The philosopher John Rawls proposes that we consider a set of rules as if we were about to take part in a game. The twist is that we don't know which side we are going to play, in this case side the insurgent sunning himself on his rooftop with his wife and children inside, or the drone pilot who is about to blow them to kingdom come. This is sometimes called "the veil of ignorance".
If you are willing to play under these rules without knowing which part you will be assigned then *you* consider these rules fair.
Right. This is a university. Find a tenured professor with an ax to grind who sits on the right committees, then set him loose on this like any sensible person would.
Except there is this thing called "implied warranty". Depending on your state, manufacturers cannot disclaim warranty for what a reasonable person would expect something he bought to do. You can't write a warranty for your hammer that says you can't drive nails with it -- that provision would be void.
Dropping your phone in the toilet is not something covered by implied warranty. Taking your phone from outdoors to indoors at normal humidity for each environment is something reasonable people expect to be able to do. And in fact this does *not* damage the phone, only the sensor. Denying warranty coverage because of a sensor that generates false positives from normal use is not reasonable.
The earphone jack is a bad place to put this sensor. It's convenient because the techs can see it without cracking the case, but topologically its part of the *exterior* of the phone.
[A Microsoft representative comes to a System Admin's place of work for a little meeting.]
MR: Thanks for making time to meet with me.
SA: No problem. So what's this all about?
MR: I don't know how to say this, but it seems that you... well you aren't entirely in control of your systems.
SA: You mean you're selling a new management tool?
MR: No, no nothing like that. It's just that there are certain things... Well let's say there are things about your system that you don't know that you really ought to be aware of.
SA: Oh, I see. You mean like undocumented registry settings, or DLLS or stuff like that.
MR: Well, sure. Technically you *could* describe it that way. It's only....
SA: Only what? How would *you* describe it.
MR: *sigh*. OK. Some Chinese hacker working for the Russian mob has been using you as his bitch.
Well, sure. unused RAM is, in a sense "wasted". But it's not so simple as that. Sometimes having RAM in reserve in case *I* need it is a use.
Microsoft's *style* seems to be one of aggressive optimization. It's not a big deal to grab RAM, unless Windows has trouble giving up that RAM fast enough to serve *my* needs. That's why I stopped using Vista for development. I had 4GB of RAM and Vista would grab almost 100% of the 3GB available. Then I'd launch a process that needed 1GB of RAM and Vista would take an enormous performance hit.
What was worse it got worse over time. That was a clue. When I looked at the paging file, it had an unbelievable number of fragments, tens of thousands of fragments. Judging from the results, I'd say that when Vista realized it needed to give me so much memory it couldn't figure out how to release the RAM it was holding fast enough, and so it just started writing pages where ever the disk head happened to be. Dropping the page file altogether stopped the random system hiccups, but still didn't help the fact that Vista was crap at allocating large blocks of memory because it's not good enough at releasing the memory it grabs. So basically I just gave up on Windows and stuck with Linux for everything.
So you can't say *necessarily* that Windows grabbing unused RAM is innocuous. that depends on Windows being able to reallocate that RAM to *my* uses when I demand it.
The behavior I've observed is symptomatic of sloppy engineering. "This thing isn't fast enough, let's make every optimization we can think of." The first reflex of a good coder when confronted with code that is not fast enough is to make the code *better*, not *faster*. "Faster" often leads to unexpected costs, while "better" often leads to unexpected benefits.
Sure you can. You use digital signatures. The problem is you need somebody with above room temperature IQ to administer the key signing infrastructure, and those folks are in short supply. That's why it's never caught on.
Not to dispute your observations, which I agree with but you can say *anything* is an illusion if you choose a sufficiently constrained definition of it.
What you are talking about is the "all or nothing" model of security, where security is regarded as a property a system either has or does not have. Any system that a determined adversary can undermine is "not secure", and of course a determined adversary (one willing and able to engage in black bag jobs and human intelligence operations) is capable of penetrating any system.
Let me propose an alternate view of security: it is a continuous dimension on which systems can be placed for purposes of evaluating them for a particular kind of use. The question then is not "is this system secure?" but rather "is this system sufficiently secure to serve this purpose in this situation?"
It's quite reasonable to ask whether a network of computers connected through the Internet and running certain services and software are sufficiently secure to run the nation's power grid. The important thing isn't the answer you get, but the *process* you go through to obtain that answer. A sound process for answering that question should result in a deeper understanding of the system's vulnerabilities. "Is this system secure?" is too vague, and is apt to lead to wishful thinking.
That's because you need more than 15 years to get statistically significant figures.
I think you said that poorly.
