Pacifism fails to reckon sometimes you need violence to forestall even more violence and other evils.
Or in this case you needed violence to ensure thirteen years of oppressive military dictatorship and frequent murder of members of the opposition, and assorted other evils.
Of course, some people believe that both Chile's and Cuba's economies would have been much better off if it hadn't been for the fact that the US spent huge resources destabilizing them.
While I don't have much sympathy for Castro - while he got rid of an oppressive dictator, he doesn't exactly seem to have kept to his ideals - Allende on the other hand was democratically elected, and still had to endure massive CIA interference, and was murdered thanks to the CIA. How that qualifies to being compared to Castro is beyond me.
And so much for the US government caring about democracy. Which makes the whole anti Castro hysteria from the US government seem rather hypocritical, but then hypocrisy seems to be a requirement to get elected in the US.
But still "revolutionary" in terms of the policies the government supported: one of very few regimes that branded itself socialist that actually stood by fundamental Marxist principles instead of just using socialist symbolism as a justification for oppression. In that sense the Allende government was certainly revolutionary.
Governments are THE largest procurers of proprietary developed one-off software solutions. Most large government systems are running on software developed particularly for that government application. Governments do that because the market doesn't build shrink wrap apps that cover their needs.
Why should they be barred from doing the same if an OS doesn't meet their needs?
And why should it be any different if their "need" is to stimulate the local economy?
I'd say a government that spends it's money on paying for local development that become freely available to anyone (including people outside their own country) is doing a hell of a lot better job than a government that spend the same amounts on proprietary development projects that will only benefit one department, and on licensing proprietary applications from a company outside their own country.
Paying for open source development is likely to do a lot more for their people.
Microsoft failing should have relatively little impact. First of all, Microsoft employs very few people compared to most companies that approach them in market cap. Microsoft employ 30-40 thousand people, compared to closing in on 300.000 for IBM and it's subsidiaries.
Secondly, Microsoft is hoarding cash. If Microsoft started losing money so fast that it would even manage to eat up the return on their capital investments, it would still take a long time before they'd collapse, so barring a sudden devastating move, it will take a VERY long time for Microsoft to end up anywhere near a bankrupcy court, meaning the economy would likely have adapted long ago.
The only way for Microsoft failing to have a major impact would be if they managed to conceal problems until they were near collapsing (lying to the SEC?), or if they were hit with a ridiculously devastating judgement in a class action lawsuit (as in 100-200 billion in damages, if lower they could likely finance their way out of it), which is highly unlikely.
More likely than Microsoft failing is Microsoft morphing. Due to it's financial strength, they keep on buying and investing outside their core areas. If Windows and Office, their two main cash cows, start collapsing on them, they'd likely have plenty of time to shift focus, and given their enormeous amount of cash and cash equivalents they'd likely have no problem surviving until they could live on other products, though probably not with anywhere near the same market cap.
That is a very simplistic view, and blatantly ignores that one of the points of the article you responded to was lack of proprietary software, that is often unlikely to support cluster configurations using PVM or MOSIX as well.
It also blatantly ignores that for a cluster, YOUR APPLICATION need to explicitly manage availability and reliability, or a single node can potentially take down everything if it fails, whereas a mainframe has extensive availability features built in. If you have a 16 node cluster, with 99.9% availability for node, your cluster as a whole would have an availability of approximately 98.4%. You'd be hard pressed to find a mainframe with availability that low.
A cluster is a valid solution where you are willing to deal with that issue, but for many types of applications availability like that simply isn't an option.
It's unclear to me what you want. If you want to make many machines look like one, then what you're looking for is Single System Image clustering - HP/Compaq/Whatever-their-name-is-today is working on providing that for Linux. Putting UML on OpenMosix simply doesn't make sense, as UML would need to virtualize a hell of a lot more than today and make multiple UML instances cooperate over OpenMosix just to run ONE instance of Linux spread over multiple machines. To run multiple instances of Linux, you would have to have multiple UML instances on each node you'd want this hypotetical OpenMosix enabled UML to spread a Linux instance out over.
