There is no reason why 99% of applications that rely on the current configure && make && make install
could not be wrapped with a graphical front end.
Well, one reason: regular users probably won't have installed "C/C++ Development" at their RedHat install time, and so don't have gcc, etc. available. Building from source is going to be a needless complication for most users, with really limited utility since they will just be reproducing what their distro's binary packages would provide anyway.
It seems to me that this whole thread is off on the wrong tangent. The equivalent of Windows setup.exe is double-clicking on your distribution's provided RPM file, and having KDE/Gnome be smart enough to run rpm -i for you. "Regular users" should only be installing from RPMs provided by their distribution maintainer and guaranteed to handle dependencies properly. Any time you get users that have to compile something from source or set up something from a foreign distro, you are way off of the parallel of Windows' setup.exe and are talking about something more akin to porting from one version of Windows to another.
Good question. There are arenas that are tough to get a foothold in without some monetary support, and I would be more confident in my argument if I could point to the preponderance of open source projects being new and innovative ideas, rather than free clones of commercial software.
Not all commercial software are proprietary. There's a huge difference between open-source software and "based
on open standards software". While I agree that something totally proprietary is bad, this is a different topic.
You are entirely correct; I read your post too quickly and assumed that you were making a leap that I've seen many others here make. Mea culpa.
However, since you go on to argue that it would be difficult to support a software industry economically based on entirely open source software, it sounds like you are arguing that "commercial" will essentially come out to be "proprietary" anyway.
I think the assumption that you are making is that the survival of a particular software industry is what's important for computing. I don't think that is really the case - otherwise the swallowing of the computing world by Microsoft would be the best outcome possible, rather than the worst, right? What is important is the computing experience which is provided to the users of the software; whether that experience can be better supplied by a commercial software industry or by the free software world is the big question. With a large enough user base, free software can drive itself forward - not built just during leisure time, but also as other users of the software hack on it to get it to work for their business needs. Considering the relative sizes of the current free software and commercial software communities, free software is already growing almost as fast, and is going much faster in some arenas. If free software were all that there were, the improvement would be even greater.
Sorry, but it's only open standards that made the difference. Remember: the first web browsers were commercial, but
it's the open standards that allowed competition. Even open-source software may not be based on open standards, so
they could be a pain to learn, manage and change. Learn to use Linux, then tell me if it's easy to learn. Same thing for
Microsoft Office.
Linux was easy to learn, because I already knew HP-UX, and Linux was based on the same POSIX standards. I can't speak to Microsoft Office, not having tried to do anything complicated in it. And the POSIX standards were open for years, but cheap and reliable POSIX clones did not become readily available until people started handing around the source code to them. Open standards are good; open source to implement them with is better.
I haven't seen one example where making the software open-source made it's development faster. And I know what
I'm talking about (see my site).
That's not exactly what I said. I said "computing has grown faster", not that a particular project has developed faster. Open source means more people contribute to projects, more people fork and produce interesting variants of projects, and user choice improves - thus computing itself, measured from the user perspective, grows faster. Although I will also point out that for large enough open source projects the project itself will be developed faster as development and debugging is done in parallel by many smarter people than one company could hire. Most projects are not Mozilla:) And even Mozilla is growing and blossoming in unpredictable ways now that there is a seasoned cadre of hackers that understand and make use of it.
I didn't say Microsoft was the only exception. And I'll point out that in many of those cases those companies have been hurt by their unethical or illegal actions. Just reeling off names does not disprove my point.
As implied, their goal is that your whole computing experience is based only on "free" software. This is crazy.
Computing can grow only if you have both commercial and "free" software. Remove one of the two and you have
something like technology without science, or science without technology. We have to be realistic.
I think this is begging the question - do we know for sure that an entirely free-software-based computing sector cannot grow and thrive just as well as a partly or entirely proprietary one? It seems to me that computing has grown faster as free software (and even open source software) has become more prominent. The only question is whether we have currently found the best "mix" between open and proprietary, or whether we should continue on to more and more degrees of freedom in order to improve software even more.
