I too want to do cool stuff, but the reality is that there is cool stuff and stuff that will make the company money.
The trick, really, is doing the research to find places to work where the distance between what you perceive as "cool stuff" (or, less informally, desirable features of the job) and what will serve your employers interest (which isn't always "make the company money"; not every job is in a private, for-profit firm) is minimized.
And, of course, in putting in the effort to stand out so that where the stuff you want out of the job is something lots of other people want and only few will get, you'll be one of the few who gets it.
For years, FOSS advocates have talked about how freely redistributable open source software is the better model, and how it's still possible to make money off open source software. Well, if open source is the better, more powerful model for developing software, the certification fee shouldn't be a problem.
Certification requirements are a government-imposed market distortion that, if imposed in a way which attaches the cost to the developer systematically disadvantages FOSS software. Of course, since certification requirements, however loudly they are trumpeted as serving a public interests, are almost invariably crafted with the cooperation of the major commercial vendors in the field with the primary intent of reinforcing their position against any possible upstart competitors, this isn't exactly an unintended feature.
If the law states that there should be a 'view but not save/copy/print' right (like here in the Netherlands), how could you enforce that *and* be truly open source?
The same way you would do that for commercial programs.
Being open source doesn't mean that there is an absence of government regulations that restrict your ability to distribute and/or use modified versions.
You have to certificate each and every release of the full software on a source code level (and provide authorization based on the (i.e.) md5 sum of the executable) to enforce such rights. One simple edit & recompile and you can save/print those x-ray pics, which is against the law.
And...so, what? Its always possible for the user to modify either the software, the software environment in which the application software runs, or the hardware platform on which the software runs to avoid such restrictions. Certification of software only provides assurance for the software in the form it is sold, not anything that is done by the purchaser after they have received it. Other enforcement measures, like on-site audits, are necessary, whether or not the software is open source, to assure that the user uses the software in a manner which complies with the law.
At the very least, forking, maintaining your own version and fixing bugs for your (employer's) own use is either impossible or very expensive.
It doesn't add any cost that isn't added to purchased software, even if the certification requirement is on software used and not software sold for a purpose, since you are going to be paying the cost of certification for any software you purchase, as well. OTOH, since the modification and use of software in house is part of the internal practices, it makes more sense to include those in whatever regulation and certification requirements exist for internal practices, rather than in the kind of certification requirements that are imposed on software sold for a regulated purpose.
It's pretty commonly accepted that certification groups that get their budget from fees have a pretty significant conflict of interest wrt. properly executing their duties.
That's actually the point of such certification groups; they serve their paying clients by creating a competitive advantage for the existing big players that any new competitor has trouble meeting. That's even moreso, often, the point of such groups when certification through them is required by government regulation, as such regulations are shaped often crafted to serve the interests of the existing major players in the industry, and its more effective of a barrier to competition when those who can't afford the certification don't just have more trouble selling their product, but are outright banned from doing so.
Well, it's more that the law is terrible than the ruling
I think the ruling is terrible, whether or not the law is: the principle of the truth as an absolute defense is an application of the First Amendment freedom of speech, and thus should not be trumped by State law in any case.
Success depends on a lot more than just a person's innate "specialness".
Sure, so what? That it depends on more than a person's "specialness" does not refute that "specialness" matters.
Or I can be more blunt: They just happened to be in the right place, at the right time, and had what was needed.
Yes, and to be equally blunt, hundreds of millions or billions of other people were around at the right time, very many of them at the same right place or one equally right; what was key is the "had what was needed" parted.
Sure, without any one of a small number of people, nuclear weapons would still have been developed, just later and perhaps by a different country. The effect on world history would, potentially, have been pretty significant.
More to the point at hand, had the preferences of gopher-preferring administrators prevailed in the early 1990s, we still eventually would probably have multimedia over popular network systems -- but with the internet already going through a slow growth before the explosion due to the web, it might have been after big firms got farther ahead of the ball, and missed much of the disruptive impact that the explosion of the internet that occurred because of the utility of the web had.
The power of the 'web isn't in multi-media delivery. That's not to say it isn't important. But there is a more fundimental feature; ubiquity.
