Do you all really think that Mastercard hasn't already thought of this and solved it???
Sorry, are we talking about the same people whose previous major innovation in security was to print an extra three- or four-digit PIN directly on the card? And the people who have rolled out pretty much nothing while identity theft went from a minor problem to a giant national clusterfuck?
Yes, I expect them to do something retarded. From the announcement, they're talking about how these super-neat cards will get them revenue growth through added convenience, not how they'll increase security. I think their main criterion is that this should suck no more than magnetic stripes. And for a few years, until theives make fancy new hardware, it probably will meet that.
Of course, they could actually do something pretty smart, in which case I'll be pleasantly surprised.
A simple solution would be to have an RSA key + engine on the card, so that the 'scanner' issues a challenge to the card and if the card can supply the decrypted string then it passes.
That protects you against capture and playback attacks, but not man-in-the-middle attacks.
What's the second factor? What people typically mean by two-facto auth is a physical token plus a PIN or password, but that doesn't square with the marketing literature or their "easier than cash" claims.
You can't steal the card number because the card doesn't transmit the card number.
Is it some sort of challenge/response thing? If so, where does the challenge come from? It seems like the challenge would have to come from the clearing firm, yes? But wouldn't that require a hot connection with good response time to get the exchange done during a wave?
Isn't the proper column to look at RSS? The docs for ps describe that as "the non-swapped physical memory that a task has used (in kiloBytes)." VSZ, on the other hand, is "virtual memory usage of entire process (vm_lib + vm_exe + vm_data + vm_stack)". If I add ups the VSZ column for everything I have running, it's well more than the physical RAM I have, so that doesn't seem useful for comparison.
And really, 112 meg for an IDE (I presume that's Eclipse you're using) doesn't sound bad. My copy of Firefox weighs in at 92 meg RSS, and I just have 2 windows with three tabs each open.
Let's call that "uncommon". The two best Java IDEs, IntelliJ IDEA and the Eclipse project, are both written in Java. They work fine. And as the other fellow pointed out, they aren't the only examples.
I can't run Azureus for more than a few hours without it eating all of my RAM and bringing down my entire system.
Just for another data point I run Azureus under Linux (FC3, JDK 1.5.0_02) for weeks at a time without problem. After 10 days of running, the thing right now weighs in at 187 MB. That seems kinda piggy for what I do with it, but my 1 GB machine is perfectly usable. Azureus reliably checks RSS feeds and downloads stuff automatically.
I wish it used less, but that's an entire $25 of RAM, so I'm not sweating it.
The real problem is that while your competitors were rewriting their products in Java, you were sticking with VB. And now you want to blame Microsoft for that.
Having used VB on exactly one project, I must admit that writing a large app in it suggests masochistic tendencies. But I think it's fair to blame Microsoft for following Microsoft's advice.
The fact that many vendors are basically untrustworthy does not mean that you shouldn't yell at vendors every time the shiv you. The vendors can learn from the pain, and your colleagues learn to fear the vendor when they hear you yelling.
And.NET does not necessarily mean platform/vendor lock, either.
You mean in the weak sense that some vendor could theoretically reimplement it someday? That was true for VB, too. Doesn't seem to be working out so well, does it?
Theoretically, I could add comments to a brace that indicate what type of structure ends at the respective point, but that's a waste of time, and if I modify an algorithm, then the function of those closing braces may change.
Why would you need such a thing?
Reading your comments, my first reaction is that if you don't know which blocks end where, your blocks are too big. Whenever I start wondering which closing brace is which, it's time to do an Extract Method refactoring.
it really helps to declare types on function parameters; it serves as a kind of documentation of the intended use and operation of a function.
It's true. Especially if you don't have named parameters, the typing makes things much clearer.
Another thing I like about static typing is the IDE magic that it makes possible. There are a number of automated refactoring tricks that I'm very fond of that I don't think could be done without static typing.