To top that off, you would have multiple machines, each with pretty high error rates (compared to a mainframe), which would increase the error rate rather than lower it - unless you do a LOT of work, any one of the machines failing could potentially take down any Linux instance partly depending on that node.
Overcoming those issues is hard to do at OS level without tight hardware integration, which is why it's expensive and still only found in mainframes.
If it's "trickier to admin" then you've defeated the main advantage of these mainframes. Anyone can build a system with the IO throughput or computing power of a mainframe. The trick is to do it in a what thats cheap to admin.
If your work is batch oriented, or individual transactions/jobs aren't critical, then a mainframe is overkill, because a lot of your admin headaches go away simply by ignoring them - a failed node is simply replaced or restarted, possibly semi-automtically, failed hardware is just replaced etc.
The moment you need to start working around this to ensure 100% uninterrupted operation, a cluster starts becoming rapidly more expensive to admin.
Of course you need redundancy, but instead of having umpteen different servers you need backup for, you need only two Z-series servers, or you can quite likely achieve the redundancy you need by outfitting the Z-series machine properly.
The Z-series supports taking CPUs out of comission for replacement without downtime. Same for RAM. Multiple hot-swappable SCSI controllers connected to a fully redundant storage system such as the ESS/Shark (where you can connect to two separate banks of controllers, so that any one of them can be offline without causing problems, and which has two separate AIX servers handling requests, supports RAID and synchroneous mirroring over fiber to a backup ESS), multiple hot swappable network cards, multiple power supplies, and you start getting pretty safe.
Yes, it will cost money, but so will providing all of the above for standalone servers. The Z-series is marketed primarily as a way of reducing maintenance work by consolidating your "servers" on one or two physical platforms, not for it's purchase price - it's an expensive beast.
No big ships use the Thames anymore, at least not up as far as London. Docklands, what used to be some of the largest docks in Europe, have mostly been converted to luxury offices and housing. Large ships dock further east, as cheap train and road transport made the London docks uneconomical.
As for rough water, you won't find much of that either on the Thames around London
People would be mostly competing with a few barges and small tourist boats.
It's pretty well accepted by now, though, that Sauron and the ring represented the coming of machines to Middle Earth. Tolkien had a strong dislike for machines and technology, and made it pretty clear that that was where he had drawn his inspiration. So while it's not explicit in the books, it would seem quite valid to take it one step further and treat the ring as technology. Not being understood is not the same as not being understandable.
I would call Lord of the Rings a fantasy work, not sci fi, but the underlying theme of mans fear of the machines and what changes they brought is a typical sci fi theme.
And you really fail to see how this could lead to a thought provoking book, and decide to judge it on a few short sentences instead?
I haven't read this book, but sci fi is full of great books that resulted from the setting of "what if these two people or these two groups of people got a chance to meet under other circumstances".
In fact the above was Marx' core argument for the inevitability of the failure of capitalism.
The key result of capitalism is competition. The only measurement of the success of a company under capitalism is profit. Driving up profit means increasing sales, which can only be done as long as consumption increases, or the total market increases (population boom, or expanding into areas you don't currently reach).
The moment these factors are all constrained (population doesn't increase, companies reach all possible consumers, and consumers are consuming as much as they can), the ONLY remaining way to increase profit becomes fighting over market share, or reducing cost. Fighting over market share also increasingly IS an issue of reducing cost, and hence prices, as there is only so much you can do with marketing and product differentiation if someone is dramatically undercutting you.
Cutting cost inevitably boils down to reducing the amount paid to other people, because all resources and materials you pay for ultimately involve paying people, whether it is wages, licenses, purchase of property or any other transaction (even when you pay a corporation, you are then indirectly enriching the owners of the corporation, if a foundation or trust the beneficiaries, if a government, the people)
The logical conclusion is a strong push to cut workforces and/or cut pay. Often the second is a result of the former: People in areas where work is short, or with skills that are becoming obsolete will lower their salary expectations.