There is no sharp cut-off as to the interval size you need to be able to achieve significance. Furthermore, the *meaning* of significance is confusing when we talk about a single interval's importance in falsifying a hypothesis about the distribution of a random variable (global average temperatures)
Imagine we play a game of coin toss with a coin I provide. I take heads, you take tails. We play four rounds, and heads comes up every time. You, naturally, suspect I'm cheating. Then our friend Dr. Jones points out that four sequential heads does not meet the 95% standard for statistical significance. You need no more greater probability for an event than p(1/20), but we only have a p(1/16) event here.
What the deniers are doing with Dr. Jones remarks is like saying, "Four heads in a row is not a statistically significant result, which PROVES the coin is fair."
In any case, *random sampling* is integral to the very notion of statistical significance. In a sequence of trials of a random variable, you can *always* choose an interval that makes the point you want to make: increase/no change/decrease. And technically, your interval *will* be significantly increasing or decreasing as you like.
So basically significance or non-significance of any single sample of a random sequence doesn't prove or disprove anything, if the sample is small and the chooser gets to pick the size of the interval.
Well, I think the "troll" moderation is pretty unfair, but I don't agree with your points, which seem to envision a program to solve all our problems by building lots and lots of nukes of the same design we were using forty years ago. I agree that's a really bad idea.
If you look forward to America's future energy needs and supplies, we're going to have a tough slog as global oil demand rises and supplies fall, but it's not going to be like shutting off the spigot. Keeping our head above the water is going to be a matter of margins. Diversifying our energy sources will protect us from fuel dependency of any sort, and buy us time to develop the energy technologies (both production and conservation) we'll need to continue to grow.
An Apollo style crash program to build hundreds of nukes would be a really bad idea, that might saddle us with a lot of problematic, outdated plants. If there are serious problems with the program or the technology it will be politically impossible to change anything due to the sunk costs and political investment.
When your "peers" appear to have been actively engaged in hiding their data from public scrutiny, actively engaged in quashing any dissenting papers from getting published (including threats to publishers),...
Nice try. You don't really know what "peer review" means, do you?
Scientists don't talk about their "peers". They talk about their "reviewers". Often in language that is not fit for public consumption.
A better term than "peer review" would be "competitor review". It gets ugly. There is occasional misbehavior of course, but its often just plain rough and feelings get hurt.
As for science being "groupthink", you're halfway there. It's groupthink with negative feedback which alters the group consensus when it strays too far from the facts. The feedback mechanism is peer review.
You really don't understand how science works, do you?
Scientists *want* to prove each other wrong. They *want* to see the status quo broken. That's why they hate seeing a half-assed job of it being done. Nobody would be more thrilled than a physicist if a working perpetual motion machine could be created, or less thrilled with the job of having to sort through all such proposed devices to find ones that might work.
The scientific consensus can be changed, but you've got to start small. You can't waltz in the party at the last minute and tell everyone that everything they've learned in the last hundred years is wrong. You've got to start with individual cases and work your way up. Papers beating up on previously published results happen all the time -- "even" in climate science.
As to this victimization fantasy of the monied interests backing AGW... that's almost too weird to respond to. The number one funder of research in this country is the US government, which would *love it* if scientists discovered that burning all the coal we can get our hands on was the best thing for the environment.
Easy. Show that the record of *global average* temperature over the last century that has been reported in the literature has systematic errors that account for most of the warming.
It has been attempted repeatedly in the published literature. There hasn't exactly been a cover up, it's just that the warming issue per se issue was largely settled in the literature over twenty years ago.
Falsification doesn't mean having to refute any straw man caricature of a hypothesis anyone can dream up. Climate change does not require average temperature at every locality to increase, for seasonal variance at locales to remain constant as average temps there increase, or for there never to be cold years at a locality or for temps to go up every single year. It certainly doesn't preclude unusually large snowstorms on the eastern seaboard.
In any case anyone who's lived in a cold, snowy climate knows snow is weakly correlated at best to temperature. It has to be cold enough to freeze water at high altitudes, then you need plenty of moisture which near the coast is correlated with higher ocean temperatures. It's not anything new or mysterious.
By that theory you should hate beef too, because it takes a lot more than a pound of corn to create a pound of beef. Furthermore the varieties of corn sold for feed are not normally sold for human consumption.
So by buying beef you are diverting corn production from human food into animal feed, which reduces the net food available.
In any case, environmentalists aren't the ones behind corn based ethanol. It's agribusiness.
Seriously, any kind of major underwater operation by a government has to be suspected as a cover story for developing, testing or *using* some kind of spook gear.