Ubiquity, a feature of the internet, was a consequence of multimedia, a feature of the web -- almost anyone could get access to the internet for many years before the web was popular (I remember first looking into local ISP options in ~1991.) Comaparatively few people did until the web was popular because there was no appeal to most people. The internet, which had been around for quite sometime, became omnipresent because it offered something which rapidly drew wide interest, and that was the multimedia offered by the the web.
Even if Gopher had dominated due to filtering (a premise I don't agree with), multimedia capabilities would have eventually been added to the protocol out of demand. We'd have the same web we have today.
Eventually, maybe, but exposure drives demand; if it had stalled long-enough for, say, cable and phone companies to deliver substantial non-free interactive multimedia outside of the context of the web first, its very likely that nothing socially like the current web would have existed any time near now, even if many of the individual features that are important about the web were available in one form or another on some networked electonic system that was widely available elsewhere.
Funny thing is, if one uses the styles in Word correctly, you get a WYSIWYM editor, just never, EVER touch the bold, italic, underscore button. And the sad thing is it's much, much easier to do this in word 2000 then in newer versions.
Personally, I find it much easier to do it Word 2003 and later than in older versions, since they have fairly well designed "click and go" style application with visual preview, which puts the structured way more on par with the seductive, but ultimately evil, appearance-oriented unstructured way.
But as long as its not much easier to do it the structured way, and to see that it was done that way, almost everyone is going to do it the unstructured way with Word, and the few people who want it right are going to be fighting a losing battle. Which, really, means that as long as its WYSIWYG-focussed, structure won't be the preferred mechanism.
Much of the complexity of LaTeX can be abstracted away in lighter text-based formats that are compiled to LaTeX to produce print-destined output (ReStructured Text, used in python Docutils, is one example.) If you are concerned about the combined complexity of LaTeX + version control, that could help reduce the overall complexity.
There was a grassroots effort in the 80s to pass what was called the Equal Opportunity Amendment. It was approved by somewhat more than 20 states before being killed by the National Organization of Women, who were outraged that the special rights of women would be stripped away in favor of the equal rights of all.
Some support is needed for each of the claims made here.
If our government is making treaties without the consent of the governed, then we should convene congress in our respective states and vote to remove from the constitution the power of the Federal Congress to make treaties without the consent and approval of the state legislatures.
the "Federal Congress" doesn't have the power to make treaties in any case, with or without the consent or approval of state legislatures.
In the article, they put it correctly: "Only 15% of respondents answered this question with the exactly correct answer of 70%."
Of course, the scientific illiteracy extends to anyone who portrays that as the "exactly correct answer", since it is, of course, an approximation (accurate to the nearest 2%; to two significant figures, the figure is 71%.)
Ah, another Google service that might one day disappear because Google don't think it's valuable enough.
That's true of every service, from every company, whether its in Beta or not. The only "guarantee" you have is with a contract, and only for the life of the contract, and even then its just them weighing the cost of buying off your contract or suffering an action for breach if you refuse to be bought off against the cost of maintaining the service that is not valuable to them.
The submitter already requires a user account to use his system. Hence the part in the story about spending a lot of time deleting accounts.
Sure, but there wouldn't be account deletion on the grounds of non-locality if only people who met whatever the criteria for "locality" are were able to create accounts in the first place. Personally, I think a big problem with automating this is that there doesn't appear to be a concrete set of criteria. Given that, the best idea may be to require an in-person, face-to-face meeting at which it can be verified that the applicant is a genuine "local" prior to creating an account. To scale up, allow certain privileged users with the trust of the site owner to validate accounts rather than requiring the site owner to do all the face-to-face meetings himself.
If the concept of "local" involved is one which requires an ongoing connection, you might also require all users to be revalidated by an administrator periodically (e.g., annually) or have their accounts automatically suspended.
Basically, the government will have to throw out or severely limit the use of most medical software, and enforce its replacement with something standard if they want to make health information electronically available to any provider.
Which is what the government has done (more than once) with existing billing software, by specifying (and then updating) standards for electronic health care claims and related transactions under the HIPAA Transactions and Code Sets rule. (These standards aren't open, but they were for quite some time free-as-in-beer for interested parties to get electronic copies of and use, until the government stopped subsidizing them.)