Of course, as another person pointed out, if you have good unit and acceptance tests, a lot of the save-me-from-myself features of static typing become harder to justify.
As a Java programmer, it is exciting to see these developments in C#, it makes me wonder whether Java is destined to fall behind C# - it sure looks like that is happening....
Yep. Java 1.5 certainly was a big step forward, but it seemed to mainly be catching up, rather than getting ahead. And they seem to have backed off in 1.6; I don't see anything as radical as the templating stuff. I was really hoping for closures, for example. And it certainly doesn't seem as ambitious as C# 3.0.
Do Xtremers have any particular choice of language?
I think the short answer is no.
A lot of the original XP people have a background in Smalltalk, but from the mailing list and talking to people at the conferences, there are few people doing current work in that. Java and C# are probably most common, but that generally seems to be a business decision, not a technical one. I hear a lot of excitement about Python and Ruby, too, both of which Paul Graham recommends as reasonable Lisp alternatives.
Personally, I do most of my work in Java, both for business reasons and because of the good tool support. Automated refactoring tools like IntelliJ's IDEA make a refactoring-driven approach to design much more workable. But I'm hoping that Ruby continues to grow in both library and tool support.
THE THINGS YOU OWN THEY END UP OWNING YOU Just blow it all up.
Neither should you be owned by your fear of stuff.
Which reminds me that the extreme end of being owned by stuff is known as hoarding, and can be a symptom of a medical problem. If you know somebody with a serious too-much-stuff problem, it's worth looking into it. There are a variety of medications that can help.
I used to be of this opinion; up until a few years ago I thought most IDEs were worthless. The one that changed my mind is IntelliJ's IDEA, a Java IDE. It turns the code base into a giant linked hypertext document, and has a lot of other nice things to make life easier.
But the real killer feature is automated refactorings. The most basic is rename; you can safely rename a class or method without disturbing other things that have the same name. But there are many more, and they take a lot of the drudgery out of programming.
I keep hearing that Eclipse is just as good, but last I did a project with Eclipse (about 18 months ago) the UI wasn't nearly as well made. Nothing major, but a lot of minor differences that really added up during extended usage.
A lot of things you mention I don't care much about. But I recommend ridiculous amounts of RAM. Even if you get more than you think you'll need, you'll find a use for it.
My latest giant RAM sink is VMWare. I run a virtual copy of Windows for browser testing, and a couple more for virtual servers. Virtual servers are much better for testing than real ones: when you're done trying something out, you can revert the virtual disk back to a known clean configuration.
It has a nubmber of additional practices. The most important are probably test-driven development, refactoring, regular iterations (ideally, one week) and frequent releases.
Utterly meaningless - No matter how excited about the topic, without a focus on profit - they'll be out of a job soon.
You've consistently confused a focus on profit with a focus on keeping out of the red so that you can keep doing what you love.
To stay in business, a small company need only stay barely in the black. And owner-run businesses can and will do that for much less, as owners happily put up with low salaries and long hours because they're doing something they love. Their focus is on doing what they do and staying in business, rather than beating the market average risk-adjusted return on capital.
The difference in financial terms is relatively small. But the difference in attitude is huge. And it's not just anecdotal. There's a whole body of psychology and management literature about the differing effects of intrinsic and extrinsic rewards on motivation and performance.
Carry on with your theoretical arguments and your handwaving if you like. I've seen this time and time again, in industries from dot-com startups to magazines to nursing homes. A business has to stay in black to survive, but the best ones are focused on innovation and quality first, and profit second.
What a complete circle jerk. If they're still arguing about whether or not there is actually a problem without any idea how we can solve it then why the hell do they expect anyone to take them seriously?
For the same reason people should take weather forecasts seriously. We can't know what will happen, but with, say, Katrina, we knew there was a substantial chance of a problem.
Here we're performing a giant experiment with the only planet we have, and the best guess is that involves substantial risks. What to do about that is not a scientific question; it's a moral and political one. It's not the job of citizens to resolve those. It's the job of us, the citizens.
