You think this kind of contract and policy is signed by civil servants without the approval of a political appointee?
If anything, a civil servant is more likely to want to keep his future options open. The politicians are more interested in saying "I cut administrative overhead," without regard to whether that increases costs after the next election.
Yes, *maintenance* is a bitch. Which is why it is nice to have source code and a competitive support market.
Not all software has the same economic characteristics. An office suite, for example, is a totally different animal from a vertical market application. In government, there are *tons* of vertical niches, and that's where a "no open source" policy is a brutal cost. If one state public health department has developed a really useful disease surveillance, Minnesota can't use it unless they *buy it* from a consultant, and are married to that consultant for as long as they want to use the software.
Likewise if another state has developed a really neat distribution system for GIS data based on open standards, Minnesota will have to build their own system from scratch or drink the ESRI kool-aid, which entails the classic interlocking proprietary systems business model. Just the *training classes* they'd need exceed the cost of learning how to run the other state's system after copying it for free.
A "no open source" policy is not only just plain stupid, it's probably unenforceable. There's bound to be all kinds of BSD licensed stuff in the proprietary software and systems used by an organization as large as a state.
I like patent 6826762, in which Microsoft patents the use of hardware independent software drivers when applied to cell phones. What moron at the patent office approved that one?
Or how about 6909910 "Method and System for Managing Changes to a Contact Database". The invention amounts to this: when the user wants to save the last phone call as a contact, you look to see whether that phone number is in the contact database. If it is, you bring that contact up for editing. If it is not, you create a new contact pre-populating the phone field with the last number called.
Seriously. How in the world does the patent office grant such rubbish patents? Do they go out of their way to hire clueless people, or do they have a special training program?
And disease is a problem because they're packed like sardines into a feedlot eating food which weakens their resistance to infection...
This can go round and round endlessly. In the end it comes down to this: we'd probably be better off buying grass fed beef (for that fat content and profile if nothing else), but that would cost more and it would taste different from what we're used to, so it will never happen.
Well, there's reason to doubt the effectiveness of Twitter in the Iranian case. I don't happen to buy the author's argument (that many of the tweets were originating outside of Iran) as either here nor there. That does not mean the information necessarily originated outside of Iran, or even if it did whether it really mattered.
The real bottom line is that the government forces won; it retained power, albeit at a loss of international respect. That might change the course of history in the long run, but it hasn't yet.
That said, the author's argument amounts to this: (a) the kind of social media mediated revolution scenarios some have hypothesized haven't happened yet and (b) historical revolutions have not worked the way those scenarios are envisioned to work. Even if we accept both these statements, it tells us *nothing* about the feasibility of these scenarios, at least yet. The whole hypothesis is that social media create a new avenue for social change, and that really can't be disproven by counterexamples from before the technology existed.
In a nutshell, there's no compelling evidence for either side of this argument. In fact, I suspect there never will be.
I think it almost certain that there will be revolutions in the future (or counter-revolutions) in which social media play a seemingly dramatic part. It will certainly *look* like the technology played a decisive role in these events. But won't never know for sure what would have happened absent the technology. Perhaps somebody lives who would have died because he gets the tweet about a paramilitary roadblock. Another man dies because the tweet is traced back to his phone. These are the kinds of events that can change history, but that kind of thing has always happened. An opposition leader chooses a flight that happens to crash and the public blames the government. Does this mean aviation is responsible for the revolutionary outburst that follows?
Well, that's an interesting observation, and it leads to the following: what do you *mean* by "theory of everything"?
Let's distinguish between a theory that connects all the forces of physics (gravitation, electromagnetism, strong and weak nuclear forces) and a theory that is capable of predicting (by calculation) the outcome of any conceivable physical experiment. Let's call the first type a "unifying theory of everything" and the later an "oracular theory of everything". An oracular theory of everything can predict the outcome of any experiment in a finite amount of time, albeit possibly a very long, long time. I think that's what the layman thinks when he hears "theory of everything".
For oracular theories, we do have to take things like the incompleteness of consistent arithmetic systems into account, because it is possible to devise physical experiments whose outcome is dependent upon the truth of some arithmetic statement. The theory should be able to predict the result, therefor an oracular theory is necessarily also a description of arithmetic.
This leads to the question is whether such a description is necessarily equivalent to some kind of formal system. I think it is. It contradicts the notion of being able to predict any outcome to qualify that by saying, "but we might never find the right representation of the problem" or "an actual answer might take forever to obtain." An oracle worthy of the name must be able to answer if you merely set the problem up a certain way.
Personally, I think the halting problem is perhaps a better model to question the mathematical feasibility of an oracular "theory of everything". Running a problem on a computer is certainly a physical process, and therefore it can be regarded as an experiment. Even quantum computing, which examines large numbers of possibilities at once, still deals with *finite* sets of alternatives. Therefore such technology does not alter which calculations are possible in a finite span of time.
The same cases cannot be made as easily against a merely unifying theory of everything (for one thing you'd actually have to know something about physics). A unifying theory does not have to guarantee our ability to extrapolate anything. That seems more reasonable. We can build a physical turning machine that can easily be modeled by Newtonian mechanics, but that doesn't mean Newtonian mechanics allows us to predict whether it uses a finite amount of energy for some set of initial conditions (i.e., halts). Despite this limitation, Newtonian physics does an admirable job predicting the trajectory of a cannon ball.