It makes more sense to do that with medical records, since the benefits of interchangeability are much higher than with billing-related transactions, but it definitely needs to have a suitable lead time and support for transition costs, because you don't want to drive providers out of business with the mandate.
The problem with the economic crisis today lies with the financial and banking system.
No, the problem lies with the lack of availability of credit and the lack of consumer demand. The primary direct cause of that may have been actions in and affecting the financial services industry including the banks, but that doesn't mean that the most effective way of dealing with it is exclusively with policies directed at that industry, in the same way that bad diet and inadequate exercise may be the principal cause of a heart attack, but the best response to a heart attack may not be limited to diet and exercise changes.
Its funny how liberals were complaining that invading Iraq had nothing to do the GWOT.
Liberals, in fact, were not generally complaining about that. Liberals were complaining that Iraq (not "invading Iraq") had nothing to do with 9/11 (not "the war on terror") and that invading Iraq was directly counterproductive in (not "had nothing to do with") the war against the people who had actually attacked the United States on 9/11, and that contributed to producing more people who would be more easily recruitable by groups wanting to attack the United States through terrorism.
The first half relates to the justification, the second to utility. Confusing different parts of two distinct-though-related criticisms of the invasion of Iraq misses the point of both criticisms rather completely.
This is the liberals version of 9/11, using the crisis as a pretext to remake the US economy and set their agenda.
That doesn't make sense. The economy is broken. Liberals are proposing a particular way of fixing it that, they argue, apply both to the immediate problem and the longer-term structural problems that make problems like the immediate one both more likely to occur and more damaging to individual citizens when they occur. As you note, what they are doing is directed at the economy, which is where you admit the problem is, not at some unrelated thing. Now, you might argue that the proposals are not directed well to fix the problems in the economy, which would be a legitimate point to debate, but you fail to make that argument, instead making an argument by analogy (though, as noted, a poorly-crafted analogy that reveals poor understanding both of the immediate situation and the one to which an analogy is drawn) that seems to rely on the idea that it is not directed at principal immediate cause of the problem, rather than arguing that it is ineffective at solving the problem. But being effective at addressing a set of undesirable conditions is logically orthogonal to being directed at the events and conditions which contributed to the development of those conditions.
Basic microeconomics says that production cost doesn't dictate price: what dictates price is what consumers are willing to pay.
Actually, basic microeconomics says that supply and demand interact to dictate price, "what consumers are willing to pay" alone dictates price only in the case of a monopoly, where the producer is free to raise price as much as is beneficial without pressure from competition. In a situation of perfect competition, the market price of any product will be equal to the marginal cost of production, since as long as the marginal utility (and hence the price the next consumer is willing to pay) is greater than that, a manufacturer will be able to make additional profit by increasing production.
Now, for many online products, the marginal cost to the producer of delivering a unit of product is very close to zero, meaning that, with competition, the price of products will be brought down very near to zero. And given that entertainment products are often very close to perfect substitutes for each other (in the economic sense), so even a new game can have problems there. And if you give much of your game away for free, you may be helping spread knowledge about your game, which helps you find the people willing to pay for it, but you are simultaneously reducing the utility to those people of paying for the game, since they've already got much of what it provides for free: the cost then has to be justified by the added benefit, not the whole quality of the game, since the cost only pays for the "premium" portion from the customer's point of view.
While they use Scheme instead of Python, which you seem to have settled on, the TeachScheme! effort and the book How to Design Programs are aimed pretty much at the kind of thing you are trying to do: HtDP is designed for universal programming education (aimed at either high school students or lower division liberal arts students in college), and TeachScheme! is directly aimed at teaching in High School.
(Linked Wikipedia pages because it seems like both websites -- and the main PLT website -- are down at the moment; the website links are on the wiki pages and I'd recommend going to those when they are up.)
not only that but what exactly do you think your handheld/phone is going to do with more than 4 gig of ram per process?
"640K should be enough for anybody", updated for 2009.