Regardless of what is causing it I want to know, how bad will it be and what can we do to correct it? These are the topics that scientists never appear to talk about.
And that's a good thing. They don't know. They're honest about the fact that they don't know.
If you ask them in the right way, you can find out their personal opinions on what they think we should do. But unlike most people quoted in the media, most scientists understand the gulf between their professional opinions and their personal ones.
Even if all the magazines on the stands were each owned by an individual publisher, they'd still be mostly interested in selling magazines and making a profit.
That's probably not true. I used to do tech work for magazines, and the small specialist mags were always staffed by people who were really excited about the topic.
Think of it like restaurants. Obviously, every restaurant owner has to sell food and make a profit. But there's a world of difference between a single-location, owner-operated restaurant and a giant chain. The decision-makers in the big chains are insulated from the consequences of their decisions. And the clock-punchers on the spot don't have the same incentive to care.
Or compare it to the web. Look at how much cool stuff has appeared on the web, and look how little of it has come from large, corporate publishers. The best technical reviews don't come from USA Today or Ziff-Davis; instead I go to Tom's Hardware, AnandTech, Ars Technica, and a host of other small publishers who do it because that's what they want to do, and for whom making a profit is secondary.
I have never, ever seen a student running in a non-administrator account on their Windows PC, even though XP supports this feature.
Have you actually tried this?
I got a Windows XP laptop recently because I'm working on a project with a Windows client. For probably four weeks, I tried hard to use the administrator account only for installing stuff.
It really sucked. A number of applications wouldn't work properly unless I had administrator rights. Doing all sorts of typical laptop stuff (e.g., switching the wireless to another SSID) required superpowers. I was logging in and out all the goddamn time. Eventually, I just gave up.
This shocks me; having used various Unix variants for a long time, I'm completely comfortable with the user/root distinction, and would never just log in as root. But Windows doesn't seem to be set up to support that.
Zonealarm asks them if they want to let, say, AIM or Weatherbug access their network connect - and the user freaks out.
A lot of this could be solved with smarter interfaces. I'm using whatever firewall Symantec sells, and it pops up similar warnings and gives advice. Even better would be if it looked up access patterns against a centralized database, so that it just knows that, say, Google Desktop is composed of 3 applications with access patterns X, Y, and Z. Then it could say, "I see you're installing Google Desktop. It will get and transmit information over the Internet. We consider it safe. Should I let it run?" With the right cues, users won't freak out.
One of the points basically comes down to "write perfect code". Well, duh, why didn't I think of that before? Jeez. Patching is bad because your code should have been perfect in the first place? That's the dumbest thing I ever heard.
You completely missed his point.
He's saying that regular patching to fix security holes is a symptom of a flaw in your design process. If you've designed for security from the beginning, your security patches should be few and far between. He even gives examples of programs that don't have this problem: Postfix and Qmail.
The problem he's pointing at isn't the occasional patch; it's the attitude that regular security patching is the only way, or even a good way, to deal with security flaws.
I just cleaned out several boxes of stuff that I've been lugging around for more than a decade. Once I actually sat down and went through it, I realized that I'd never look at 95% of it again. I kept the stuff with sentimental value, and the rest is confetti. What a relief!
Which will be completely irrelevant when push comes to shove. Given a choice between a. living in the dark. or b. not living in the dark, I have few doubts that the majority of us will choose option "b". That means that nuclear power plants will get built, and that the government will indemnify them to whatever degree is required.
Yes, but why is it only nuclear that deserves such massive operational subsidies? Not just in insurance, but in waste disposal, security, and fuel supply? Not to mention R&D subsidies and construction incentives.
It seems to me that if the government is going to get involved, a carbon tax is the most obvious option; then the best tech can duke it out in the marketplace. But if it's going to play favorites, why not favor something that doesn't carry the apparently substantial risk of a Katrina-sized evacuation and cleanup effort?