It is quite possible that *fundamental* computer science and quantum physics may become intertwined in future years. The theory of computation is all about limitations on gaining knowledge through computation. Some things can't be guaranteed knowable in a finite span of time or space; other ones may not be knowable in any conceivably practical span. It is possible that we will encounter limitations on prediction of the behavior of physical systems that are usable as models of computation. This could conceivably (in my non-physicist's limited brain) lead to unifying theories that might not be capable of *any* practical predictions (e.g., we can show mathematically that our model connecting gravity to nuclear forces gives the right answer, but it might take us forever to actually perform that feat).
Obviously, I know that lifetime institutionaliation costs more than $300,000. I've seen the calculations for vector borne encephalitis when $10M/case averted was considered reasonable in the 1990s.
I'm using institutionalization as an extreme case, the right end of the scale where "paper cut" is on the left end. For sake of argument, I'm assuming we're most concerned with injuries whose responses fall in the range between first aid and a trip to the emergency room. That seems a reasonable range of severity to consider for a kit made out of common, everyday items.
The point is that the scale of distribution governs what is economically rational. If we have reasonable expectation of injuries requiring extended hospitalization, we aren't going to give that kit to a *single* user until it's been examined by somebody who really knows what he's doing. But that assumption negates the value of the thought experiment. I needed assumptions that are reasonable, yet favorable to the null hypothesis, which is that you *never* have to think about the safety of a kit that's made out of common, everyday items.
"Biodegradable" isn't some kind of exotic material property. They're using cardboard.
Non-biodegradability is something that we've engineered into materials we want to persist for a minimum length of time. We just haven't paid quite as much attention to the *maximum* lifespan because once the consumer takes Barbie out of her packaging, the fate of that packaging is so far removed from the manufacturer it is somebody else's problem.
Just to play devil's advocate here: what amounts to reasonable precautions is a function of scale, because you can amortize the cost of each expected injury over a larger number of units shipped.
As a thought experiment, suppose Miss Jones the science teacher puts together a science experiment kit for each of her 30 students. A representative kit is then sent to a safety engineering firm, which charges $10K to conclude there is a 0.2% chance of injury from the ruler, and that this could be reduced to 0.1% by using a slightly different ruler.
Now it almost certain that nobody is gong to be hurt by the offending ruler, and the engineering investment of $10K prevents an expected 0.03 injuries. That's over $300,000 per injury averted. That makes no economic sense unless the injury is horrific (e.g. requires lifetime institutionalization).
Now suppose JonesCo puts together a similar kit, and expects to ship 30 thousand units. In that case, it is almost *certain* that somebody is going to get hurt, although any *individual's* chance is quite small. The expected number of injuries saved by the engineering study is now 30. Amortizing the $10K study costs over 30 injuries means that you've spent just over 300 per injury saved. This is not quite justifiable for things like paper cuts of course, but an emergency room visit probably costs more than that.
So: the costs involved with a safety review may or may not be justifiable depending on the number of units that will be shipped.
In either case, the safety of the pre-study and post-study kits are practically indistinguishable. As a parent, I wouldn't freak out if the Miss Jones kit was used in place of the JonesCo kit, because we are talking about very, very rare accidents. But those freak accidents *do* happen and are worth considering *collectively*. I say this as a parent who has taken a toddler to the emergency room for an injury at preschool who sent that child right back to the same school the next day with the heart shaped bead he'd shoved up into his sinuses in his pocket.
See a problem? I do. You're lumping welfare and social security together. That makes no sense; not only are they very different programs, they are financially different animals.
It also misleadingly suggests that half our budget is going to welfare queens. That is simply not possible. The total budget of Administration for Families and Children (otherwise known as welfare) for 2011 is 17.48 billion, which is actually less than the 18.7 billion outlay in the fiscal year for NASA.
Medicaid is a big program, but still nowhere near 50% of the budget. For FY 2011, the Medicaid budget is 297 billion. Medicare is even bigger at 491 billion. If you added up Medicaid, Medicare and welfare, you still less than the money spent on defense, so these can hardly break the 50% of the budget mark.
To do that, you have to add social security into the mix, but that's inherently misleading from a budget balancing standpoint. Social Security brings in income. A *lot* of income. In fact it runs a surplus. To get an accurate picture, you have to look at both the expense *and* income side.
Here are the top sources of income in the US budget (in billions of dollars):
Individual Income Taxes: 1,121 or 43.7%. Social Security(payroll) Taxes: 934 or 36.4%. Corporate Income Taxes: 297 or 11.6%. Excise Taxes: 80 or 3.12%. Federal Reserve Deposits:79 or 3.08%. Customs Duties:29 or 1.13%. Estate Taxes: 24 or 0.94%. Everything Else (roughly): 10 or 0.39%.
See the problem? Since Social Security expenditures are 730 billion, if you waved a magic wand and made that program disappear, you'd add 204 billion dollar to the budget deficit. That's on the same order of magnitude as *all corporate taxes* added up. It's fairly safe to say that without the Social Security surplus, there wouldn't be 18+ billion dollars lying around to spend on NASA.
If we had a sensible approach to this, we'd set social security to one side and offset the cash influx with the expected liability for future payments. Then we'd invest the surplus in an instrument that paid interest, the goal being to ensure the cash flow remains balanced over the lifetime of the bulk of the people in the system.