However, at some point you reach a level where any reduction in cost lead to a reduction in consumption, at which point reduction in cost for one company will be increasingly hard to compensate by growth elsewhere.
Marx' thesis was that at this point, capitalism will continue to produce, and continue to cut costs, and drive consumptions among the people with capital to extreme excesses by promoting waste that people wouldn't normally consider, while more and more people are pushed into poverty by cost cutting measures.
Capitalists on the other hand, dismiss this, usually by assuming that overall consumption can continue to grow forever, hence always allowing for cost cutting to be compensated by growth in other markets.
Taken to extreme, a society where "workers" aren't needed, capitalism is unlikely to survive. How do you maintain a system based on private ownership of the means of production when it leads to immense poverty, and that poverty isn't "needed" because of scarcity?
It is hard to see a situation like that not eventually leading to growing popular unrest.
Incidentally, in "The German Ideology" Marx wrote [paraphrased] "if the revolution happens in a country with insufficient resources to meet the basic needs, the same shit will start all over again" - Marx always made it very clear that for a socialist revolution to have a chance to succeed, it must happen in a highly evolved capitalist economy, a country where a small elite have accumulated sufficient wealth that the needs of the population as a whole could be met by redistribution, and where the wast majority had been forced into poverty by the more and more extreme competition of capitalist economy.
He specifically named the UK, France and Germany originally, but in a later preface to the Communist Manifesto, he pointed to the US with it's rapid growth and expanding markets as more likely to mature to the sufficient level first....
Interestingly, he also specifically made it clear that he believed that a socialist revolution in Russia would be doomed to failure because of it's low level of development (it was a feudal dictatorship with a mostly agrarian economy).
You missed a crucial part of the post you replied to. There is precedent at least in the US that limits the copyright protection on code to the expression of the ideas, not the function of the program itself, hence the question is whether the expression is sufficiently different. Whether or not translating to a different language changes the expression sufficiently has not been tested in court.
A literary work has much stronger protection. Not only the literal content of the work is protected, but also the narrative and the characters.
Thus you could read The Fellowship of the Ring, and learn from it, and write about any ideas you gain from it, but you can't copy the narrative or the characters, which dramatically limits what you could do. As a more relevant example, consider a mathematics textbook. You can read the textbook and learn calculus for instance, and reuse that elsewhere, or even write your own textbook, so long as you don't copy the narrative or directly lift the contents.
Software is special, in that it's relatively untested in court exactly what protection it has, in that the original work is the source code, and it is very unclear how large the changes has to be before a work is no longer considered derivative. The source of this confusion is the fact that software usually is meant to express function, while a literary work for instance isn't - this makes many consider software as akin to a machine. You can't copyright the way a machine works (but you can patent it, which is incidentally part of the reason why we have the whole software patents debacle)
No, it sends the message "release a driver for ANY operating system" and we'll reverse engineer it and buy your hardware...
It's not like you'd need a Linux driver for it to be relatively straightforward to reverse engineer - most hardware drivers are relatively simple, they act as relatively thin layers to abstract out low level hardware access - and reverse engineering them isn't a big deal.
The most notable exceptions are winmodem drivers, where the drivers provide more or less a full modem protocol stack, and 3d graphics cards drivers that tend to provide quite high level interfaces to relatively low level hardware primitives.
If any hardware vendor for more simple hardware devices believe that anyone will have a problem reverse engineering their drivers, they are clueless idiots and will only benefit from learning a lesson.
You're getting it the wrong way. Canopy has been investing in Linux companies for many years before Caldera even bought SCO and morphed into a mini Evil Empire. They more or less funded Caldera from the start and throughout their Linux friendly era. They have done a lot for Linux that way.
I'd say it's to early to say whether the whole SCO thing is a sign they are changing, "just business" that they let SCO do as they wanted with, or something they were against but unable to stop.