In organizations I've worked in, I've tried to spread this idea: delegation does not divide responsibility, it multiplies it. If I delegate a task to a subordinate, he is fully responsible for completing it, but I am no less responsible.
Blaming my subordinate for his failure in no measure exonerates me. Taking the blame myself in mo measure exonerates him.
Responsibility (or blame) is not a finite resource. We don't have to conserve blame. We can blame politicians for policies that increase recidivism without blaming criminals any less for their crimes.
Most minor UI faults aren't rally that important. It's how it all *works*.
Now my take on iTunes is that it shows how Apple has succumbed to the problem that doomed Windows to be an inferior UI. Windows isn't about catering to the user. It throws enough bones to the user to put up a credible looking facade of being user friendly. But Windows is really about catering to people who are chokepoints in the buying process: vendors and people who buy and manage systems for other people. But even Microsoft doesn't have enough money to be all things to all people. Some stakeholders have to be more equal than others.
For years Apple survived on publishing and purchases by individuals and very small businesses where user satisfaction was a critical factor. It had no OEMs to appease, nor any chance in the IT game.
When Jobs returned, he did a lot for Apple's business strategy, but the Apple UI purist was over. Apple now has its own third party masters: content owners.
When I use iTunes wth the iTunes store, it's simply wonderful for doing things which are very profitable for Apple's mass market content partners. It's not so great if I want to find things that aren't the latest episode of some popular TV show. It's glitzy, but that doesn't really matter, because it's a *tool*, one that is really on the sidelines of what users want, but also a place where users can be steered to Apple favored content.
So? All Microsoft has to do is to take out its checkbook and create a partnership with a company like Teleatlas.
What would be really tough is getting the degree of street view coverage Google has.
On top of that, there is scaling this kind of technology so that it works with the volume of data and users that Google has. In many ways, Google is a data storage and distribution company. They've got something like synergy, only it's much more concrete: an infrastructure that is useful for a wide variety of applications serving data to lots and lots of people. It will be interesting to see if Microsoft can grow their apps to the scale of use Google has, and whether they can do it on the software they sell to others.
Slashdot Mirror
The philosopher John Rawls proposes that we consider a set of rules as if we were about to take part in a game. The twist is that we don't know which side we are going to play, in this case side the insurgent sunning himself on his rooftop with his wife and children inside, or the drone pilot who is about to blow them to kingdom come. This is sometimes called "the veil of ignorance".
If you are willing to play under these rules without knowing which part you will be assigned then *you* consider these rules fair.
Are you for real?
Right. This is a university. Find a tenured professor with an ax to grind who sits on the right committees, then set him loose on this like any sensible person would.
You don't understand this grant process, do you?
You always leave something for the next proposal.
Why not *no* legs.. Snakes seem to move along fine.
Except there is this thing called "implied warranty". Depending on your state, manufacturers cannot disclaim warranty for what a reasonable person would expect something he bought to do. You can't write a warranty for your hammer that says you can't drive nails with it -- that provision would be void.
Dropping your phone in the toilet is not something covered by implied warranty. Taking your phone from outdoors to indoors at normal humidity for each environment is something reasonable people expect to be able to do. And in fact this does *not* damage the phone, only the sensor. Denying warranty coverage because of a sensor that generates false positives from normal use is not reasonable.
The earphone jack is a bad place to put this sensor. It's convenient because the techs can see it without cracking the case, but topologically its part of the *exterior* of the phone.
Wikipedia contains the same "misinformation"
And you point is ...?
[A Microsoft representative comes to a System Admin's place of work for a little meeting.]
MR: Thanks for making time to meet with me.
SA: No problem. So what's this all about?
MR: I don't know how to say this, but it seems that you... well you aren't entirely in control of your systems.
SA: You mean you're selling a new management tool?
MR: No, no nothing like that. It's just that there are certain things... Well let's say there are things about your system that you don't know that you really ought to be aware of.
SA: Oh, I see. You mean like undocumented registry settings, or DLLS or stuff like that.
MR: Well, sure. Technically you *could* describe it that way. It's only....
SA: Only what? How would *you* describe it.
MR: *sigh*. OK. Some Chinese hacker working for the Russian mob has been using you as his bitch.
http://en.wikipedia.org/wiki/Knol
Well, sure. unused RAM is, in a sense "wasted". But it's not so simple as that. Sometimes having RAM in reserve in case *I* need it is a use.
Microsoft's *style* seems to be one of aggressive optimization. It's not a big deal to grab RAM, unless Windows has trouble giving up that RAM fast enough to serve *my* needs. That's why I stopped using Vista for development. I had 4GB of RAM and Vista would grab almost 100% of the 3GB available. Then I'd launch a process that needed 1GB of RAM and Vista would take an enormous performance hit.