But we don't do that. Instead we wring our hands about an entirely foreseeable and manageable problem, then take the money that could deal with that problem, the working man's 204 billion dollar contribution to deficit reduction, and throw into things that don't benefit him. But to truthful if we did manage the social security surplus responsibly, there probably wouldn't be money for NASA under that scenario either.
Now it *is* a politically conceivable scenario to get rid of Social Security and Medicaid (the notion of Medicare going away is fantasy). The 200 billion in surplus lost would be more than offset by 297 reduction in outlays. But if you think that anything like a proportionate share of that 97 billion dollars is going back into your pocket, you're either dreaming, or a member of a very small group of very wealthy people. So in that scenario, the working guy loses the programs that provide him security against tough times, but the programs that benefit the wealthy aren't going anywhere.
It's my understanding that Stuxnet was designed to only *do only* to one certain computer/system that was specifically targeted. On all other computers that do not match the signature of that computer, it leaves them alone. So what is the "havoc" that it is causing?
The havoc of trying to keep the damned thing from spreading is enough. You do not let the black hats keep the keys to your systems, especially when their tool is capable of updating itself using P2P techniques. We should probably add when we say anything about this worm "so far as we know." There's a lot of encrypted stuff whose purpose we don't know. Furthermore, it is possible that the whole ecosystem of infected machines could be updated to do something we haven't imagined yet.
One thing that occurs to me is that the whole Siemens business might be a "cover story". The worm might have completely unrelated purposes we don't know about.
Miss them? Not me. I'll miss the neighborhood video store that kept a stockpile of unusual movies, not Blockbuster with its dozens of duplicate copies of movies that were in theaters last year. That store easily had twice the number of titles in 1/5 the space.
Well, that's the peace dividend for you. We used to "stockpile" physics PhDs like they were bomb warheads. The rate slacked off after 1970, picking up briefly during the Reagan administration.
But we still have the giant physicist factories fully manned because with tenure there isn't really an effective way to dismantle them.
That's a good point, but it's hardly worth arguing that proprietary software is unethical if it is the only viable model. Since I don't think Stallman is arguing that people should do without software, I think we can take it as an implicit assumption that free software can ultimately meet peoples' needs.
while in fact majority always represents just the most marketed, the most advertised, the most imposed position.
[emphasis mine]
OK, as a first approximation, you are right, but to say "always" is to overstate the case. There is no question that people who invest in propaganda don't do so out of naiveté... they expect a return on their investment. They expect to wield public sentiment like a tool, but it is a treacherous tool.
I think propaganda works best when it is directs people's attention away from their day to day lives, as opposed to changing their assessment of those lives. You can say, "your life is hell because of the Jews" or "you are insecure because of the homosexual agenda." You can't say, "your life is actually pretty good so far as the world standard of living is concerned," even if that is true. You can't say "it's actually quite easy to get a job; people who don't have jobs are just lazy," unless you are talking to somebody with a secure job.
If you could simply manufacture the opinions you wanted, then the public would have continued to favor the Iraq war in the run up to the 2008 elections, but the war had gone on so long that people were touched by it in some way, by a family member, friend or colleague who was deployed and maybe didn't come back. Likewise the Democrats are going to pay in 2010 because they can't credibly claim to have improved peoples' lives in the twenty months they've had power. That's common in mid-term elections.
In such cases, propaganda has a way of turning on its masters.
Perhaps we should evaluate people's political sanity not on their absolute position on some political axis, but on their open or narrow mindedness. A political position becomes pernicious fantasy, no matter where it is on your favorite philosophical axis, when it willfully ignores the probable outcomes of the actions it advocates.
For example, other people with me on the left favored single payer health insurance or even a socialized medical system during the recent debates on health insurance reform. While I am philosophically well disposed to these things, I did not favor them at that time. I thought if they were enacted that existing businesses would immediately collapse, and that working public replacements could not be conjured into existence quickly enough to take their place. Now I realize many who prefer socialized medicine or single payer (not the same things at all by the way) might disagree with that assessment. They may even be right. But that's not the point I'm trying to make. I moderated my position based on a critical examination of the likely outcomes of my *ideal* solution. That examination might be faulty, but I did not twist my evaluation of the facts in order to justify my a priori position.
It's tricky to evaluate the political sanity of a figure like Stallman. He is very, very bright,and bright people have a way of finding credible sounding rationalizations for really ill considered opinions. That said, I think that Stallman's positions on the viability of free software sound a lot more credible today than they did twenty years ago. True, free software projects haven't produced viable competition in a number of important niches; but after two decades of experience with free software success, it isn't so hard to believe that a free software ecosystem could meet all the software needs of an individual or enterprise.
Well -- this article really poses something that is really more like a thought experiment than a recommendation.
Case 1: Imagine you have *no* firewall software on desktops and laptops. How would you secure your network?
Case 2: Imagine all your desktop and laptop machines have software firewalls that you completely trust to be perfect. What things would you do differently than in Case 1?
Case 3: All our desktops and laptops have firewall software on them that you don't consider perfect, but is nonetheless pretty good. You have a lot of stuff on your plate and not quite enough time and budget to do everything you ought to. What things from Case 1 do you stop doing?
Case 4: Like Case 3, but your most clueless coworker is making the decision.
Case 5: Like Case 4, but prolonged indefinitely.