Conventional power generation isn't necessarily immune to liability issues either. I mean, a whole lot of people are dying of ailments directly related to the use of coal-fired powerplants. The way I figure it, if they can sue a gun maker because a thug shot somebody, they can go after power plant operators for causing cancer and other ailments.
There are two main obstacles that I know of. One is that the causal chain is impossible to establish. I can easily prove who made the gun that killed my spouse and demonstrate that they didn't exercise due care in designing, advertising, and selling them. But I can't prove who made the particular soot particles that caused her asthma or lung cancer.
The other is that power producers are extensively regulated, emitting waste at levels the government has determined to be safe.
In the case of another Chernobyl, though, it's pretty obvious who is at fault, and a disaster like that is probably above EPA limits even these days.
But many people are deathly afraid of the idea with good reason: when nuke plants fail they fail really, really badly.
Interestingly, a while back The Economist did a long article on the pros and cons of using more nuclear power in the US. They concluded that it wasn't economical without huge government subsidies, which come in the form of free insurance. If nuke plants had to pay the full cost of the insurance to cover catastophes, they would cost well more than fossil fuels do.
They're cheap enough that someone could buy a thousand of them and distribute them to everyone in the Astrodome.
Indeed, they had already arranged a donation of 10,000 radios, so that there would be plenty for all.
Very true, but that doesn't diminish the fact that it WILL help people.
Indeed. A psychiatrist was mentioning somewhere that one of the worst things for people who have been through disasters is to sit around with nothing to do and nothing but the disaster to think about. For people developing PTSD, it can intensify and lengthen their problems.
And there's a lot to be said for community-building and morale. Heck, just the music alone would help. Imagine you're on a long road trip and the radio breaks. How sad would you be? And now imagine somebody else is driving, you don't know where you're going, and you're not sure when you'll get there.
Slashdot Mirror
Do you all really think that Mastercard hasn't already thought of this and solved it???
Sorry, are we talking about the same people whose previous major innovation in security was to print an extra three- or four-digit PIN directly on the card? And the people who have rolled out pretty much nothing while identity theft went from a minor problem to a giant national clusterfuck?
Yes, I expect them to do something retarded. From the announcement, they're talking about how these super-neat cards will get them revenue growth through added convenience, not how they'll increase security. I think their main criterion is that this should suck no more than magnetic stripes. And for a few years, until theives make fancy new hardware, it probably will meet that.
Of course, they could actually do something pretty smart, in which case I'll be pleasantly surprised.
A simple solution would be to have an RSA key + engine on the card, so that the 'scanner' issues a challenge to the card and if the card can supply the decrypted string then it passes.
That protects you against capture and playback attacks, but not man-in-the-middle attacks.
ISO/IEC 14443 has two-factor authentication.
What's the second factor? What people typically mean by two-facto auth is a physical token plus a PIN or password, but that doesn't square with the marketing literature or their "easier than cash" claims.
You can't steal the card number because the card doesn't transmit the card number.
Is it some sort of challenge/response thing? If so, where does the challenge come from? It seems like the challenge would have to come from the clearing firm, yes? But wouldn't that require a hot connection with good response time to get the exchange done during a wave?
the Java Virtual Machine eats up some 320MB
Isn't the proper column to look at RSS? The docs for ps describe that as "the non-swapped physical memory that a task has used (in kiloBytes)." VSZ, on the other hand, is "virtual memory usage of entire process (vm_lib + vm_exe + vm_data + vm_stack)". If I add ups the VSZ column for everything I have running, it's well more than the physical RAM I have, so that doesn't seem useful for comparison.
And really, 112 meg for an IDE (I presume that's Eclipse you're using) doesn't sound bad. My copy of Firefox weighs in at 92 meg RSS, and I just have 2 windows with three tabs each open.
desktop java is completely absent
Let's call that "uncommon". The two best Java IDEs, IntelliJ IDEA and the Eclipse project, are both written in Java. They work fine. And as the other fellow pointed out, they aren't the only examples.