What was worse it got worse over time. That was a clue. When I looked at the paging file, it had an unbelievable number of fragments, tens of thousands of fragments. Judging from the results, I'd say that when Vista realized it needed to give me so much memory it couldn't figure out how to release the RAM it was holding fast enough, and so it just started writing pages where ever the disk head happened to be. Dropping the page file altogether stopped the random system hiccups, but still didn't help the fact that Vista was crap at allocating large blocks of memory because it's not good enough at releasing the memory it grabs. So basically I just gave up on Windows and stuck with Linux for everything.
So you can't say *necessarily* that Windows grabbing unused RAM is innocuous. that depends on Windows being able to reallocate that RAM to *my* uses when I demand it.
The behavior I've observed is symptomatic of sloppy engineering. "This thing isn't fast enough, let's make every optimization we can think of." The first reflex of a good coder when confronted with code that is not fast enough is to make the code *better*, not *faster*. "Faster" often leads to unexpected costs, while "better" often leads to unexpected benefits.
Sure you can. You use digital signatures. The problem is you need somebody with above room temperature IQ to administer the key signing infrastructure, and those folks are in short supply. That's why it's never caught on.
Not to dispute your observations, which I agree with but you can say *anything* is an illusion if you choose a sufficiently constrained definition of it.
What you are talking about is the "all or nothing" model of security, where security is regarded as a property a system either has or does not have. Any system that a determined adversary can undermine is "not secure", and of course a determined adversary (one willing and able to engage in black bag jobs and human intelligence operations) is capable of penetrating any system.
Let me propose an alternate view of security: it is a continuous dimension on which systems can be placed for purposes of evaluating them for a particular kind of use. The question then is not "is this system secure?" but rather "is this system sufficiently secure to serve this purpose in this situation?"
It's quite reasonable to ask whether a network of computers connected through the Internet and running certain services and software are sufficiently secure to run the nation's power grid. The important thing isn't the answer you get, but the *process* you go through to obtain that answer. A sound process for answering that question should result in a deeper understanding of the system's vulnerabilities. "Is this system secure?" is too vague, and is apt to lead to wishful thinking.
That's because you need more than 15 years to get statistically significant figures.
I think you said that poorly.
There is no sharp cut-off as to the interval size you need to be able to achieve significance. Furthermore, the *meaning* of significance is confusing when we talk about a single interval's importance in falsifying a hypothesis about the distribution of a random variable (global average temperatures)
Imagine we play a game of coin toss with a coin I provide. I take heads, you take tails. We play four rounds, and heads comes up every time. You, naturally, suspect I'm cheating. Then our friend Dr. Jones points out that four sequential heads does not meet the 95% standard for statistical significance. You need no more greater probability for an event than p(1/20), but we only have a p(1/16) event here.
What the deniers are doing with Dr. Jones remarks is like saying, "Four heads in a row is not a statistically significant result, which PROVES the coin is fair."
In any case, *random sampling* is integral to the very notion of statistical significance. In a sequence of trials of a random variable, you can *always* choose an interval that makes the point you want to make: increase/no change/decrease. And technically, your interval *will* be significantly increasing or decreasing as you like.
So basically significance or non-significance of any single sample of a random sequence doesn't prove or disprove anything, if the sample is small and the chooser gets to pick the size of the interval.
Well, I think the "troll" moderation is pretty unfair, but I don't agree with your points, which seem to envision a program to solve all our problems by building lots and lots of nukes of the same design we were using forty years ago. I agree that's a really bad idea.
If you look forward to America's future energy needs and supplies, we're going to have a tough slog as global oil demand rises and supplies fall, but it's not going to be like shutting off the spigot. Keeping our head above the water is going to be a matter of margins. Diversifying our energy sources will protect us from fuel dependency of any sort, and buy us time to develop the energy technologies (both production and conservation) we'll need to continue to grow.
An Apollo style crash program to build hundreds of nukes would be a really bad idea, that might saddle us with a lot of problematic, outdated plants. If there are serious problems with the program or the technology it will be politically impossible to change anything due to the sunk costs and political investment.
Whereas ignorant knuckleheads just instinctively know the truth.
When your "peers" appear to have been actively engaged in hiding their data from public scrutiny, actively engaged in quashing any dissenting papers from getting published (including threats to publishers),...
Nice try. You don't really know what "peer review" means, do you?
Scientists don't talk about their "peers". They talk about their "reviewers". Often in language that is not fit for public consumption.
A better term than "peer review" would be "competitor review". It gets ugly. There is occasional misbehavior of course, but its often just plain rough and feelings get hurt.