I think this guy is making the case that desktop firewalls are harmful in that they induce network administrators to put their trust in them. Naturally, if you are Mr. Anal Retentive NetAdmin (as you ought to be), this doesn't apply to you.
sometimes making "simple" solutions actually work is more complicated than the "complex" alternative. As an engineer you run into this all the time, the manager who's so enamored of his brilliance he can't see the flaws in his idea.
This guy is talking about creating artificial peat bogs. It actually *is* an intriguing idea, but I don't see it as "simple". It certainly isn't an "alternative" to government subsidies or regulation. Somebody is going to have to pay the farmers to do this, and to buy the land and transport the waste there, and to deal with the effects of removing so much biomass from th cropland.
Things like grassahol subsidies are supposed to incent the development of new technologies. If those technologies ever make grassahol cheaper than oil, then the subsidies will have bootstrapped a new private grassahol market. That's a big "if", but so is research in energy technologies like fusion. What is problematic is that the farm lobby distorts the program, just as it would a farm waste sequestration program. So it's not a politically simpler solution.
$100 million dollars to scrub 1.5% of the carbon out of the atmosphere sounds like a huge amount of money, until you consider this. A single F22 Raptor cost half again as much (150 million), and we've managed to purchase 166 of those. Most people would admit that is a lot of money, but there are still people that think buying a few more would be a good investment. When you're talking about an entire national economy like the US, 100 million to accomplish something important isn't that much.
Now consider the damage figure for Hurricane Katrina, which stands at 81 billion. Now you can't say that any hurricane was *caused* by global warming, but severe hurricanes are more *frequent* under an AGW scenario. If the frequency of such hurricanes increases, a few billion dollars to dial that down wouldn't be that much money, much less 100 million.
OK, what will you build this "simple" "seawall" out of? Riprap? Nope. These are coral atolls we're talking about. Very *remote* coral atolls. You'd have quarry enough granite to wrap 33 islands somewhere then transport it to the middle of the Pacific Ocean. Not cheap. Not easy. Not simple.
OK, what about concrete? Well, maybe they can use coral for aggregate, but they don't have limestone, so they've got to import cement. Any idea how much it would cost to import enough Portland cement to build a wall around 33 islands to the required height? And the steel? You're going to have a *huge* and *very expensive* engineering project just transporting and staging the materials. OK, so a concrete wall around the county is not simple either.
But of course, what you're really talking about is creating a system of dikes to transform the country into 33 empolderments. What are dikes made out of? Well, they are primarily earthworks stabilized with vegetation. What do these people have to work with? Coral sand. How do you propose to make it stay where it's put?
It's easy to sit in your armchair and call the people who have to solve this problem *for real* "childish" and call their problems "whining". In fact, the less you know the easier it is.
Furthermore, you obviously haven't read the article. The problem with the marine sanctuary is that fishing permits (for foreign tuna fleets -- a detail the article should mention) are a major part of *government* revenue. Kiribati thinks that it deserves some revenue for preserving this huge hunk of the Earth for everyone else, and they're right.
True. On the other hand it has the least orthogonal trigger language of any commonly used RDBMS, and has an absolutely dreadful parser that can't be trusted with unusual cases (like bound variables in subqueries) and is unusually restrictive about the use of column aliases. These are issues that don't matter to most MS-SQL users, who really like the integration with the MS tool stack.
On the other hand, the Oracle peculiarity about '' and NULL although aesthetically ugly, has almost no practical importance.
"there are always ways that all parties can resolve their legitmate conflicts to the greater benefit of everyone," was thrown out the door and stomped on by Hitler following the partition of Czechoslovakia.
That would only be demonstrated by your example if the outcome of WW2 was the best possible outcome for Germany. I'd argue that Hitler's strategy was less than optimal for Germany, since it led to regime change, partition, and territorial loss. That was a less than perfect outcome not only for the nation, but for the regime, so it actually supports the hypothesis that warmongering is irrational.
The interesting question is this: given that Nazi Germany pursued a cynically self-interested policy with respect to rearmament and territorial acquisition through the thirties, why didn't they continue to pursue their self interest when it required moderating their aggressive tendencies?
I'd argue that it's a bit like the entrepreneur who starts a promising business but somehow can't make the transition from start-up to viable business. The skills and culture needed to seize power and begin the acquisition of an empire are different from those needed to run an empire. By 1939 they were locked into behavior that had succeeded brilliantly for them over the past decade at a time when they would have been better served by slowing down. If Hitler had given Chamberlain what Chamberlain promised his people -- "peace in our time" -- if he'd put Poland on the back burner for a few years, he *might* have had a chance to stay in power for decades longer. If he played his cards skillfully, he might have seen a pro-Nazi government in Britain and an anti-communist government in the US that would have welcomed his challenging the expansion of a Soviet sphere of influence in Eastern Europe.
I believe that kind of regime tends to end up riding the tiger. Their need for a steady stream of scapegoats and conquests makes them aggressive and successful in the short term, like cancer.
So it is both true that there is always a better way to settle differences than war, but it is also true that you can't count on people pursuing their rational self interests.
Slashdot Mirror
You think this kind of contract and policy is signed by civil servants without the approval of a political appointee?
If anything, a civil servant is more likely to want to keep his future options open. The politicians are more interested in saying "I cut administrative overhead," without regard to whether that increases costs after the next election.
Yes, *maintenance* is a bitch. Which is why it is nice to have source code and a competitive support market.