Plus, TFS specifically mentions netbooks as part of where the chips would be used, which are more likely to make that particular feature likely than handhelds/phones, as the use cases for those are pretty much those for a general purpose laptop -- but with more focus on battery life (hence, power consumption), weight, and cost. I suspect the same issues that Nvidia sees that would make it make sense to "at some point" go from ARM to x86 for integrated chips would also at some point make it make sense to go to x86-64 as well, though it might be at a later point than going to x86.
No, I do not want government software to be open source.
I want government software, like all software, to be good.
Let me clarify: I do not mean to say that it is essential that all government software should be open source, what I meant is that there open source is, all other things being equal, a positive thing in many areas of government, where the same advantage may not apply equally in the private sector (this is particularly true, for instance, in software where government sponsors the development, remembering that the government exists to serve the public interest, not a competitive private interest, so that whereas for a private company it is a benefit to gain a competitive advantage over other users with similar needs -- and moreso the more similar the needs are -- with government it is not.)
IOW, very often, open source is good for government in ways that it might not be as good for a private business, simply because the purpose of government is not the same as the purpose of a private business.
We have various agencies offering tax breaks and increased spending to help the economy. The reality is that spending Trillions more than you bring in may help short-term. However, long term we have to work harder just to pay the debt we owe everyone.
If, as you say, the tax breaks and increased spending actually do help, then, by definition, more wealth has been created than otherwise would have existed with which to pay off the debt. That's the whole theory behind accepting short-term and potentially very significant increases in debt load (proportional to GDP) during a recession (not willy-nilly, but with targeted policies designed to reduce the impact of and reverse the downturn) while working out of the debt load during expansions. The problems generally are (1) proper targeting of stimulus policy during recessions (which is a comparatively minor problem in practice; while it can be done very badly, usually it is done in a way which is a net benefit even if it could be improved), and, particularly recently, (2) maintaining the discipline to work debt loads down during expansions.
#2 is the big problem, and its not a problem that has always existed and that shows you just can't trust politicians; US national debt-to-GDP ratio was managed well from the end of WWII through the 1970s, working down pretty consistently, with occasional short-term increases; exactly what you'd expect from responsible fiscal policy. In the 1980s under Reagan and Bush I it exploded -- not only during the recessionary periods, but during the expansionary periods in between, as well. Under Clinton the growth of the debt compared to the GDP was slowed and reversed, buy it grew again under Bush II, not just during the short recession at the beginning of his term and the longer one at the end, but through the expansion in the middle as well (to be fair to Bush the Younger, the debt increase in his term was at nothing like the rate under Reagan and Bush the Elder.)
(OTOH, its worth noting that, despite all the hyperventilating about the size of the national debt now, its currently a little over half the share of GDP it was when it peaked in the 1940s; while we certainly need to get better discipline about working it down during expansions, the apocalyptic portrayals of unprecedent runaway debt are misleading.)
If you want to help open source, require that government software makes widespread use of open specifications.
Open specifications and open source are almost orthogonal; I would say you want government software to specifically use open source, and government-issued software standards (e.g., standard mandated data formats) to be open specifications.
Force the projects to be GPL-licensed and only GPL licensed. That way it prevents companies opening an BSD-style licensed project to get tax cuts on the early development and then finishing it closed source.
Companies don't get tax cuts, the contributors do. The companies can already deduct their expenditures as business expenses. And, if that was an issue, your proposal wouldn't help, since it wouldn't stop them from starting a GPL-licensed-with-copyright-assignment-required-to-contribute and then finishing it closed-source.
Sure, all the original contributed code would have at one point been available under the GPL, just as in the other case the original contributed could would have at one point been available under the BSD-like license, but the license of the final product is controlled by whoever controls the copyright, however they may have licensed preliminary versions.
Slashdot Mirror
The trick, really, is doing the research to find places to work where the distance between what you perceive as "cool stuff" (or, less informally, desirable features of the job) and what will serve your employers interest (which isn't always "make the company money"; not every job is in a private, for-profit firm) is minimized.
And, of course, in putting in the effort to stand out so that where the stuff you want out of the job is something lots of other people want and only few will get, you'll be one of the few who gets it.
