the EULA? Not really: unless you try to do something with the software that is somehow visible to other people, nobody knows that you are using the software, let alone that you have agreed to the EULA, so many provisions are meaningless. If you plan on reverse engineering the software, building applications with it, or linking with it, you might have to worry about it, but then, you might be better off not living in the US. And the law already protects you against highly one-sided contracts (Bill Gates may write into his EULA that you agree to be be his towel boy for a year, but he'd have a hard time enforcing that).
Most of these problems and issues with EULAs are unrelated to their readability; even if they were highly readable, there still wouldn't be much point in reading them.
I mean, where do you go from there? Or what about medical doctors? Or plumbers? Or construction workers? Or lawyers?
The notion of "promotion" is seriously overrated anyway. Do you really want to spend your days talking to whiny investment bankers, composing meaningless vision statements, having half your company snicker about you behind your back, having all stress and no free time, and managing people problems? If you do, go right ahead and aspire to that management position. But there is a reason those positions are paid highly: it's hazard pay for dirty work most people don't want to do.
Seriously, people do what they like, what they get paid for, and what they are good at. Many people who aren't qualified as programmers would love to have a $80k/year "dead-end job" with full benefits.
As for the supposed age limit, jobs going off-shore, and all that, in my opinion, Matloff is a loony. His claims are poorly supported by data and contradict what people who actually try to hire programmers experience. Sure, occasionally, you'll see age discrimination, and occasionally you'll see companies taking advantage of immigration issues. But the former is already covered by non-discrimination statutes, and the latter has been addressed with H1B portability and faster green-card approvals. Jobs will probably continue to go off-shore, but the best way of stemming that is to bring the qualified programmers from those other countries to the US; if you force them to go back to their countries of origin, they won't become farmers, they'll create a thriving and competitive software industry there.
Assuming a lower cost barrier to entry and an ease of unsubscribing with no penalties, it benefits consumers to buy into this model.
The company wouldn't be doing this if they expected to make less money. So, the two ways in which it can help them is (1) attract more customers, or (2) get more money from the same customers they would get at their current model. I'm not so sure that this will attract a lot more customers...
Slashdot Mirror
Let's hope the Eastern European legislatures crack down on this kind of problem. When Windows XP costs $200, as God intended it to, then open source software will look a lot more attractive to those people.
What are you doing while sitting at your screen??? Do you practice different positions from the Kama Sutra while programming or--web browsing?
This is such a modest update that it won't cause a lot of people to upgrade. For under $2000, you already get PC laptops with 1440x1050, and some of the higher end PC laptops have 1600x1200 screens for less money than the PowerBook. I think Apple really needs to come out with a PowerBook that has a 1600x1024 screen and at least a 1GHz processor.
And that's my point: yes, the Exxon Valdez spill generated lots of jobs and economic activities, but generating economic activity isn't useful in and of itself if there are externalities or opportunity costs involved. As another example, East Germany used to have lots of jobs just to keep people employed, like people who would stand at escalators to warn people about reaching the end of the escalator.
By analogy, deploying Microsoft operating systems for a nation-wide infrastructure may generate lots of MS sysadmin jobs, but thousands of Mexicans clicking away mindlessly at Microsoft dialog boxes is not productive work when the same tasks can be automated on Linux. The same people would be better off learning real computer science and programming skills and developing the next generation of killer apps and operating systems.
Windows machines are not easier to administer if you have to administer a lot of them. (It is debatable whether they are easier to administer individually; tools like Webmin actually present a more unified and streamlined view of Linux administration than anything Microsoft offers, IMO.)
IE6 offers a state of the art web browsing experience. I recall the first time I browsed with IE after a few years of Netscape 4.x on linux -- I realized where all the hype about the internet came from -- IE was a richer experience. Mozilla has come a long way, but IE still takes the cake in my opinion... don't get me wrong, I really want to start liking Mozilla best... the lizzard is inching closer and closer.
Mozilla 1.0 is as good, if not better than, IE6, and there are several other good web browsers for Linux out there. In the Mexican case, much of the content is going to be authored for the chosen browser anyway.
But, what's more important, Mozilla on Linux is much easier to transform into a reliable and robust public, multiuser web client platform than IE on Windows.
Have you actually tried using public Windows web access terminals? They are full of security holes and privacy problems, in large part because Windows simply has no good infrastructure for supporting multiple users. I have yet to see a single public Windows-based web access terminal that I would trust with important information.
Are you saying that Mexico will be better off if the deal with Microsoft doesn't happen?
Absolutely. Microsoft contributes $6 million, but actually likely generates many times that in business and revenue, money that will ultimately be paid by the Mexican tax payer. Overall, the deal is very costly for Mexico in the long run. And for what? A platform that is less secure, requires more costly hardware, and is harder to administer (at least in bulk) than Linux.