Not all software has the same economic characteristics. An office suite, for example, is a totally different animal from a vertical market application. In government, there are *tons* of vertical niches, and that's where a "no open source" policy is a brutal cost. If one state public health department has developed a really useful disease surveillance, Minnesota can't use it unless they *buy it* from a consultant, and are married to that consultant for as long as they want to use the software.
Likewise if another state has developed a really neat distribution system for GIS data based on open standards, Minnesota will have to build their own system from scratch or drink the ESRI kool-aid, which entails the classic interlocking proprietary systems business model. Just the *training classes* they'd need exceed the cost of learning how to run the other state's system after copying it for free.
A "no open source" policy is not only just plain stupid, it's probably unenforceable. There's bound to be all kinds of BSD licensed stuff in the proprietary software and systems used by an organization as large as a state.
I like patent 6826762, in which Microsoft patents the use of hardware independent software drivers when applied to cell phones. What moron at the patent office approved that one?
Or how about 6909910 "Method and System for Managing Changes to a Contact Database". The invention amounts to this: when the user wants to save the last phone call as a contact, you look to see whether that phone number is in the contact database. If it is, you bring that contact up for editing. If it is not, you create a new contact pre-populating the phone field with the last number called.
Seriously. How in the world does the patent office grant such rubbish patents? Do they go out of their way to hire clueless people, or do they have a special training program?
They can't make a decision on how to tie their collective shoes together...
I believe someday they'll manage tie their shoes together, and collectively plunge forward into a brave new world.
And disease is a problem because they're packed like sardines into a feedlot eating food which weakens their resistance to infection...
This can go round and round endlessly. In the end it comes down to this: we'd probably be better off buying grass fed beef (for that fat content and profile if nothing else), but that would cost more and it would taste different from what we're used to, so it will never happen.
Well, there's reason to doubt the effectiveness of Twitter in the Iranian case. I don't happen to buy the author's argument (that many of the tweets were originating outside of Iran) as either here nor there. That does not mean the information necessarily originated outside of Iran, or even if it did whether it really mattered.
The real bottom line is that the government forces won; it retained power, albeit at a loss of international respect. That might change the course of history in the long run, but it hasn't yet.
That said, the author's argument amounts to this: (a) the kind of social media mediated revolution scenarios some have hypothesized haven't happened yet and (b) historical revolutions have not worked the way those scenarios are envisioned to work. Even if we accept both these statements, it tells us *nothing* about the feasibility of these scenarios, at least yet. The whole hypothesis is that social media create a new avenue for social change, and that really can't be disproven by counterexamples from before the technology existed.
In a nutshell, there's no compelling evidence for either side of this argument. In fact, I suspect there never will be.
I think it almost certain that there will be revolutions in the future (or counter-revolutions) in which social media play a seemingly dramatic part. It will certainly *look* like the technology played a decisive role in these events. But won't never know for sure what would have happened absent the technology. Perhaps somebody lives who would have died because he gets the tweet about a paramilitary roadblock. Another man dies because the tweet is traced back to his phone. These are the kinds of events that can change history, but that kind of thing has always happened. An opposition leader chooses a flight that happens to crash and the public blames the government. Does this mean aviation is responsible for the revolutionary outburst that follows?
Well, that's an interesting observation, and it leads to the following: what do you *mean* by "theory of everything"?
Let's distinguish between a theory that connects all the forces of physics (gravitation, electromagnetism, strong and weak nuclear forces) and a theory that is capable of predicting (by calculation) the outcome of any conceivable physical experiment. Let's call the first type a "unifying theory of everything" and the later an "oracular theory of everything". An oracular theory of everything can predict the outcome of any experiment in a finite amount of time, albeit possibly a very long, long time. I think that's what the layman thinks when he hears "theory of everything".
For oracular theories, we do have to take things like the incompleteness of consistent arithmetic systems into account, because it is possible to devise physical experiments whose outcome is dependent upon the truth of some arithmetic statement. The theory should be able to predict the result, therefor an oracular theory is necessarily also a description of arithmetic.
This leads to the question is whether such a description is necessarily equivalent to some kind of formal system. I think it is. It contradicts the notion of being able to predict any outcome to qualify that by saying, "but we might never find the right representation of the problem" or "an actual answer might take forever to obtain." An oracle worthy of the name must be able to answer if you merely set the problem up a certain way.
Personally, I think the halting problem is perhaps a better model to question the mathematical feasibility of an oracular "theory of everything". Running a problem on a computer is certainly a physical process, and therefore it can be regarded as an experiment. Even quantum computing, which examines large numbers of possibilities at once, still deals with *finite* sets of alternatives. Therefore such technology does not alter which calculations are possible in a finite span of time.
The same cases cannot be made as easily against a merely unifying theory of everything (for one thing you'd actually have to know something about physics). A unifying theory does not have to guarantee our ability to extrapolate anything. That seems more reasonable. We can build a physical turning machine that can easily be modeled by Newtonian mechanics, but that doesn't mean Newtonian mechanics allows us to predict whether it uses a finite amount of energy for some set of initial conditions (i.e., halts). Despite this limitation, Newtonian physics does an admirable job predicting the trajectory of a cannon ball.