But your paranoia is misplaced - it's not as if they've just suddenly started investing in Linux.
To expand on this: If Brazilians spend that amount on licenses for a US product, they can only spend that amount once, and it will be in the pockets of Microsoft, which is unlikely to spend much, if any, of it in Brazil.
If they spend it locally, on hiring developers to improve open source products or on entirely other things (like education etc.), they can spend it multiple times: Part of it will be taxed as it is spent (VAT/sales tax, income tax on salaries etc.), and the government will spend the money again - likely mostly in Brazil. Parts of it will be paid as salaries to people who will spend the money mostly locally. And so on.
Any money kept locally as opposed to paid out of the country will likely circulate in Brazil several times, and may help build value (some of it may end up being invested in companies that generate a profit, or may be used to purchase goods from companies that as a result can afford to expand their business).
So the value of being able to keep the money locally is a lot higher than just the monetary value of the cash that stays in the country.
We win regardless. Either Microsoft will keep cutting prices (and each deal it makes will make other customers start pressuring them), with the resulting reduced margins and hopefully dropping profitability, or people will see through them and they'll lose customers. Both ways leads to a weakened Microsoft. What Microsoft doesn't seem to understand is that cutting prices to the bone only really works when you are fighting against a small group of smaller companies that you can bleed dry by consistently undercutting.
You seem to assume (wrongly), that most software engineers work on shrink wrapped software for companies that live off selling mass produced software, when in fact most software engineers work on in house applications and one off deliveries. Microsoft is an anomaly. Most software engineers already work in positions where open source would not affect their jobs.
However, the shrinkwrap applications that are in common use account for a disproportionate amount of software spending, and by encouraging the use of open source one would free up huge amounts of money otherwise spent on license fees that could be spent on hiring people to adapt various software packages to your specific conditions.
For Brazil that would be hugely beneficial, as most money for shrinkwrap software end up in the US, while software engineers hired to add features to open source software would be more likely to be local.
But even though the US stand to lose short term, it too stand to lose longer term as open source over time reduce the cost to develop software (because of the increasing amount of software that can be used as building blocks free of charge). Looking at the history of software development, this is unlikely to reduce the number of available jobs for software engineers overall - in fact it is likely to make more jobs available, as cheaper software development means significantly more projects become cost effective.
It's like car manufacturing - Ford massively automated the process, but that didn't put people making cars out of work, it massively grew the industry since the lower cost lead to lower prices, which led to a huge increase in demand.
For what purpose? If you want it that distributed, run your own Jabber server locally on your own machine. But how exactly would that make a difference?
Add to that that each step "up the ladder" for these nations will erode their cost advantage. Taiwan used to be cheap. Now they are quickly losing low end manufacturing jobs to Korea and China because their boom has driven salaries up. At some point there will be noone left to lose jobs to, and things will start to stabilize.
How is calling it Red Flag nationalistic? The red flag has a history as a symbol of socialism and revolution and the labor movement worldwide ever since the French revolution, and is still used worldwide. You may not like the symbolism or it's use, but how one could label it nationalistic is beyond me.
What makes you think that a regular user will download individual packages instead of just popping a Redhat, Suse, Mandrake or similar CD in their machine, let the installer take ower, and be done with it?
Mainstream users will no more be building their own packages than Windows users will format their drives and copy in individual files manually from their Windows install CD.
This wasn't a review or tutorial for end users, as should be obvious from the outset of the article.
You don't need rabbit-human hybrids. However, medical researchers do need stem cells to pursue a huge number of research goals, including growing new organs, repairing damaged tissue, regrowing nerves etc. That research is being hampered by concerns from a lot of groups about harvesting stemcells from aborted embryos. Hence this research to create an alternative source of stem cells.
In other words, this IS important research, that has the potential of providing important material for research projects that might take medical science huge leaps forwards.
Slashdot Mirror
Or in this case you needed violence to ensure thirteen years of oppressive military dictatorship and frequent murder of members of the opposition, and assorted other evils.