If Gnome is to be successcul, they will need
a board of like minded, energetic people to lead them. Do you think there was anyone on Microsquish's board who
said "hey, I think we are doing the wrong thing here?".
If there was any justice in the world (or in the U.S. government) then Microsoft would right now be wishing that they'd had such a conscience. Most of the time, keeping a business focused on ethics (or at least on staying within the law) is a good business practice, not a mistake. It saves you money and time in the long run. Microsoft just happens to have been a giant exception to this so far.
You've made some good points about matching the cost to expected future revenue based on those costs, and I can't dispute them on accounting terms. I guess in the end I feel that the damages in a case should actually impose some sort of punishment. Including Microsoft's standard markup from blank cdroms to boxed software effectively decreases the amount of punishment by about the markup amount, minus the sales which they would lose by the giveaway. Maybe it is only because the markup for software is so much higher than other goods that it bugs me, but I don't feel that the proposed penalty is just.
Here's how you can tell, in fact: give Microsoft the option to instead be fined $1.1 billion in cash, and then the cash will be given to a third party to buy software and hardware for the schools. Even if this third party were willing to buy all Microsoft software, and even if you made this alternate penalty 5% less painful, Microsoft would still go for the original "software giveaway" punishment, because it really is less painful for them to run off a few extra copies of Windows and lose a few sales (to poor schools who probably would have gone without or pirated anyway) than it would be for them to give up cash money. Every business will realistically account current cash money higher than an equivalent amount of yesterday's IP creation, no matter how much they report that R&D to be worth on their tax forms.
From your argument, I admit that price they set for the software is somewhat related to its production cost. But I think that as long as Microsoft is willing to give away the software rather than just pay a similarly-sized cash fine, they themselves are reputing your argument for the true cost of their product. I think I might be more willing to accept their offer if they would provide a real accounting of how much of their market prices are R&D, production, and profit, and would remove the profit percentage from their costs for the basis of this penalty calculation.
And this is all apart from the unquantifiable product lock-in opportunities which the "software giveaway" provides, of course. That in and of itself makes this punishment more of a "get out of jail free" than anything.
It does matter, because it allows the US to respond in ways other than WMD and with less than
overwhelming force to the small scale attacks that are more likely in the future. You are living
pre-1990, not post 2000. (As the quote below shows)
I think that the suggestion that the U.S. would not respond with overwhelming force in the face of any nuclear attack is a mistaken one to give to the world or to ourselves. The whole point of maintaining superiority in terms of this kind of weapon is to make sure that the consequences of a known nuclear attack on the U.S. are certain. The situation has not changed since the cold war: any nuclear attack on the U.S. which can be traced back to a particular country will be met with a massive and devastating response. To respond with any less than that would invite more attacks.
BMD is *not* meant to stop currently extremely unlikely 'massive attack'. It's meant to blunt or stop
the increasingly likely 1-10 missile attacks.
Any missile attack is traceable, and would invite a similar or greater retaliation from the U.S. No country is going to attack the U.S. with nuclear missiles unless they embrace destruction. Missile attacks themselves are pre-1990s, not post-2000 (or post-09-11-2001 I might say).
Sorry, but it does. Even a 50% defense doubles the amount of missiles an attacker must launch to
ensure a given level of damage. This is very simple math. (And once the supporting infrastructure is in
place and debugged, adding interceptors is far cheaper to us than buying ICBM's and warheads are to
any attacker.)
Read more carefully: the amount of damage doesn't really matter. A nuclear attack via missile on the U.S. will result in the nuclear destruction of the attacker. We are already using MAD to effectively handle this sort of threat. This threat will never go away, since there are already nations existing which could totally destroy the U.S. if they wanted to.
And heck, for any up-front expenditure X, it is possible to say that it's "cheaper" to just pay an additional %Y for so much more protection. The question is whether the upfront expenditure is worth it, or whether there are better ways to achieve the same goals.
Clue:We are doing so, but that does not defend us against the numerous nations already posessing such
a weapon or it's base technologies.