Certification requirements are a government-imposed market distortion that, if imposed in a way which attaches the cost to the developer systematically disadvantages FOSS software. Of course, since certification requirements, however loudly they are trumpeted as serving a public interests, are almost invariably crafted with the cooperation of the major commercial vendors in the field with the primary intent of reinforcing their position against any possible upstart competitors, this isn't exactly an unintended feature.
The same way you would do that for commercial programs.
Being open source doesn't mean that there is an absence of government regulations that restrict your ability to distribute and/or use modified versions.
And...so, what? Its always possible for the user to modify either the software, the software environment in which the application software runs, or the hardware platform on which the software runs to avoid such restrictions. Certification of software only provides assurance for the software in the form it is sold, not anything that is done by the purchaser after they have received it. Other enforcement measures, like on-site audits, are necessary, whether or not the software is open source, to assure that the user uses the software in a manner which complies with the law.
It doesn't add any cost that isn't added to purchased software, even if the certification requirement is on software used and not software sold for a purpose, since you are going to be paying the cost of certification for any software you purchase, as well. OTOH, since the modification and use of software in house is part of the internal practices, it makes more sense to include those in whatever regulation and certification requirements exist for internal practices, rather than in the kind of certification requirements that are imposed on software sold for a regulated purpose.
That's actually the point of such certification groups; they serve their paying clients by creating a competitive advantage for the existing big players that any new competitor has trouble meeting. That's even moreso, often, the point of such groups when certification through them is required by government regulation, as such regulations are shaped often crafted to serve the interests of the existing major players in the industry, and its more effective of a barrier to competition when those who can't afford the certification don't just have more trouble selling their product, but are outright banned from doing so.
I think the ruling is terrible, whether or not the law is: the principle of the truth as an absolute defense is an application of the First Amendment freedom of speech, and thus should not be trumped by State law in any case.
Sure, so what? That it depends on more than a person's "specialness" does not refute that "specialness" matters.
Yes, and to be equally blunt, hundreds of millions or billions of other people were around at the right time, very many of them at the same right place or one equally right; what was key is the "had what was needed" parted.
Sure, without any one of a small number of people, nuclear weapons would still have been developed, just later and perhaps by a different country. The effect on world history would, potentially, have been pretty significant.
More to the point at hand, had the preferences of gopher-preferring administrators prevailed in the early 1990s, we still eventually would probably have multimedia over popular network systems -- but with the internet already going through a slow growth before the explosion due to the web, it might have been after big firms got farther ahead of the ball, and missed much of the disruptive impact that the explosion of the internet that occurred because of the utility of the web had.
Ubiquity, a feature of the internet, was a consequence of multimedia, a feature of the web -- almost anyone could get access to the internet for many years before the web was popular (I remember first looking into local ISP options in ~1991.) Comaparatively few people did until the web was popular because there was no appeal to most people. The internet, which had been around for quite sometime, became omnipresent because it offered something which rapidly drew wide interest, and that was the multimedia offered by the the web.
Eventually, maybe, but exposure drives demand; if it had stalled long-enough for, say, cable and phone companies to deliver substantial non-free interactive multimedia outside of the context of the web first, its very likely that nothing socially like the current web would have existed any time near now, even if many of the individual features that are important about the web were available in one form or another on some networked electonic system that was widely available elsewhere.
Personally, I find it much easier to do it Word 2003 and later than in older versions, since they have fairly well designed "click and go" style application with visual preview, which puts the structured way more on par with the seductive, but ultimately evil, appearance-oriented unstructured way.
But as long as its not much easier to do it the structured way, and to see that it was done that way, almost everyone is going to do it the unstructured way with Word, and the few people who want it right are going to be fighting a losing battle. Which, really, means that as long as its WYSIWYG-focussed, structure won't be the preferred mechanism.
Much of the complexity of LaTeX can be abstracted away in lighter text-based formats that are compiled to LaTeX to produce print-destined output (ReStructured Text, used in python Docutils, is one example.) If you are concerned about the combined complexity of LaTeX + version control, that could help reduce the overall complexity.
Some support is needed for each of the claims made here.
the "Federal Congress" doesn't have the power to make treaties in any case, with or without the consent or approval of state legislatures.