As for science being "groupthink", you're halfway there. It's groupthink with negative feedback which alters the group consensus when it strays too far from the facts. The feedback mechanism is peer review.
You really don't understand how science works, do you?
Scientists *want* to prove each other wrong. They *want* to see the status quo broken. That's why they hate seeing a half-assed job of it being done. Nobody would be more thrilled than a physicist if a working perpetual motion machine could be created, or less thrilled with the job of having to sort through all such proposed devices to find ones that might work.
The scientific consensus can be changed, but you've got to start small. You can't waltz in the party at the last minute and tell everyone that everything they've learned in the last hundred years is wrong. You've got to start with individual cases and work your way up. Papers beating up on previously published results happen all the time -- "even" in climate science.
As to this victimization fantasy of the monied interests backing AGW... that's almost too weird to respond to. The number one funder of research in this country is the US government, which would *love it* if scientists discovered that burning all the coal we can get our hands on was the best thing for the environment.
Easy. Show that the record of *global average* temperature over the last century that has been reported in the literature has systematic errors that account for most of the warming.
It has been attempted repeatedly in the published literature. There hasn't exactly been a cover up, it's just that the warming issue per se issue was largely settled in the literature over twenty years ago.
Falsification doesn't mean having to refute any straw man caricature of a hypothesis anyone can dream up. Climate change does not require average temperature at every locality to increase, for seasonal variance at locales to remain constant as average temps there increase, or for there never to be cold years at a locality or for temps to go up every single year. It certainly doesn't preclude unusually large snowstorms on the eastern seaboard.
In any case anyone who's lived in a cold, snowy climate knows snow is weakly correlated at best to temperature. It has to be cold enough to freeze water at high altitudes, then you need plenty of moisture which near the coast is correlated with higher ocean temperatures. It's not anything new or mysterious.
By that theory you should hate beef too, because it takes a lot more than a pound of corn to create a pound of beef. Furthermore the varieties of corn sold for feed are not normally sold for human consumption.
So by buying beef you are diverting corn production from human food into animal feed, which reduces the net food available.
In any case, environmentalists aren't the ones behind corn based ethanol. It's agribusiness.
while they're at it?
Seriously, any kind of major underwater operation by a government has to be suspected as a cover story for developing, testing or *using* some kind of spook gear.
In organizations I've worked in, I've tried to spread this idea: delegation does not divide responsibility, it multiplies it. If I delegate a task to a subordinate, he is fully responsible for completing it, but I am no less responsible.
Blaming my subordinate for his failure in no measure exonerates me. Taking the blame myself in mo measure exonerates him.
Responsibility (or blame) is not a finite resource. We don't have to conserve blame. We can blame politicians for policies that increase recidivism without blaming criminals any less for their crimes.
Just like Blackwater renamed itself to "Xe" after its employees were implicated in (among other things) massacring civilians.
Maybe "X" is the scarlet letter of corporate misconduct.
Most minor UI faults aren't rally that important. It's how it all *works*.
Now my take on iTunes is that it shows how Apple has succumbed to the problem that doomed Windows to be an inferior UI. Windows isn't about catering to the user. It throws enough bones to the user to put up a credible looking facade of being user friendly. But Windows is really about catering to people who are chokepoints in the buying process: vendors and people who buy and manage systems for other people. But even Microsoft doesn't have enough money to be all things to all people. Some stakeholders have to be more equal than others.
For years Apple survived on publishing and purchases by individuals and very small businesses where user satisfaction was a critical factor. It had no OEMs to appease, nor any chance in the IT game.
When Jobs returned, he did a lot for Apple's business strategy, but the Apple UI purist was over. Apple now has its own third party masters: content owners.
When I use iTunes wth the iTunes store, it's simply wonderful for doing things which are very profitable for Apple's mass market content partners. It's not so great if I want to find things that aren't the latest episode of some popular TV show. It's glitzy, but that doesn't really matter, because it's a *tool*, one that is really on the sidelines of what users want, but also a place where users can be steered to Apple favored content.
thought biomedical researchers were "playing God".
And which one of you wanted the clean glass?
So? All Microsoft has to do is to take out its checkbook and create a partnership with a company like Teleatlas.
What would be really tough is getting the degree of street view coverage Google has.
On top of that, there is scaling this kind of technology so that it works with the volume of data and users that Google has. In many ways, Google is a data storage and distribution company. They've got something like synergy, only it's much more concrete: an infrastructure that is useful for a wide variety of applications serving data to lots and lots of people. It will be interesting to see if Microsoft can grow their apps to the scale of use Google has, and whether they can do it on the software they sell to others.