MAD protects us against nations that launch missiles. I agree that we are not currently very well protected against the use of nuclear weapons for domestic terrorism (non-ICBM-launched) by those nations or other groups, and that's exactly where I think the defense budget should be going. It is very obvious to the rest of the world, friends and enemies alike, that using an ICBM to strike the U.S. is about the toughest and least rewarding means of attack.
Clue: The problem already exists. It's not going away. Oh, and another clue, nobody is proposing to
not test the system before deploying it fully. The parts being proposed for early deployment are the
radars and other well developed bits and pieces.
The ICBM problem exists, but it is no longer the most important problem to be solved, and there are better ways to solve it than when ICBMs are already in flight. The problem is effectively solved by MAD, just as it has been for 40 years or so.
I'm not saying that the system has not been tested before deployment, I'm more worried that it has not been tested prior to us being asked to pay for it. The feasability of the system has not been proven and will not be proven completely for quite a while, perhaps even after it is fully deployed.
And if the commandos are not sucessful? Then what? We've tipped our hand. And to suggest that we
would launch a pre-emptive strike (along with the fantasy about commandos), is juvenile at best.
I don't think that it tips our hand for an adversary to find out that we have acted to remove their opportunity to make a nuclear threat against us; that is exactly what a rational person would expect the U.S. to do. It's about as juvenile as, say, dropping U.S. special operations troops into Afghanistan, or flying high-altitude spy planes right over the Soviet Union, or - oh wait, we did do those things. Sometimes life just imitates Tom Clancy, doesn't it:)
But seriously, I admit that there are difficulties in ensuring that attacks on the U.S. do not succeed. I just think it will be cheaper and easier to prevent those attacks before they occur, rather than while they are in progress. Especially since other means of attack are still quite available.
In the real world, possesion of nuclear arms implies the capability to influence events. They provide
influence by their mere existence, without overt threats of use.
Sure, but the U.S. is not significantly more at risk as long as that existence is known. It's the folks out there who have nuclear weapons that we don't know about that we should be more worried about, or perhaps the folks that do have nuclear weapons and then find their governments wobbling on the edge of collapse, like the USSR or modern-day Pakistan. This month's new nuclear state can be added to the U.S. MAD list quite easily. This month's new terrorist group with a suitcase nuke cannot be so targeted or threatened.
In the real world people posessing delicate, expensive, dangerous things like to keep them under lock
and key and close control. They also like the ability to use them with minimum warning and maximum
influence. That means ICBM's. While the threat of alternate delivery systems exists, in the real world,
the one where many nations are developing ICBM's, that threat is very small. In the real world we need
BMD to defend against real, existing, growing, threats.
I think it would be great to defend against everything - missile attacks, domestic nuclear terrorist attacks, chemical/biological attacks, and even box cutters on airliners. But I think a reasonable military planner will first defend their most likely avenues of attack against the most possible threats. And I think recent events have demonstrated that ICBM attack is not as likely a threat as we have planned for, and other types of attack are much more likely and much easier to carry out than we had thought. We would be remiss if we did not first shut down those avenues of attack (through intelligence gathering, actions against terrorist groups, nuclear nonproliferation treaties, international nuclear arms inspection agreements, and even preemptive attacks against credible present nuclear threats) before moving on to trying to shoot down fast-moving suborbital bits of metal with other such bits of metal. We may even find that those first efforts have reduced or eliminated the amount of missile defense which is required.
But what BMD does do is introduce an enormous amount of uncertainty into the attack.
The attacker will no longer be able to calculate 'I can launch x amount of my missiles (knowing they have y
reliability) and have a z percentage chance of destroying my targets'. A new and difficult term has been added to
the equation, and even though subtle it's important to military planners and politicians.
I don't see what this matters - an attacker is guaranteed MAD if they make such an attack. The certainty of their destruction is not in question. There isn't a whole lot of difference between 50% destruction of the U.S. and 100% destruction in my mind - either would be counted a great victory by an adversary in the minutes they had left to live. A missile defense that is not a near-perfect defense does not materially alter the situation militarily (well, other than the money that it's taken away from more worthwhile defensive efforts).