Remember, we are not talking here about someone with a home machine running zillions of games, oddball multimedia plug-ins, or requiring nitty-gritty office suite compatibility. We are talking about a robust, widely-deployed, multi-user, secure Internet access infrastructure. Something that doesn't require upgrades every few months. Something that you can put down and that works for years to come. Microsoft simply has nothing to offer in that space. UNIX, and by extension Linux, have served exactly that need for nearly two decades, and UNIX and Linux have been hardened in generations of attacks on college campuses, in financial institutions, and on the Internet.
It seems the ignorant natives are easily amused by colorful glass beads.
No, what is really absurd is your brand of bean-counting cost-benefit analysis. If you look at its overall contribution to the economy, the Exxon Valdez disaster was a boon for the economy: it created jobs and stimulated economic activity. You need to look at the details of some endeavor in order to determine its effect, not just at money.
Microsoft's envisioned role is in contributing software to run Internet servers and web access terminals. But Windows machines are not the most important Internet server platform. Furthermore, Windows machines make lousy web access terminals: they are hard to secure and unreliable. On top of that, maintenance of Windows servers and Windows clients is very costly and labor-intensive, and while that creates jobs, the labor and money that is wasted on maintaining those machines could be more usefully redirected to other purposes.
Linux handles both tasks, Internet servers and public web access terminals, beautifully and at a much better TCO than Windows.
You do need to sign over your copyright to the FSF if you want the FSF to distribute the software and assume maintenance for it. That has nothing to do with the GPL.
It is decidedly not rude, however, to explain to others what the problems with a self-proclaimed "open source" license are and why they shouldn't use the code either. It is also not rude to explain to the authors, politely, why one can't use the license the way it is; that may help the authors figure out how they might be able to grow their user community.
I didn't make that comparison, you did. At issue isn't how many people Stalin killed, but his style of economic organization. Stalin built the Soviet economy by single-mindedly focussing on political unity, productivity, and centralization, which is quite analogous to the style of non-competition and centralization Gates apparently wants in the computer industry. And that's a quite valid comparison because, in the end, whether pushed by Gates or Stalin, this economic strategy is doomed to failure, even if it looks attractive at first. What's particularly obnoxious about Gates is that he proclaims himself a free-market champion when what he is arguing for is Soviet-style centralization in the computer software industry.
There are lots of different versions of Windows as well, but Gates will decide which one you get.
Unfortunately, the Plan 9 license is unacceptable, as Stallman and Myers point out. And it doesn't look like that's going to change either.
Microsoft is the same way: they don't give you much, but they are going to fight tooth and nail to keep you from getting confused by too much choice. Come to papa Gates, he'll take care of you, just like papa Stalin did before.
I think a lot of today's claims to something like Star Wars are based on trademarks. So, preserving legal protection for their trademark characters may be important to them.
Maybe now he'll volunteer for immediate contribution to the visible human project. Hey, it's only fair.
You can do taxes over the web from Linux and other free operating systems just fine. But taxes are a special case anyway (highly legalistic, highly time constrained, of no independent interest to scientists or programmers). Scientific and educational software is about as different as you can get.
True, you can get lots of programs from the open source world, but the more specialized the programs get, the less likely you will find a free alternative.
There are plenty of very specialized programs that you can only get for free. In fact, most research software starts out that way before some company picks it up, makes it closed source, and generally ends up making it much less useful.
These programs normally take a higher expertise level (ie, you need to be a chemistry expert to design a feasible chemistry app), and the open source need just isn't there.
Scientists who develop software as part of publically funded grants, or who want to publish results related to their software, should be required to make the software available for free: it's necessary for experimental reproducibility, and why should the tax payer fund private software companies anyway?
Many scientists appreciate those reasons. And many scientists don't want to become software entrepreneurs anyway and publish their software even if they could commercialize it.
And your average unversity isn't going to spend tens of thousands of dollars in salary to develop a complex app and then give it away for free to their competitors (ie, other universities).
Universities generally don't spend money on developing science-related software; funding agencies do. Universities are trying to get into the act by asserting rights to software they didn't pay for, but we shouldn't let them get away with that. In fact, these days, it's often the universities that try to close source against the wishes of researchers and funding agencies.
Besides, the issues are unrelated. Even if you have paid for a Windows site license, you might still want to switch over as many machines to Linux as possible: it saves you adminstrative costs and gives you better performance. And maybe, eventually, you can drop the Windows site license.
You don't have to go completely open source either. Keep a few Windows PCs and Macs with the proprietary stuff and let the BSA worry about those. You can fix the bulk of the problem by converting the bulk of the machines completely to open source software; the BSA can spend as much time as they want crawling around those machines.