While I don't have much sympathy for Castro - while he got rid of an oppressive dictator, he doesn't exactly seem to have kept to his ideals - Allende on the other hand was democratically elected, and still had to endure massive CIA interference, and was murdered thanks to the CIA. How that qualifies to being compared to Castro is beyond me.
And so much for the US government caring about democracy. Which makes the whole anti Castro hysteria from the US government seem rather hypocritical, but then hypocrisy seems to be a requirement to get elected in the US.
But still "revolutionary" in terms of the policies the government supported: one of very few regimes that branded itself socialist that actually stood by fundamental Marxist principles instead of just using socialist symbolism as a justification for oppression. In that sense the Allende government was certainly revolutionary.
Why should they be barred from doing the same if an OS doesn't meet their needs?
And why should it be any different if their "need" is to stimulate the local economy?
I'd say a government that spends it's money on paying for local development that become freely available to anyone (including people outside their own country) is doing a hell of a lot better job than a government that spend the same amounts on proprietary development projects that will only benefit one department, and on licensing proprietary applications from a company outside their own country.
Paying for open source development is likely to do a lot more for their people.
Secondly, Microsoft is hoarding cash. If Microsoft started losing money so fast that it would even manage to eat up the return on their capital investments, it would still take a long time before they'd collapse, so barring a sudden devastating move, it will take a VERY long time for Microsoft to end up anywhere near a bankrupcy court, meaning the economy would likely have adapted long ago.
The only way for Microsoft failing to have a major impact would be if they managed to conceal problems until they were near collapsing (lying to the SEC?), or if they were hit with a ridiculously devastating judgement in a class action lawsuit (as in 100-200 billion in damages, if lower they could likely finance their way out of it), which is highly unlikely.
More likely than Microsoft failing is Microsoft morphing. Due to it's financial strength, they keep on buying and investing outside their core areas. If Windows and Office, their two main cash cows, start collapsing on them, they'd likely have plenty of time to shift focus, and given their enormeous amount of cash and cash equivalents they'd likely have no problem surviving until they could live on other products, though probably not with anywhere near the same market cap.
It also blatantly ignores that for a cluster, YOUR APPLICATION need to explicitly manage availability and reliability, or a single node can potentially take down everything if it fails, whereas a mainframe has extensive availability features built in. If you have a 16 node cluster, with 99.9% availability for node, your cluster as a whole would have an availability of approximately 98.4%. You'd be hard pressed to find a mainframe with availability that low.
A cluster is a valid solution where you are willing to deal with that issue, but for many types of applications availability like that simply isn't an option.
To top that off, you would have multiple machines, each with pretty high error rates (compared to a mainframe), which would increase the error rate rather than lower it - unless you do a LOT of work, any one of the machines failing could potentially take down any Linux instance partly depending on that node.
Overcoming those issues is hard to do at OS level without tight hardware integration, which is why it's expensive and still only found in mainframes.
If it's "trickier to admin" then you've defeated the main advantage of these mainframes. Anyone can build a system with the IO throughput or computing power of a mainframe. The trick is to do it in a what thats cheap to admin.
If your work is batch oriented, or individual transactions/jobs aren't critical, then a mainframe is overkill, because a lot of your admin headaches go away simply by ignoring them - a failed node is simply replaced or restarted, possibly semi-automtically, failed hardware is just replaced etc.
The moment you need to start working around this to ensure 100% uninterrupted operation, a cluster starts becoming rapidly more expensive to admin.
The Z-series supports taking CPUs out of comission for replacement without downtime. Same for RAM. Multiple hot-swappable SCSI controllers connected to a fully redundant storage system such as the ESS/Shark (where you can connect to two separate banks of controllers, so that any one of them can be offline without causing problems, and which has two separate AIX servers handling requests, supports RAID and synchroneous mirroring over fiber to a backup ESS), multiple hot swappable network cards, multiple power supplies, and you start getting pretty safe.