It is quite possible that *fundamental* computer science and quantum physics may become intertwined in future years. The theory of computation is all about limitations on gaining knowledge through computation. Some things can't be guaranteed knowable in a finite span of time or space; other ones may not be knowable in any conceivably practical span. It is possible that we will encounter limitations on prediction of the behavior of physical systems that are usable as models of computation. This could conceivably (in my non-physicist's limited brain) lead to unifying theories that might not be capable of *any* practical predictions (e.g., we can show mathematically that our model connecting gravity to nuclear forces gives the right answer, but it might take us forever to actually perform that feat).
Obviously, I know that lifetime institutionaliation costs more than $300,000. I've seen the calculations for vector borne encephalitis when $10M/case averted was considered reasonable in the 1990s.
I'm using institutionalization as an extreme case, the right end of the scale where "paper cut" is on the left end. For sake of argument, I'm assuming we're most concerned with injuries whose responses fall in the range between first aid and a trip to the emergency room. That seems a reasonable range of severity to consider for a kit made out of common, everyday items.
The point is that the scale of distribution governs what is economically rational. If we have reasonable expectation of injuries requiring extended hospitalization, we aren't going to give that kit to a *single* user until it's been examined by somebody who really knows what he's doing. But that assumption negates the value of the thought experiment. I needed assumptions that are reasonable, yet favorable to the null hypothesis, which is that you *never* have to think about the safety of a kit that's made out of common, everyday items.
"Biodegradable" isn't some kind of exotic material property. They're using cardboard.
Non-biodegradability is something that we've engineered into materials we want to persist for a minimum length of time. We just haven't paid quite as much attention to the *maximum* lifespan because once the consumer takes Barbie out of her packaging, the fate of that packaging is so far removed from the manufacturer it is somebody else's problem.
Just to play devil's advocate here: what amounts to reasonable precautions is a function of scale, because you can amortize the cost of each expected injury over a larger number of units shipped.
As a thought experiment, suppose Miss Jones the science teacher puts together a science experiment kit for each of her 30 students. A representative kit is then sent to a safety engineering firm, which charges $10K to conclude there is a 0.2% chance of injury from the ruler, and that this could be reduced to 0.1% by using a slightly different ruler.
Now it almost certain that nobody is gong to be hurt by the offending ruler, and the engineering investment of $10K prevents an expected 0.03 injuries. That's over $300,000 per injury averted. That makes no economic sense unless the injury is horrific (e.g. requires lifetime institutionalization).
Now suppose JonesCo puts together a similar kit, and expects to ship 30 thousand units. In that case, it is almost *certain* that somebody is going to get hurt, although any *individual's* chance is quite small. The expected number of injuries saved by the engineering study is now 30. Amortizing the $10K study costs over 30 injuries means that you've spent just over 300 per injury saved. This is not quite justifiable for things like paper cuts of course, but an emergency room visit probably costs more than that.
So: the costs involved with a safety review may or may not be justifiable depending on the number of units that will be shipped.
In either case, the safety of the pre-study and post-study kits are practically indistinguishable. As a parent, I wouldn't freak out if the Miss Jones kit was used in place of the JonesCo kit, because we are talking about very, very rare accidents. But those freak accidents *do* happen and are worth considering *collectively*. I say this as a parent who has taken a toddler to the emergency room for an injury at preschool who sent that child right back to the same school the next day with the heart shaped bead he'd shoved up into his sinuses in his pocket.
See a problem? I do. You're lumping welfare and social security together. That makes no sense; not only are they very different programs, they are financially different animals.
It also misleadingly suggests that half our budget is going to welfare queens. That is simply not possible. The total budget of Administration for Families and Children (otherwise known as welfare) for 2011 is 17.48 billion, which is actually less than the 18.7 billion outlay in the fiscal year for NASA.
Medicaid is a big program, but still nowhere near 50% of the budget. For FY 2011, the Medicaid budget is 297 billion. Medicare is even bigger at 491 billion. If you added up Medicaid, Medicare and welfare, you still less than the money spent on defense, so these can hardly break the 50% of the budget mark.
To do that, you have to add social security into the mix, but that's inherently misleading from a budget balancing standpoint. Social Security brings in income. A *lot* of income. In fact it runs a surplus. To get an accurate picture, you have to look at both the expense *and* income side.
Here are the top sources of income in the US budget (in billions of dollars):
Individual Income Taxes: 1,121 or 43.7%.
Social Security(payroll) Taxes: 934 or 36.4%.
Corporate Income Taxes: 297 or 11.6%.
Excise Taxes: 80 or 3.12%.
Federal Reserve Deposits:79 or 3.08%.
Customs Duties:29 or 1.13%.
Estate Taxes: 24 or 0.94%.
Everything Else (roughly): 10 or 0.39%.
See the problem? Since Social Security expenditures are 730 billion, if you waved a magic wand and made that program disappear, you'd add 204 billion dollar to the budget deficit. That's on the same order of magnitude as *all corporate taxes* added up. It's fairly safe to say that without the Social Security surplus, there wouldn't be 18+ billion dollars lying around to spend on NASA.
If we had a sensible approach to this, we'd set social security to one side and offset the cash influx with the expected liability for future payments. Then we'd invest the surplus in an instrument that paid interest, the goal being to ensure the cash flow remains balanced over the lifetime of the bulk of the people in the system.