I can't run Azureus for more than a few hours without it eating all of my RAM and bringing down my entire system.
Just for another data point I run Azureus under Linux (FC3, JDK 1.5.0_02) for weeks at a time without problem. After 10 days of running, the thing right now weighs in at 187 MB. That seems kinda piggy for what I do with it, but my 1 GB machine is perfectly usable. Azureus reliably checks RSS feeds and downloads stuff automatically.
I wish it used less, but that's an entire $25 of RAM, so I'm not sweating it.
I can't believe this got rated insightful.
.NET does not necessarily mean platform/vendor lock, either.
The real problem is that while your competitors were rewriting their products in Java, you were sticking with VB. And now you want to blame Microsoft for that.
Having used VB on exactly one project, I must admit that writing a large app in it suggests masochistic tendencies. But I think it's fair to blame Microsoft for following Microsoft's advice.
The fact that many vendors are basically untrustworthy does not mean that you shouldn't yell at vendors every time the shiv you. The vendors can learn from the pain, and your colleagues learn to fear the vendor when they hear you yelling.
And
You mean in the weak sense that some vendor could theoretically reimplement it someday? That was true for VB, too. Doesn't seem to be working out so well, does it?
Theoretically, I could add comments to a brace that indicate what type of structure ends at the respective point, but that's a waste of time, and if I modify an algorithm, then the function of those closing braces may change.
Why would you need such a thing?
Reading your comments, my first reaction is that if you don't know which blocks end where, your blocks are too big. Whenever I start wondering which closing brace is which, it's time to do an Extract Method refactoring.
it really helps to declare types on function parameters; it serves as a kind of documentation of the intended use and operation of a function.
It's true. Especially if you don't have named parameters, the typing makes things much clearer.
Another thing I like about static typing is the IDE magic that it makes possible. There are a number of automated refactoring tricks that I'm very fond of that I don't think could be done without static typing.
Of course, as another person pointed out, if you have good unit and acceptance tests, a lot of the save-me-from-myself features of static typing become harder to justify.
As a Java programmer, it is exciting to see these developments in C#, it makes me wonder whether Java is destined to fall behind C# - it sure looks like that is happening....
Yep. Java 1.5 certainly was a big step forward, but it seemed to mainly be catching up, rather than getting ahead. And they seem to have backed off in 1.6; I don't see anything as radical as the templating stuff. I was really hoping for closures, for example. And it certainly doesn't seem as ambitious as C# 3.0.
Do Xtremers have any particular choice of language?
I think the short answer is no.
A lot of the original XP people have a background in Smalltalk, but from the mailing list and talking to people at the conferences, there are few people doing current work in that. Java and C# are probably most common, but that generally seems to be a business decision, not a technical one. I hear a lot of excitement about Python and Ruby, too, both of which Paul Graham recommends as reasonable Lisp alternatives.
Personally, I do most of my work in Java, both for business reasons and because of the good tool support. Automated refactoring tools like IntelliJ's IDEA make a refactoring-driven approach to design much more workable. But I'm hoping that Ruby continues to grow in both library and tool support.
THE THINGS YOU OWN THEY END UP OWNING YOU Just blow it all up.
Neither should you be owned by your fear of stuff.
Which reminds me that the extreme end of being owned by stuff is known as hoarding, and can be a symptom of a medical problem. If you know somebody with a serious too-much-stuff problem, it's worth looking into it. There are a variety of medications that can help.
I used to be of this opinion; up until a few years ago I thought most IDEs were worthless. The one that changed my mind is IntelliJ's IDEA, a Java IDE. It turns the code base into a giant linked hypertext document, and has a lot of other nice things to make life easier.
But the real killer feature is automated refactorings. The most basic is rename; you can safely rename a class or method without disturbing other things that have the same name. But there are many more, and they take a lot of the drudgery out of programming.