Also from a military point of view, jetliners and biochem attacks are
rather unpredictable and undependable in their effects, not a good thing. Also threats of those types of attacks can
not be used in advance to influence the behavior of the target nation, because once the threat is issued, security
gets stepped up, and the window of opportunity closes.
Exactly why we need to spend defensive money on security ahead of time, nuclear nonproliferation efforts worldwide, and intelligence to figure out what entities might have nuclear attack capabilities. Better to nip the problem in the bud than to rely on a questionable and largely unproven defense against a weapon (the ICBM) which will be used as much in the 21st century as the bow and arrow was in the 20th.
Nations and militaries *don't* build offense to execute
bolt-out-of-the-blue attacks, but build capabilities that allow them to influence events. (There are exceptions to
the rule, but they are just that exceptions.) International diplomacy isn't Civilization, but is far more complex and
subtle.
Any entity which does not have nuclear parity with the U.S. and makes nuclear threats against the U.S. will find themselves preemptively stopped, either by a team of commandos who destroy their nuclear potential or by a preemptive nuclear strike if necessary. Any state with nuclear parity is subject to MAD (and wouldn't be stopped by a missile shield anyway). The U.S. is thus only effectively more at risk from an unannounced nuclear attack from a small entity or group (the unannounced part means that "influencing events" beforehand isn't high on their agenda). A missile defense won't stop that adversary, since they'll just fly in a nuclear device on a jet airplane, private plane, or construct it within the U.S. itself. The only way to stop such a small and determined adversary is to use military intelligence and political savvy to defuse such attacks before they start. A missile defense will just be a "Maginot Line" in space for the 21st century - easily bypassed by those who really want to do harm in the U.S.
But leaving development machines exposed to the 'net (or even to much of your internal hardened network) is almost as egregious a sin. For example, in our lab any development machines with weak/no passwords are behind a router which will only let them talk to a few internal machines and won't let them talk to the outside world or the rest of the company at all.
Actually, that's backwards. Iraq was one of the most secular Middle Eastern states before they were on the outs with the U.S., and even though they're currently ruled by a despot, he's not really a religious despot. Communism in China is more of a state religion than Islam in Iraq.
As far as missile defense goes: it would not be a sufficient defense against an adversary with more than a few warheads, and so wouldn't prevent attacks by China at all. And as we've discovered, smaller adversaries don't need ICBMs at all to cause mass destruction in the U.S. We need a jetliner shield and a realistic chemical/biological threat prevention mechanism more than a missile shield at this point.
Well, the funny part was that they wanted Microsoft to pay $1.1 billion for the computers to run it on:) These links should really be their own/. story, or at least a SlashBack:
It's really more complicated than that, though, since the software costs are all up front and there is little ongoing cost to increased production of a particular software title. It's like printing money: once you've bought the presses, you can make almost any amount. And requiring this giveaway doesn't hurt Microsoft in its funding of the creation of new software because it can still sell the titles that it was forced to give away. There are a few cases where the schools would have bought software but won't now because they get it for free, but since these were poor schools to begin with this isn't a large amount of software or money. When you get right down to it, the giveaway any amount of a non-scarce item is never going to hurt Microsoft.
While you are correct from accepted business accounting standards, I think it would be a mistake to apply those standards to this case. In reality, Microsoft will be harmed very little by this giveaway, and they will also gain an opportunity to lock schools into Windows and shut out Apple. It's win-win-win all around for Microsoft. If this is contrary to standard accounting practices, then maybe our accounting practices should be altered to take into account business whose products are mostly intellectual property stamped onto cheap bits of plastic.
$1.1 billion worth of software does not cost Microsoft anything. It's essentially free for Microsoft to crank out more software since the R&D has already been paid for. That reduces this so-called "settlement" to just a Microsoft marketing campaign.