But we don't do that. Instead we wring our hands about an entirely foreseeable and manageable problem, then take the money that could deal with that problem, the working man's 204 billion dollar contribution to deficit reduction, and throw into things that don't benefit him. But to truthful if we did manage the social security surplus responsibly, there probably wouldn't be money for NASA under that scenario either.
Now it *is* a politically conceivable scenario to get rid of Social Security and Medicaid (the notion of Medicare going away is fantasy). The 200 billion in surplus lost would be more than offset by 297 reduction in outlays. But if you think that anything like a proportionate share of that 97 billion dollars is going back into your pocket, you're either dreaming, or a member of a very small group of very wealthy people. So in that scenario, the working guy loses the programs that provide him security against tough times, but the programs that benefit the wealthy aren't going anywhere.
NEVER underestimate a determined hacker.
Er...So that means we should *correctly estimate* a determined hacker, right?
Which groups of people should we estimate incorrectly? That's probably a shorter list to remember.
It's my understanding that Stuxnet was designed to only *do only* to one certain computer/system that was specifically targeted. On all other computers that do not match the signature of that computer, it leaves them alone. So what is the "havoc" that it is causing?
The havoc of trying to keep the damned thing from spreading is enough. You do not let the black hats keep the keys to your systems, especially when their tool is capable of updating itself using P2P techniques. We should probably add when we say anything about this worm "so far as we know." There's a lot of encrypted stuff whose purpose we don't know. Furthermore, it is possible that the whole ecosystem of infected machines could be updated to do something we haven't imagined yet.
One thing that occurs to me is that the whole Siemens business might be a "cover story". The worm might have completely unrelated purposes we don't know about.
Miss them? Not me. I'll miss the neighborhood video store that kept a stockpile of unusual movies, not Blockbuster with its dozens of duplicate copies of movies that were in theaters last year. That store easily had twice the number of titles in 1/5 the space.
Well, that's the peace dividend for you. We used to "stockpile" physics PhDs like they were bomb warheads. The rate slacked off after 1970, picking up briefly during the Reagan administration.
But we still have the giant physicist factories fully manned because with tenure there isn't really an effective way to dismantle them.
Sure you can vacuum up the money, but how do you launder it afterward?
That's a good point, but it's hardly worth arguing that proprietary software is unethical if it is the only viable model. Since I don't think Stallman is arguing that people should do without software, I think we can take it as an implicit assumption that free software can ultimately meet peoples' needs.
while in fact majority always represents just the most marketed, the most advertised, the most imposed position.
[emphasis mine]
OK, as a first approximation, you are right, but to say "always" is to overstate the case. There is no question that people who invest in propaganda don't do so out of naiveté... they expect a return on their investment. They expect to wield public sentiment like a tool, but it is a treacherous tool.
I think propaganda works best when it is directs people's attention away from their day to day lives, as opposed to changing their assessment of those lives. You can say, "your life is hell because of the Jews" or "you are insecure because of the homosexual agenda." You can't say, "your life is actually pretty good so far as the world standard of living is concerned," even if that is true. You can't say "it's actually quite easy to get a job; people who don't have jobs are just lazy," unless you are talking to somebody with a secure job.
If you could simply manufacture the opinions you wanted, then the public would have continued to favor the Iraq war in the run up to the 2008 elections, but the war had gone on so long that people were touched by it in some way, by a family member, friend or colleague who was deployed and maybe didn't come back. Likewise the Democrats are going to pay in 2010 because they can't credibly claim to have improved peoples' lives in the twenty months they've had power. That's common in mid-term elections.
In such cases, propaganda has a way of turning on its masters.
Perhaps we should evaluate people's political sanity not on their absolute position on some political axis, but on their open or narrow mindedness. A political position becomes pernicious fantasy, no matter where it is on your favorite philosophical axis, when it willfully ignores the probable outcomes of the actions it advocates.
For example, other people with me on the left favored single payer health insurance or even a socialized medical system during the recent debates on health insurance reform. While I am philosophically well disposed to these things, I did not favor them at that time. I thought if they were enacted that existing businesses would immediately collapse, and that working public replacements could not be conjured into existence quickly enough to take their place. Now I realize many who prefer socialized medicine or single payer (not the same things at all by the way) might disagree with that assessment. They may even be right. But that's not the point I'm trying to make. I moderated my position based on a critical examination of the likely outcomes of my *ideal* solution. That examination might be faulty, but I did not twist my evaluation of the facts in order to justify my a priori position.
It's tricky to evaluate the political sanity of a figure like Stallman. He is very, very bright,and bright people have a way of finding credible sounding rationalizations for really ill considered opinions. That said, I think that Stallman's positions on the viability of free software sound a lot more credible today than they did twenty years ago. True, free software projects haven't produced viable competition in a number of important niches; but after two decades of experience with free software success, it isn't so hard to believe that a free software ecosystem could meet all the software needs of an individual or enterprise.
Well -- this article really poses something that is really more like a thought experiment than a recommendation.
Case 1: Imagine you have *no* firewall software on desktops and laptops. How would you secure your network?
Case 2: Imagine all your desktop and laptop machines have software firewalls that you completely trust to be perfect. What things would you do differently than in Case 1?
Case 3: All our desktops and laptops have firewall software on them that you don't consider perfect, but is nonetheless pretty good. You have a lot of stuff on your plate and not quite enough time and budget to do everything you ought to. What things from Case 1 do you stop doing?
Case 4: Like Case 3, but your most clueless coworker is making the decision.