I keep hearing that Eclipse is just as good, but last I did a project with Eclipse (about 18 months ago) the UI wasn't nearly as well made. Nothing major, but a lot of minor differences that really added up during extended usage.
A lot of things you mention I don't care much about. But I recommend ridiculous amounts of RAM. Even if you get more than you think you'll need, you'll find a use for it.
My latest giant RAM sink is VMWare. I run a virtual copy of Windows for browser testing, and a couple more for virtual servers. Virtual servers are much better for testing than real ones: when you're done trying something out, you can revert the virtual disk back to a known clean configuration.
It has a nubmber of additional practices. The most important are probably test-driven development, refactoring, regular iterations (ideally, one week) and frequent releases.
Utterly meaningless - No matter how excited about the topic, without a focus on profit - they'll be out of a job soon.
You've consistently confused a focus on profit with a focus on keeping out of the red so that you can keep doing what you love.
To stay in business, a small company need only stay barely in the black. And owner-run businesses can and will do that for much less, as owners happily put up with low salaries and long hours because they're doing something they love. Their focus is on doing what they do and staying in business, rather than beating the market average risk-adjusted return on capital.
The difference in financial terms is relatively small. But the difference in attitude is huge. And it's not just anecdotal. There's a whole body of psychology and management literature about the differing effects of intrinsic and extrinsic rewards on motivation and performance.
Carry on with your theoretical arguments and your handwaving if you like. I've seen this time and time again, in industries from dot-com startups to magazines to nursing homes. A business has to stay in black to survive, but the best ones are focused on innovation and quality first, and profit second.
What a complete circle jerk. If they're still arguing about whether or not there is actually a problem without any idea how we can solve it then why the hell do they expect anyone to take them seriously?
For the same reason people should take weather forecasts seriously. We can't know what will happen, but with, say, Katrina, we knew there was a substantial chance of a problem.
Here we're performing a giant experiment with the only planet we have, and the best guess is that involves substantial risks. What to do about that is not a scientific question; it's a moral and political one. It's not the job of citizens to resolve those. It's the job of us, the citizens.
Regardless of what is causing it I want to know, how bad will it be and what can we do to correct it? These are the topics that scientists never appear to talk about.
And that's a good thing. They don't know. They're honest about the fact that they don't know.
If you ask them in the right way, you can find out their personal opinions on what they think we should do. But unlike most people quoted in the media, most scientists understand the gulf between their professional opinions and their personal ones.
Even if all the magazines on the stands were each owned by an individual publisher, they'd still be mostly interested in selling magazines and making a profit.
That's probably not true. I used to do tech work for magazines, and the small specialist mags were always staffed by people who were really excited about the topic.
Think of it like restaurants. Obviously, every restaurant owner has to sell food and make a profit. But there's a world of difference between a single-location, owner-operated restaurant and a giant chain. The decision-makers in the big chains are insulated from the consequences of their decisions. And the clock-punchers on the spot don't have the same incentive to care.
Or compare it to the web. Look at how much cool stuff has appeared on the web, and look how little of it has come from large, corporate publishers. The best technical reviews don't come from USA Today or Ziff-Davis; instead I go to Tom's Hardware, AnandTech, Ars Technica, and a host of other small publishers who do it because that's what they want to do, and for whom making a profit is secondary.
I have never, ever seen a student running in a non-administrator account on their Windows PC, even though XP supports this feature.
Have you actually tried this?
I got a Windows XP laptop recently because I'm working on a project with a Windows client. For probably four weeks, I tried hard to use the administrator account only for installing stuff.
It really sucked. A number of applications wouldn't work properly unless I had administrator rights. Doing all sorts of typical laptop stuff (e.g., switching the wireless to another SSID) required superpowers. I was logging in and out all the goddamn time. Eventually, I just gave up.
This shocks me; having used various Unix variants for a long time, I'm completely comfortable with the user/root distinction, and would never just log in as root. But Windows doesn't seem to be set up to support that.