Of course, the scientific illiteracy extends to anyone who portrays that as the "exactly correct answer", since it is, of course, an approximation (accurate to the nearest 2%; to two significant figures, the figure is 71%.)
That's true of every service, from every company, whether its in Beta or not. The only "guarantee" you have is with a contract, and only for the life of the contract, and even then its just them weighing the cost of buying off your contract or suffering an action for breach if you refuse to be bought off against the cost of maintaining the service that is not valuable to them.
Sure, but there wouldn't be account deletion on the grounds of non-locality if only people who met whatever the criteria for "locality" are were able to create accounts in the first place. Personally, I think a big problem with automating this is that there doesn't appear to be a concrete set of criteria. Given that, the best idea may be to require an in-person, face-to-face meeting at which it can be verified that the applicant is a genuine "local" prior to creating an account. To scale up, allow certain privileged users with the trust of the site owner to validate accounts rather than requiring the site owner to do all the face-to-face meetings himself.
If the concept of "local" involved is one which requires an ongoing connection, you might also require all users to be revalidated by an administrator periodically (e.g., annually) or have their accounts automatically suspended.
Which is what the government has done (more than once) with existing billing software, by specifying (and then updating) standards for electronic health care claims and related transactions under the HIPAA Transactions and Code Sets rule. (These standards aren't open, but they were for quite some time free-as-in-beer for interested parties to get electronic copies of and use, until the government stopped subsidizing them.)
It makes more sense to do that with medical records, since the benefits of interchangeability are much higher than with billing-related transactions, but it definitely needs to have a suitable lead time and support for transition costs, because you don't want to drive providers out of business with the mandate.
No, the problem lies with the lack of availability of credit and the lack of consumer demand. The primary direct cause of that may have been actions in and affecting the financial services industry including the banks, but that doesn't mean that the most effective way of dealing with it is exclusively with policies directed at that industry, in the same way that bad diet and inadequate exercise may be the principal cause of a heart attack, but the best response to a heart attack may not be limited to diet and exercise changes.
Liberals, in fact, were not generally complaining about that. Liberals were complaining that Iraq (not "invading Iraq") had nothing to do with 9/11 (not "the war on terror") and that invading Iraq was directly counterproductive in (not "had nothing to do with") the war against the people who had actually attacked the United States on 9/11, and that contributed to producing more people who would be more easily recruitable by groups wanting to attack the United States through terrorism.
The first half relates to the justification, the second to utility. Confusing different parts of two distinct-though-related criticisms of the invasion of Iraq misses the point of both criticisms rather completely.
That doesn't make sense. The economy is broken. Liberals are proposing a particular way of fixing it that, they argue, apply both to the immediate problem and the longer-term structural problems that make problems like the immediate one both more likely to occur and more damaging to individual citizens when they occur. As you note, what they are doing is directed at the economy, which is where you admit the problem is, not at some unrelated thing. Now, you might argue that the proposals are not directed well to fix the problems in the economy, which would be a legitimate point to debate, but you fail to make that argument, instead making an argument by analogy (though, as noted, a poorly-crafted analogy that reveals poor understanding both of the immediate situation and the one to which an analogy is drawn) that seems to rely on the idea that it is not directed at principal immediate cause of the problem, rather than arguing that it is ineffective at solving the problem. But being effective at addressing a set of undesirable conditions is logically orthogonal to being directed at the events and conditions which contributed to the development of those conditions.
Actually, basic microeconomics says that supply and demand interact to dictate price, "what consumers are willing to pay" alone dictates price only in the case of a monopoly, where the producer is free to raise price as much as is beneficial without pressure from competition. In a situation of perfect competition, the market price of any product will be equal to the marginal cost of production, since as long as the marginal utility (and hence the price the next consumer is willing to pay) is greater than that, a manufacturer will be able to make additional profit by increasing production.
Now, for many online products, the marginal cost to the producer of delivering a unit of product is very close to zero, meaning that, with competition, the price of products will be brought down very near to zero. And given that entertainment products are often very close to perfect substitutes for each other (in the economic sense), so even a new game can have problems there. And if you give much of your game away for free, you may be helping spread knowledge about your game, which helps you find the people willing to pay for it, but you are simultaneously reducing the utility to those people of paying for the game, since they've already got much of what it provides for free: the cost then has to be justified by the added benefit, not the whole quality of the game, since the cost only pays for the "premium" portion from the customer's point of view.