Best solution: they must contribute $billion or so of cold, hard, cash to a fund for school technology improvement. Then independent technical experts and educators can suggest uses for the money that don't necessarily benefit Microsoft. This settlement is a total victory for Microsoft - I'd hate to see what happens when they actually win a case...
So, if I use the text-based version of their website because I have java* turned off and their page doesn't load right, I get directed to the "lowbrow" page. Somehow I don't imagine that a surly and insulting user interface is really the direction to go in...
Of course something can be done about it. Unless you've let the government take away your guns. In that case, you may have some problems pulling off your revolution without a lot of civilian deaths. But that's the price of freedom, I guess.
That is exactly right - always keep all the packaging, and never let your Mom or (later) your wife toss them out or store them on a damp basement floor:) You never know when you'll need to ship and/or return something.
Unless they say that they were only iterating through the email address namespace, rather than scanning your particular web page. I'm not sure who the burden of proof would be on in that case.
Contrariwise, no anonymizing service is going to be able to retain legal services to fend off attacks on anonymity without having some form of income. So either some wealthy benefactor pays for "free" anonymity because they believe in it, or else everyone has to chip in to preserve their own privacy.
But on the other hand if you can say to the user "don't set it to do such-and-such since that's illegal", then I don't see the problem. At that point it's no longer the developer's problem; they can just expect the user to use it legally, just as your modem came with similar instructions.
Almost all installation boot disks seem to use ash, for pretty much the same reasons. I'm writing an installer right now an am using ash, in fact. For simple shell scripting, the extra size of bash just isn't worth it.
Slashdot Mirror
Well, one reason: regular users probably won't have installed "C/C++ Development" at their RedHat install time, and so don't have gcc, etc. available. Building from source is going to be a needless complication for most users, with really limited utility since they will just be reproducing what their distro's binary packages would provide anyway.
It seems to me that this whole thread is off on the wrong tangent. The equivalent of Windows setup.exe is double-clicking on your distribution's provided RPM file, and having KDE/Gnome be smart enough to run rpm -i for you. "Regular users" should only be installing from RPMs provided by their distribution maintainer and guaranteed to handle dependencies properly. Any time you get users that have to compile something from source or set up something from a foreign distro, you are way off of the parallel of Windows' setup.exe and are talking about something more akin to porting from one version of Windows to another.
Remember, preview before you troll :)
Good question. There are arenas that are tough to get a foothold in without some monetary support, and I would be more confident in my argument if I could point to the preponderance of open source projects being new and innovative ideas, rather than free clones of commercial software.
You are entirely correct; I read your post too quickly and assumed that you were making a leap that I've seen many others here make. Mea culpa.
However, since you go on to argue that it would be difficult to support a software industry economically based on entirely open source software, it sounds like you are arguing that "commercial" will essentially come out to be "proprietary" anyway.
I think the assumption that you are making is that the survival of a particular software industry is what's important for computing. I don't think that is really the case - otherwise the swallowing of the computing world by Microsoft would be the best outcome possible, rather than the worst, right? What is important is the computing experience which is provided to the users of the software; whether that experience can be better supplied by a commercial software industry or by the free software world is the big question. With a large enough user base, free software can drive itself forward - not built just during leisure time, but also as other users of the software hack on it to get it to work for their business needs. Considering the relative sizes of the current free software and commercial software communities, free software is already growing almost as fast, and is going much faster in some arenas. If free software were all that there were, the improvement would be even greater.
Linux was easy to learn, because I already knew HP-UX, and Linux was based on the same POSIX standards. I can't speak to Microsoft Office, not having tried to do anything complicated in it. And the POSIX standards were open for years, but cheap and reliable POSIX clones did not become readily available until people started handing around the source code to them. Open standards are good; open source to implement them with is better.