Zonealarm asks them if they want to let, say, AIM or Weatherbug access their network connect - and the user freaks out.
A lot of this could be solved with smarter interfaces. I'm using whatever firewall Symantec sells, and it pops up similar warnings and gives advice. Even better would be if it looked up access patterns against a centralized database, so that it just knows that, say, Google Desktop is composed of 3 applications with access patterns X, Y, and Z. Then it could say, "I see you're installing Google Desktop. It will get and transmit information over the Internet. We consider it safe. Should I let it run?" With the right cues, users won't freak out.
One of the points basically comes down to "write perfect code". Well, duh, why didn't I think of that before? Jeez. Patching is bad because your code should have been perfect in the first place? That's the dumbest thing I ever heard.
You completely missed his point.
He's saying that regular patching to fix security holes is a symptom of a flaw in your design process. If you've designed for security from the beginning, your security patches should be few and far between. He even gives examples of programs that don't have this problem: Postfix and Qmail.
The problem he's pointing at isn't the occasional patch; it's the attitude that regular security patching is the only way, or even a good way, to deal with security flaws.
Let's also consider the humble shredder.
I just cleaned out several boxes of stuff that I've been lugging around for more than a decade. Once I actually sat down and went through it, I realized that I'd never look at 95% of it again. I kept the stuff with sentimental value, and the rest is confetti. What a relief!
Which will be completely irrelevant when push comes to shove. Given a choice between a. living in the dark. or b. not living in the dark, I have few doubts that the majority of us will choose option "b". That means that nuclear power plants will get built, and that the government will indemnify them to whatever degree is required.
Yes, but why is it only nuclear that deserves such massive operational subsidies? Not just in insurance, but in waste disposal, security, and fuel supply? Not to mention R&D subsidies and construction incentives.
It seems to me that if the government is going to get involved, a carbon tax is the most obvious option; then the best tech can duke it out in the marketplace. But if it's going to play favorites, why not favor something that doesn't carry the apparently substantial risk of a Katrina-sized evacuation and cleanup effort?
Conventional power generation isn't necessarily immune to liability issues either. I mean, a whole lot of people are dying of ailments directly related to the use of coal-fired powerplants. The way I figure it, if they can sue a gun maker because a thug shot somebody, they can go after power plant operators for causing cancer and other ailments.
There are two main obstacles that I know of. One is that the causal chain is impossible to establish. I can easily prove who made the gun that killed my spouse and demonstrate that they didn't exercise due care in designing, advertising, and selling them. But I can't prove who made the particular soot particles that caused her asthma or lung cancer.
The other is that power producers are extensively regulated, emitting waste at levels the government has determined to be safe.
In the case of another Chernobyl, though, it's pretty obvious who is at fault, and a disaster like that is probably above EPA limits even these days.
But many people are deathly afraid of the idea with good reason: when nuke plants fail they fail really, really badly.
Interestingly, a while back The Economist did a long article on the pros and cons of using more nuclear power in the US. They concluded that it wasn't economical without huge government subsidies, which come in the form of free insurance. If nuke plants had to pay the full cost of the insurance to cover catastophes, they would cost well more than fossil fuels do.
Can you imagine of most of our protocols were named by picking some vaguely related word in literature somewhere?
That's a good point. It has certainly worked well for "quark".
They're cheap enough that someone could buy a thousand of them and distribute them to everyone in the Astrodome.
Indeed, they had already arranged a donation of 10,000 radios, so that there would be plenty for all.
Very true, but that doesn't diminish the fact that it WILL help people.
Indeed. A psychiatrist was mentioning somewhere that one of the worst things for people who have been through disasters is to sit around with nothing to do and nothing but the disaster to think about. For people developing PTSD, it can intensify and lengthen their problems.
And there's a lot to be said for community-building and morale. Heck, just the music alone would help. Imagine you're on a long road trip and the radio breaks. How sad would you be? And now imagine somebody else is driving, you don't know where you're going, and you're not sure when you'll get there.