While they use Scheme instead of Python, which you seem to have settled on, the TeachScheme! effort and the book How to Design Programs are aimed pretty much at the kind of thing you are trying to do: HtDP is designed for universal programming education (aimed at either high school students or lower division liberal arts students in college), and TeachScheme! is directly aimed at teaching in High School.
(Linked Wikipedia pages because it seems like both websites -- and the main PLT website -- are down at the moment; the website links are on the wiki pages and I'd recommend going to those when they are up.)
"640K should be enough for anybody", updated for 2009.
Plus, TFS specifically mentions netbooks as part of where the chips would be used, which are more likely to make that particular feature likely than handhelds/phones, as the use cases for those are pretty much those for a general purpose laptop -- but with more focus on battery life (hence, power consumption), weight, and cost. I suspect the same issues that Nvidia sees that would make it make sense to "at some point" go from ARM to x86 for integrated chips would also at some point make it make sense to go to x86-64 as well, though it might be at a later point than going to x86.
Let me clarify: I do not mean to say that it is essential that all government software should be open source, what I meant is that there open source is, all other things being equal, a positive thing in many areas of government, where the same advantage may not apply equally in the private sector (this is particularly true, for instance, in software where government sponsors the development, remembering that the government exists to serve the public interest, not a competitive private interest, so that whereas for a private company it is a benefit to gain a competitive advantage over other users with similar needs -- and moreso the more similar the needs are -- with government it is not.)
IOW, very often, open source is good for government in ways that it might not be as good for a private business, simply because the purpose of government is not the same as the purpose of a private business.
Correct. Which is why I said, in the post you responded to here, that the two were "almost orthogonal".
You are repeating yourself, to which I repeat the above response.
If, as you say, the tax breaks and increased spending actually do help, then, by definition, more wealth has been created than otherwise would have existed with which to pay off the debt. That's the whole theory behind accepting short-term and potentially very significant increases in debt load (proportional to GDP) during a recession (not willy-nilly, but with targeted policies designed to reduce the impact of and reverse the downturn) while working out of the debt load during expansions. The problems generally are (1) proper targeting of stimulus policy during recessions (which is a comparatively minor problem in practice; while it can be done very badly, usually it is done in a way which is a net benefit even if it could be improved), and, particularly recently, (2) maintaining the discipline to work debt loads down during expansions.
#2 is the big problem, and its not a problem that has always existed and that shows you just can't trust politicians; US national debt-to-GDP ratio was managed well from the end of WWII through the 1970s, working down pretty consistently, with occasional short-term increases; exactly what you'd expect from responsible fiscal policy. In the 1980s under Reagan and Bush I it exploded -- not only during the recessionary periods, but during the expansionary periods in between, as well. Under Clinton the growth of the debt compared to the GDP was slowed and reversed, buy it grew again under Bush II, not just during the short recession at the beginning of his term and the longer one at the end, but through the expansion in the middle as well (to be fair to Bush the Younger, the debt increase in his term was at nothing like the rate under Reagan and Bush the Elder.)
(OTOH, its worth noting that, despite all the hyperventilating about the size of the national debt now, its currently a little over half the share of GDP it was when it peaked in the 1940s; while we certainly need to get better discipline about working it down during expansions, the apocalyptic portrayals of unprecedent runaway debt are misleading.)
Open specifications and open source are almost orthogonal; I would say you want government software to specifically use open source, and government-issued software standards (e.g., standard mandated data formats) to be open specifications.
Companies don't get tax cuts, the contributors do. The companies can already deduct their expenditures as business expenses. And, if that was an issue, your proposal wouldn't help, since it wouldn't stop them from starting a GPL-licensed-with-copyright-assignment-required-to-contribute and then finishing it closed-source.
Sure, all the original contributed code would have at one point been available under the GPL, just as in the other case the original contributed could would have at one point been available under the BSD-like license, but the license of the final product is controlled by whoever controls the copyright, however they may have licensed preliminary versions.