That's not exactly what I said. I said "computing has grown faster", not that a particular project has developed faster. Open source means more people contribute to projects, more people fork and produce interesting variants of projects, and user choice improves - thus computing itself, measured from the user perspective, grows faster. Although I will also point out that for large enough open source projects the project itself will be developed faster as development and debugging is done in parallel by many smarter people than one company could hire. Most projects are not Mozilla :) And even Mozilla is growing and blossoming in unpredictable ways now that there is a seasoned cadre of hackers that understand and make use of it.
I didn't say Microsoft was the only exception. And I'll point out that in many of those cases those companies have been hurt by their unethical or illegal actions. Just reeling off names does not disprove my point.
No.
I think this is begging the question - do we know for sure that an entirely free-software-based computing sector cannot grow and thrive just as well as a partly or entirely proprietary one? It seems to me that computing has grown faster as free software (and even open source software) has become more prominent. The only question is whether we have currently found the best "mix" between open and proprietary, or whether we should continue on to more and more degrees of freedom in order to improve software even more.
If there was any justice in the world (or in the U.S. government) then Microsoft would right now be wishing that they'd had such a conscience. Most of the time, keeping a business focused on ethics (or at least on staying within the law) is a good business practice, not a mistake. It saves you money and time in the long run. Microsoft just happens to have been a giant exception to this so far.
You've made some good points about matching the cost to expected future revenue based on those costs, and I can't dispute them on accounting terms. I guess in the end I feel that the damages in a case should actually impose some sort of punishment. Including Microsoft's standard markup from blank cdroms to boxed software effectively decreases the amount of punishment by about the markup amount, minus the sales which they would lose by the giveaway. Maybe it is only because the markup for software is so much higher than other goods that it bugs me, but I don't feel that the proposed penalty is just.
Here's how you can tell, in fact: give Microsoft the option to instead be fined $1.1 billion in cash, and then the cash will be given to a third party to buy software and hardware for the schools. Even if this third party were willing to buy all Microsoft software, and even if you made this alternate penalty 5% less painful, Microsoft would still go for the original "software giveaway" punishment, because it really is less painful for them to run off a few extra copies of Windows and lose a few sales (to poor schools who probably would have gone without or pirated anyway) than it would be for them to give up cash money. Every business will realistically account current cash money higher than an equivalent amount of yesterday's IP creation, no matter how much they report that R&D to be worth on their tax forms.
From your argument, I admit that price they set for the software is somewhat related to its production cost. But I think that as long as Microsoft is willing to give away the software rather than just pay a similarly-sized cash fine, they themselves are reputing your argument for the true cost of their product. I think I might be more willing to accept their offer if they would provide a real accounting of how much of their market prices are R&D, production, and profit, and would remove the profit percentage from their costs for the basis of this penalty calculation.
And this is all apart from the unquantifiable product lock-in opportunities which the "software giveaway" provides, of course. That in and of itself makes this punishment more of a "get out of jail free" than anything.
I think that the suggestion that the U.S. would not respond with overwhelming force in the face of any nuclear attack is a mistaken one to give to the world or to ourselves. The whole point of maintaining superiority in terms of this kind of weapon is to make sure that the consequences of a known nuclear attack on the U.S. are certain. The situation has not changed since the cold war: any nuclear attack on the U.S. which can be traced back to a particular country will be met with a massive and devastating response. To respond with any less than that would invite more attacks.
Any missile attack is traceable, and would invite a similar or greater retaliation from the U.S. No country is going to attack the U.S. with nuclear missiles unless they embrace destruction. Missile attacks themselves are pre-1990s, not post-2000 (or post-09-11-2001 I might say).
Read more carefully: the amount of damage doesn't really matter. A nuclear attack via missile on the U.S. will result in the nuclear destruction of the attacker. We are already using MAD to effectively handle this sort of threat. This threat will never go away, since there are already nations existing which could totally destroy the U.S. if they wanted to.
And heck, for any up-front expenditure X, it is possible to say that it's "cheaper" to just pay an additional %Y for so much more protection. The question is whether the upfront expenditure is worth it, or whether there are better ways to achieve the same goals.