We'd all like free software. However, with very rare exceptions, the best (or all) software in most domains is closed. Why? Because I can't find enough chemistry people and programmers who will cooperate to make me specialized software of superb quality unless I unload a big pile of cash.
That's complete BS. There is plenty of excellent, free software in almost all areas. There are some specialized areas where the only choice may be something proprietary and closed source, but most work in most departments at a university can run completely on open source software. That includes computer science, engineering, language tutoring, graphic arts, writing, chemistry, and other fields. Of course, it does take a little more thought by a faculty member to select a non-advertised open source solution over an equivalent heavily-advertised closed-source solution.
If I was still in high school, it would be a no-brainer to decide not to go to any school that didn't use any proprietary software.
Yes, and that demonstrates just about how much brains you have.
I was looking at doing some audio hacks with a compressed format. I was trying hard to find an Ogg Vorbis specification but couldn't find any. Eventually, I gave up and used MP3. Even if I had reverse-engineered the Ogg Vorbis en/decoder, without an official specification, there would have been no guarantee that things would have remained compatible or that I wouldn't have missed some important subtlety.
Electrically assisted bicycles like this are in a legal limbo: you can't legally use them on bike paths in many places, you can't ride them as fast as a bicycle, and some variants require motorcycle licenses. Yet, they are suitable and highly efficient for travel and commuting.
Similarly, innovative scooters like the BMW C1 (a scooter with a secure passenger cell) fail to get approval in the US, even though they passed lots of safety and real-world tests in Europe.
Yet, a high-tech gizmo like the Segway just breezes through regulatory approvals, probably based on the excellent marketing and celebrity endorsements behind it. First, drivers have to engage in an arms-race with SUVs on the road; will pedestrians now all have to upgrade to Segways in order to use sidewalks safely?
Many H1B workers are US-educated: they have come here on student visas and then transition into the workforce. The ones that aren't US-educated usually come from excellent universities in their home countries. So, in my experience, these people tend to be highly qualified. Note that the H1B visa program requires employers to document worker qualifications and to make an effort to find similarly qualified US workers prior to hiring a foreign candidate. I won't pretend that the system is perfect, but the government does make an effort.
In different words, nanotechnology is a lot of hot air. It has utterly failed to deliver on its promises: universal replicators, and the like, and there is no light at the end of the tunnel. The successes attributed to nanotechnology have instead come from traditional fields like materials science, physics, VLSI, micromachines, and molecular biology.
Often, the EULA that comes with the software is out of date anyway: companies reserve the right to change terms and conditions at any point, so what you read and agree to may already be out of date. Do you have the time and interest to check whether there is a newer version? Most people don't.
For most web sites, you are a fool if you give them any information you care about. For the few web sites where you need to give a correct address and a credit card number, reputation probably counts more than EULA, but there is one case where you might want to check about any unexpected charges and marketing tie-ins.
Now, can anybody hold you to the EULA? Not really: unless you try to do something with the software that is somehow visible to other people, nobody knows that you are using the software, let alone that you have agreed to the EULA, so many provisions are meaningless. If you plan on reverse engineering the software, building applications with it, or linking with it, you might have to worry about it, but then, you might be better off not living in the US. And the law already protects you against highly one-sided contracts (Bill Gates may write into his EULA that you agree to be be his towel boy for a year, but he'd have a hard time enforcing that).
Most of these problems and issues with EULAs are unrelated to their readability; even if they were highly readable, there still wouldn't be much point in reading them.
The notion of "promotion" is seriously overrated anyway. Do you really want to spend your days talking to whiny investment bankers, composing meaningless vision statements, having half your company snicker about you behind your back, having all stress and no free time, and managing people problems? If you do, go right ahead and aspire to that management position. But there is a reason those positions are paid highly: it's hazard pay for dirty work most people don't want to do.
Seriously, people do what they like, what they get paid for, and what they are good at. Many people who aren't qualified as programmers would love to have a $80k/year "dead-end job" with full benefits.
As for the supposed age limit, jobs going off-shore, and all that, in my opinion, Matloff is a loony. His claims are poorly supported by data and contradict what people who actually try to hire programmers experience. Sure, occasionally, you'll see age discrimination, and occasionally you'll see companies taking advantage of immigration issues. But the former is already covered by non-discrimination statutes, and the latter has been addressed with H1B portability and faster green-card approvals. Jobs will probably continue to go off-shore, but the best way of stemming that is to bring the qualified programmers from those other countries to the US; if you force them to go back to their countries of origin, they won't become farmers, they'll create a thriving and competitive software industry there.
The company wouldn't be doing this if they expected to make less money. So, the two ways in which it can help them is (1) attract more customers, or (2) get more money from the same customers they would get at their current model. I'm not so sure that this will attract a lot more customers...