Yes, it will cost money, but so will providing all of the above for standalone servers. The Z-series is marketed primarily as a way of reducing maintenance work by consolidating your "servers" on one or two physical platforms, not for it's purchase price - it's an expensive beast.
As for rough water, you won't find much of that either on the Thames around London
People would be mostly competing with a few barges and small tourist boats.
I would call Lord of the Rings a fantasy work, not sci fi, but the underlying theme of mans fear of the machines and what changes they brought is a typical sci fi theme.
I haven't read this book, but sci fi is full of great books that resulted from the setting of "what if these two people or these two groups of people got a chance to meet under other circumstances".
The key result of capitalism is competition. The only measurement of the success of a company under capitalism is profit. Driving up profit means increasing sales, which can only be done as long as consumption increases, or the total market increases (population boom, or expanding into areas you don't currently reach).
The moment these factors are all constrained (population doesn't increase, companies reach all possible consumers, and consumers are consuming as much as they can), the ONLY remaining way to increase profit becomes fighting over market share, or reducing cost. Fighting over market share also increasingly IS an issue of reducing cost, and hence prices, as there is only so much you can do with marketing and product differentiation if someone is dramatically undercutting you.
Cutting cost inevitably boils down to reducing the amount paid to other people, because all resources and materials you pay for ultimately involve paying people, whether it is wages, licenses, purchase of property or any other transaction (even when you pay a corporation, you are then indirectly enriching the owners of the corporation, if a foundation or trust the beneficiaries, if a government, the people)
The logical conclusion is a strong push to cut workforces and/or cut pay. Often the second is a result of the former: People in areas where work is short, or with skills that are becoming obsolete will lower their salary expectations.
However, at some point you reach a level where any reduction in cost lead to a reduction in consumption, at which point reduction in cost for one company will be increasingly hard to compensate by growth elsewhere.
Marx' thesis was that at this point, capitalism will continue to produce, and continue to cut costs, and drive consumptions among the people with capital to extreme excesses by promoting waste that people wouldn't normally consider, while more and more people are pushed into poverty by cost cutting measures.
Capitalists on the other hand, dismiss this, usually by assuming that overall consumption can continue to grow forever, hence always allowing for cost cutting to be compensated by growth in other markets.
Taken to extreme, a society where "workers" aren't needed, capitalism is unlikely to survive. How do you maintain a system based on private ownership of the means of production when it leads to immense poverty, and that poverty isn't "needed" because of scarcity?
It is hard to see a situation like that not eventually leading to growing popular unrest.
Incidentally, in "The German Ideology" Marx wrote [paraphrased] "if the revolution happens in a country with insufficient resources to meet the basic needs, the same shit will start all over again" - Marx always made it very clear that for a socialist revolution to have a chance to succeed, it must happen in a highly evolved capitalist economy, a country where a small elite have accumulated sufficient wealth that the needs of the population as a whole could be met by redistribution, and where the wast majority had been forced into poverty by the more and more extreme competition of capitalist economy.
He specifically named the UK, France and Germany originally, but in a later preface to the Communist Manifesto, he pointed to the US with it's rapid growth and expanding markets as more likely to mature to the sufficient level first....
Interestingly, he also specifically made it clear that he believed that a socialist revolution in Russia would be doomed to failure because of it's low level of development (it was a feudal dictatorship with a mostly agrarian economy).
A literary work has much stronger protection. Not only the literal content of the work is protected, but also the narrative and the characters.
Thus you could read The Fellowship of the Ring, and learn from it, and write about any ideas you gain from it, but you can't copy the narrative or the characters, which dramatically limits what you could do. As a more relevant example, consider a mathematics textbook. You can read the textbook and learn calculus for instance, and reuse that elsewhere, or even write your own textbook, so long as you don't copy the narrative or directly lift the contents.