MAD protects us against nations that launch missiles. I agree that we are not currently very well protected against the use of nuclear weapons for domestic terrorism (non-ICBM-launched) by those nations or other groups, and that's exactly where I think the defense budget should be going. It is very obvious to the rest of the world, friends and enemies alike, that using an ICBM to strike the U.S. is about the toughest and least rewarding means of attack.
The ICBM problem exists, but it is no longer the most important problem to be solved, and there are better ways to solve it than when ICBMs are already in flight. The problem is effectively solved by MAD, just as it has been for 40 years or so.
I'm not saying that the system has not been tested before deployment, I'm more worried that it has not been tested prior to us being asked to pay for it. The feasability of the system has not been proven and will not be proven completely for quite a while, perhaps even after it is fully deployed.
I don't think that it tips our hand for an adversary to find out that we have acted to remove their opportunity to make a nuclear threat against us; that is exactly what a rational person would expect the U.S. to do. It's about as juvenile as, say, dropping U.S. special operations troops into Afghanistan, or flying high-altitude spy planes right over the Soviet Union, or - oh wait, we did do those things. Sometimes life just imitates Tom Clancy, doesn't it :)
But seriously, I admit that there are difficulties in ensuring that attacks on the U.S. do not succeed. I just think it will be cheaper and easier to prevent those attacks before they occur, rather than while they are in progress. Especially since other means of attack are still quite available.
Sure, but the U.S. is not significantly more at risk as long as that existence is known. It's the folks out there who have nuclear weapons that we don't know about that we should be more worried about, or perhaps the folks that do have nuclear weapons and then find their governments wobbling on the edge of collapse, like the USSR or modern-day Pakistan. This month's new nuclear state can be added to the U.S. MAD list quite easily. This month's new terrorist group with a suitcase nuke cannot be so targeted or threatened.
I think it would be great to defend against everything - missile attacks, domestic nuclear terrorist attacks, chemical/biological attacks, and even box cutters on airliners. But I think a reasonable military planner will first defend their most likely avenues of attack against the most possible threats. And I think recent events have demonstrated that ICBM attack is not as likely a threat as we have planned for, and other types of attack are much more likely and much easier to carry out than we had thought. We would be remiss if we did not first shut down those avenues of attack (through intelligence gathering, actions against terrorist groups, nuclear nonproliferation treaties, international nuclear arms inspection agreements, and even preemptive attacks against credible present nuclear threats) before moving on to trying to shoot down fast-moving suborbital bits of metal with other such bits of metal. We may even find that those first efforts have reduced or eliminated the amount of missile defense which is required.
I don't see what this matters - an attacker is guaranteed MAD if they make such an attack. The certainty of their destruction is not in question. There isn't a whole lot of difference between 50% destruction of the U.S. and 100% destruction in my mind - either would be counted a great victory by an adversary in the minutes they had left to live. A missile defense that is not a near-perfect defense does not materially alter the situation militarily (well, other than the money that it's taken away from more worthwhile defensive efforts).
Exactly why we need to spend defensive money on security ahead of time, nuclear nonproliferation efforts worldwide, and intelligence to figure out what entities might have nuclear attack capabilities. Better to nip the problem in the bud than to rely on a questionable and largely unproven defense against a weapon (the ICBM) which will be used as much in the 21st century as the bow and arrow was in the 20th.
Any entity which does not have nuclear parity with the U.S. and makes nuclear threats against the U.S. will find themselves preemptively stopped, either by a team of commandos who destroy their nuclear potential or by a preemptive nuclear strike if necessary. Any state with nuclear parity is subject to MAD (and wouldn't be stopped by a missile shield anyway). The U.S. is thus only effectively more at risk from an unannounced nuclear attack from a small entity or group (the unannounced part means that "influencing events" beforehand isn't high on their agenda). A missile defense won't stop that adversary, since they'll just fly in a nuclear device on a jet airplane, private plane, or construct it within the U.S. itself. The only way to stop such a small and determined adversary is to use military intelligence and political savvy to defuse such attacks before they start. A missile defense will just be a "Maginot Line" in space for the 21st century - easily bypassed by those who really want to do harm in the U.S.