Case 5: Like Case 4, but prolonged indefinitely.
I think this guy is making the case that desktop firewalls are harmful in that they induce network administrators to put their trust in them. Naturally, if you are Mr. Anal Retentive NetAdmin (as you ought to be), this doesn't apply to you.
Could you explain that again, but using a car analogy?
sometimes making "simple" solutions actually work is more complicated than the "complex" alternative. As an engineer you run into this all the time, the manager who's so enamored of his brilliance he can't see the flaws in his idea.
This guy is talking about creating artificial peat bogs. It actually *is* an intriguing idea, but I don't see it as "simple". It certainly isn't an "alternative" to government subsidies or regulation. Somebody is going to have to pay the farmers to do this, and to buy the land and transport the waste there, and to deal with the effects of removing so much biomass from th cropland.
Things like grassahol subsidies are supposed to incent the development of new technologies. If those technologies ever make grassahol cheaper than oil, then the subsidies will have bootstrapped a new private grassahol market. That's a big "if", but so is research in energy technologies like fusion. What is problematic is that the farm lobby distorts the program, just as it would a farm waste sequestration program. So it's not a politically simpler solution.
$100 million dollars to scrub 1.5% of the carbon out of the atmosphere sounds like a huge amount of money, until you consider this. A single F22 Raptor cost half again as much (150 million), and we've managed to purchase 166 of those. Most people would admit that is a lot of money, but there are still people that think buying a few more would be a good investment. When you're talking about an entire national economy like the US, 100 million to accomplish something important isn't that much.
Now consider the damage figure for Hurricane Katrina, which stands at 81 billion. Now you can't say that any hurricane was *caused* by global warming, but severe hurricanes are more *frequent* under an AGW scenario. If the frequency of such hurricanes increases, a few billion dollars to dial that down wouldn't be that much money, much less 100 million.
Wow.
OK, what will you build this "simple" "seawall" out of? Riprap? Nope. These are coral atolls we're talking about. Very *remote* coral atolls. You'd have quarry enough granite to wrap 33 islands somewhere then transport it to the middle of the Pacific Ocean. Not cheap. Not easy. Not simple.
OK, what about concrete? Well, maybe they can use coral for aggregate, but they don't have limestone, so they've got to import cement. Any idea how much it would cost to import enough Portland cement to build a wall around 33 islands to the required height? And the steel? You're going to have a *huge* and *very expensive* engineering project just transporting and staging the materials. OK, so a concrete wall around the county is not simple either.
But of course, what you're really talking about is creating a system of dikes to transform the country into 33 empolderments. What are dikes made out of? Well, they are primarily earthworks stabilized with vegetation. What do these people have to work with? Coral sand. How do you propose to make it stay where it's put?
It's easy to sit in your armchair and call the people who have to solve this problem *for real* "childish" and call their problems "whining". In fact, the less you know the easier it is.
Furthermore, you obviously haven't read the article. The problem with the marine sanctuary is that fishing permits (for foreign tuna fleets -- a detail the article should mention) are a major part of *government* revenue. Kiribati thinks that it deserves some revenue for preserving this huge hunk of the Earth for everyone else, and they're right.
True. On the other hand it has the least orthogonal trigger language of any commonly used RDBMS, and has an absolutely dreadful parser that can't be trusted with unusual cases (like bound variables in subqueries) and is unusually restrictive about the use of column aliases. These are issues that don't matter to most MS-SQL users, who really like the integration with the MS tool stack.
On the other hand, the Oracle peculiarity about '' and NULL although aesthetically ugly, has almost no practical importance.
"there are always ways that all parties can resolve their legitmate conflicts to the greater benefit of everyone," was thrown out the door and stomped on by Hitler following the partition of Czechoslovakia.
That would only be demonstrated by your example if the outcome of WW2 was the best possible outcome for Germany. I'd argue that Hitler's strategy was less than optimal for Germany, since it led to regime change, partition, and territorial loss. That was a less than perfect outcome not only for the nation, but for the regime, so it actually supports the hypothesis that warmongering is irrational.
The interesting question is this: given that Nazi Germany pursued a cynically self-interested policy with respect to rearmament and territorial acquisition through the thirties, why didn't they continue to pursue their self interest when it required moderating their aggressive tendencies?
I'd argue that it's a bit like the entrepreneur who starts a promising business but somehow can't make the transition from start-up to viable business. The skills and culture needed to seize power and begin the acquisition of an empire are different from those needed to run an empire. By 1939 they were locked into behavior that had succeeded brilliantly for them over the past decade at a time when they would have been better served by slowing down. If Hitler had given Chamberlain what Chamberlain promised his people -- "peace in our time" -- if he'd put Poland on the back burner for a few years, he *might* have had a chance to stay in power for decades longer. If he played his cards skillfully, he might have seen a pro-Nazi government in Britain and an anti-communist government in the US that would have welcomed his challenging the expansion of a Soviet sphere of influence in Eastern Europe.
I believe that kind of regime tends to end up riding the tiger. Their need for a steady stream of scapegoats and conquests makes them aggressive and successful in the short term, like cancer.
So it is both true that there is always a better way to settle differences than war, but it is also true that you can't count on people pursuing their rational self interests.
Tell me one single task that Outlook can do that the combo Lotus Notes/Domino can not do.
Get the job done without pissing off its users. Granted, that's a minor point.