Software is special, in that it's relatively untested in court exactly what protection it has, in that the original work is the source code, and it is very unclear how large the changes has to be before a work is no longer considered derivative. The source of this confusion is the fact that software usually is meant to express function, while a literary work for instance isn't - this makes many consider software as akin to a machine. You can't copyright the way a machine works (but you can patent it, which is incidentally part of the reason why we have the whole software patents debacle)
It's not like you'd need a Linux driver for it to be relatively straightforward to reverse engineer - most hardware drivers are relatively simple, they act as relatively thin layers to abstract out low level hardware access - and reverse engineering them isn't a big deal.
The most notable exceptions are winmodem drivers, where the drivers provide more or less a full modem protocol stack, and 3d graphics cards drivers that tend to provide quite high level interfaces to relatively low level hardware primitives.
If any hardware vendor for more simple hardware devices believe that anyone will have a problem reverse engineering their drivers, they are clueless idiots and will only benefit from learning a lesson.
I'd say it's to early to say whether the whole SCO thing is a sign they are changing, "just business" that they let SCO do as they wanted with, or something they were against but unable to stop.
But your paranoia is misplaced - it's not as if they've just suddenly started investing in Linux.
If they spend it locally, on hiring developers to improve open source products or on entirely other things (like education etc.), they can spend it multiple times: Part of it will be taxed as it is spent (VAT/sales tax, income tax on salaries etc.), and the government will spend the money again - likely mostly in Brazil. Parts of it will be paid as salaries to people who will spend the money mostly locally. And so on.
Any money kept locally as opposed to paid out of the country will likely circulate in Brazil several times, and may help build value (some of it may end up being invested in companies that generate a profit, or may be used to purchase goods from companies that as a result can afford to expand their business).
So the value of being able to keep the money locally is a lot higher than just the monetary value of the cash that stays in the country.
We win regardless. Either Microsoft will keep cutting prices (and each deal it makes will make other customers start pressuring them), with the resulting reduced margins and hopefully dropping profitability, or people will see through them and they'll lose customers. Both ways leads to a weakened Microsoft. What Microsoft doesn't seem to understand is that cutting prices to the bone only really works when you are fighting against a small group of smaller companies that you can bleed dry by consistently undercutting.
However, the shrinkwrap applications that are in common use account for a disproportionate amount of software spending, and by encouraging the use of open source one would free up huge amounts of money otherwise spent on license fees that could be spent on hiring people to adapt various software packages to your specific conditions.
For Brazil that would be hugely beneficial, as most money for shrinkwrap software end up in the US, while software engineers hired to add features to open source software would be more likely to be local.
But even though the US stand to lose short term, it too stand to lose longer term as open source over time reduce the cost to develop software (because of the increasing amount of software that can be used as building blocks free of charge). Looking at the history of software development, this is unlikely to reduce the number of available jobs for software engineers overall - in fact it is likely to make more jobs available, as cheaper software development means significantly more projects become cost effective.
It's like car manufacturing - Ford massively automated the process, but that didn't put people making cars out of work, it massively grew the industry since the lower cost lead to lower prices, which led to a huge increase in demand.
For what purpose? If you want it that distributed, run your own Jabber server locally on your own machine. But how exactly would that make a difference?
Add to that that each step "up the ladder" for these nations will erode their cost advantage. Taiwan used to be cheap. Now they are quickly losing low end manufacturing jobs to Korea and China because their boom has driven salaries up. At some point there will be noone left to lose jobs to, and things will start to stabilize.
WP != WPS
How is calling it Red Flag nationalistic? The red flag has a history as a symbol of socialism and revolution and the labor movement worldwide ever since the French revolution, and is still used worldwide. You may not like the symbolism or it's use, but how one could label it nationalistic is beyond me.
You are making the assumption that all the software they distribute as part of their Enterprise offering is GPL'd, which is wrong.
Mainstream users will no more be building their own packages than Windows users will format their drives and copy in individual files manually from their Windows install CD.
This wasn't a review or tutorial for end users, as should be obvious from the outset of the article.
In other words, this IS important research, that has the potential of providing important material for research projects that might take medical science huge leaps forwards.