But leaving development machines exposed to the 'net (or even to much of your internal hardened network) is almost as egregious a sin. For example, in our lab any development machines with weak/no passwords are behind a router which will only let them talk to a few internal machines and won't let them talk to the outside world or the rest of the company at all.
Actually, that's backwards. Iraq was one of the most secular Middle Eastern states before they were on the outs with the U.S., and even though they're currently ruled by a despot, he's not really a religious despot. Communism in China is more of a state religion than Islam in Iraq.
As far as missile defense goes: it would not be a sufficient defense against an adversary with more than a few warheads, and so wouldn't prevent attacks by China at all. And as we've discovered, smaller adversaries don't need ICBMs at all to cause mass destruction in the U.S. We need a jetliner shield and a realistic chemical/biological threat prevention mechanism more than a missile shield at this point.
Well, the funny part was that they wanted Microsoft to pay $1.1 billion for the computers to run it on :) These links should really be their own /. story, or at least a SlashBack:
http://linuxtoday.com/news_story.php3?ltsn=2001-11 -20-017-20-PR-RH
http://www.redhat.com/opensourcenow/
It's really more complicated than that, though, since the software costs are all up front and there is little ongoing cost to increased production of a particular software title. It's like printing money: once you've bought the presses, you can make almost any amount. And requiring this giveaway doesn't hurt Microsoft in its funding of the creation of new software because it can still sell the titles that it was forced to give away. There are a few cases where the schools would have bought software but won't now because they get it for free, but since these were poor schools to begin with this isn't a large amount of software or money. When you get right down to it, the giveaway any amount of a non-scarce item is never going to hurt Microsoft.
While you are correct from accepted business accounting standards, I think it would be a mistake to apply those standards to this case. In reality, Microsoft will be harmed very little by this giveaway, and they will also gain an opportunity to lock schools into Windows and shut out Apple. It's win-win-win all around for Microsoft. If this is contrary to standard accounting practices, then maybe our accounting practices should be altered to take into account business whose products are mostly intellectual property stamped onto cheap bits of plastic.
$1.1 billion worth of software does not cost Microsoft anything. It's essentially free for Microsoft to crank out more software since the R&D has already been paid for. That reduces this so-called "settlement" to just a Microsoft marketing campaign.
Best solution: they must contribute $billion or so of cold, hard, cash to a fund for school technology improvement. Then independent technical experts and educators can suggest uses for the money that don't necessarily benefit Microsoft. This settlement is a total victory for Microsoft - I'd hate to see what happens when they actually win a case...
So, if I use the text-based version of their website because I have java* turned off and their page doesn't load right, I get directed to the "lowbrow" page. Somehow I don't imagine that a surly and insulting user interface is really the direction to go in...
Of course something can be done about it. Unless you've let the government take away your guns. In that case, you may have some problems pulling off your revolution without a lot of civilian deaths. But that's the price of freedom, I guess.
That is exactly right - always keep all the packaging, and never let your Mom or (later) your wife toss them out or store them on a damp basement floor :) You never know when you'll need to ship and/or return something.
Unless they say that they were only iterating through the email address namespace, rather than scanning your particular web page. I'm not sure who the burden of proof would be on in that case.
Contrariwise, no anonymizing service is going to be able to retain legal services to fend off attacks on anonymity without having some form of income. So either some wealthy benefactor pays for "free" anonymity because they believe in it, or else everyone has to chip in to preserve their own privacy.
"Help, help - I'm being oppressed!"
But on the other hand if you can say to the user "don't set it to do such-and-such since that's illegal", then I don't see the problem. At that point it's no longer the developer's problem; they can just expect the user to use it legally, just as your modem came with similar instructions.
Almost all installation boot disks seem to use ash, for pretty much the same reasons. I'm writing an installer right now an am using ash, in fact. For simple shell scripting, the extra size of bash just isn't worth it.
IIRC the phantoms from Final Fantasy were a result of an impact of a Leonid meteor.