(This post isn't really about the parent or the article, I'm just getting on a soapbox).
"If the law prevents others..."
Laws never prevent anything. Laws ascribe consequences to convictions. No matter what penal deterents you put in place to make a choice less attractive, it is still a choice.
It is possible to increase security, but all increases in security necessarily incur a usability cost. Adding a guard to a knife may protect the user's fingers, but will get in the way of a skilled craftsman. Adding metal to a car to make it sturdier may interfere with visibility and handling.
It is also possible to use physical force (handcuffs, walls) and threat of violence (guns) to lock people down, such as in a concentration camp or jail. This will prevent the prisoners from doing things outside the walls of their confines, but is otherwise hardly different from locking the rest of the world up, as far as 'preventing' things goes.
As long as we put responsibility for people's actions in the hands of other people, we can expect people to behave irresponsibly. It is a self-fulfilling prophecy. I don't claim to have the answer, I claim all answers are trade-offs.
A) whitelist-only doesn't protect against spoofing B) whitelist-only doesn't protect against your friends giving you viruses C) whitelist-only still allows stupid HTML mail with 30 charachters of text and a 10k 'stationary' background image.
From scanning the article, it looks like it doesn't even use SMTP, POP, IMAP or anything else that would integrate with an email client.
In any case, as usual the/. article is completly mis-directing. From the article:
Hardware accepts dmail is not an "alternative to email" as the release claims, and cannot therefore eradicate spam. "I have a dmail account, but I could not do without email," he says. "I still get about 200 spams a day."
Think of it like that cell-phone service with the walkie-talkie feature. It's not a replacement, it's an addition.
They could even use taxes to enforce voting. When people fill in and submit their ballots, they could have a unique number given to them by the voting attendants which they could put on their tax form to give them an exemption from the "didn't vote tax". Those who supply an invalid number or a duplicate number get audited. Those who supply no number get 'fined' by paying the full tax. Those who don't submit taxes are already at risk for being audited.
This would have the added benefit of reminding people about the relationship between voting and the use of their taxes.
Maybe he's waiting for Parrot to be done so he can host the first version on that?
I find the Parrot (perl6 vm) family of languages far more promising anyway. Arc is interesting in the same way Paul's essays are interesting: they get me thinking about things I like to think about. Then I go do something else.
I agree with your premise, but good luck getting anything done about it.
Enforcing rules of consentual interaction in a prison is extremely expensive because the inmates quickly develop a culture for working around any enforcement scheme. For example, an inmate is welcome to complain about abuse from other inmates, and will probably be transferred to another prison in that event. They will be no better off at the new prison, and their 'tattling' will probably be 'leaked' at the new prison, where the inmate in question will be 'punished' for complaining.
To counter this, the prisons and procedures would have to be re-designed to not just keep prisoners in, but to keep them from getting at each other. Cells would have to have their own bathing facilities, and would have to not be shared. It would also be a huge hassle for the already stressed out staff.
I agree that it's a worthy goal, but you're going to have a hard time convincing the masses that 'guilty' prisoners deserve the expense of protection.
Schroedinger's Stock Market? The market is both Bull and Bear until you trade? Or the Market has an equal probability of being alive or dead until you open for business?
Really all you need is a shared database of form filling information and a client-side module for generating it. Then you could generate new info every time you visit a site, which would have the added benefit of preventing them from even tracking users between visits. If the information submission module were integrated into a browser, it might be able to 'watch' someone fill out a form and learn from that as a means of populating the form database.
Even if they add 'captchas' to their registrations, this tool could reduce the 'registration' process to just identifying a captcha.
A user is only new to a tool once, but if the tool is any good the user will use the tool for the rest of her life, making the time spent learning the tool insignificant. Rather than making interfaces 'discoverable', I'd like to see interfaces made 'efficient' and augmented with a tutorial mode. There is no reason to combine tutorial mode with day-to-day usage mode. This applies equally well to software as to anything else.
What if the Xbox2 could be had for less than $150? The rumors say it's going to run on a PowerPC and have no hard drive. That sounds a lot like a GameCube to me. Throwing away all the PC crap in the X1 might allow them to drop the cost drastically. It might also simplify development, making the platform slightly more attractive to console game writers. Obviously this is pure speculation, but maybe the gambit they're making is that they can come out with something which is the cheapest, most beautiful, and easiest to develop for, and that more games will be ported to it if it's that accessible.
As the devil's advocate, I must remind you that Those Who Favor This Sort Of Thing, contend that copying data is stealing. The argument they make is that the party who created the data has a claim to all potential revenue from all the copies of the data. Any copying which does not include revenue for the creator is stealing from that creator because the copier is getting something for nothing.
Of course we all know what the fallacy is here. It's the idea that the creator can tell when someone has 'stolen' from them. With normal property the creator no longer has the thing which was stollen from them. With data this is not the case.
The reason I bring this up is that some of us will get into arguments with people who don't understand this distinction. If we merely tell them that noone is hurt buy copying, they will think we are stupid or ignorant because CLEARLY the author is being cheated out of earned revenue. Rather than basing our arguments on these contentious notions, we must start from positions we can agree with them on.
Unfortunately, I don' know of a way to state our position in a way they can understand. "They can't tell it's been copied" won't work because the people we are arguing with will say, "sure they can - they didn't get money for the copy." This sounds ludicrous to us because we know you can't test for things not happening if you don't have a reason to expect them to happen, but they don't think that way.
In any case, it's important that we not descend into a "does not / does too" argument with these people.
Every object in the universe is unique, for if two objects were identical in every way, they would be the same object. Furthermore, no objects are unique because they all have something in common with every other object: they are all unique.
Degrees of uniqueness exist because there are different contexts for evaluating an object's differences from its peers. Every evaluation has a context within which it has more than two degrees.
If one in a billion oranges looks like Elvis, that's pretty unique. If one in a billion of those Elvis-shaped oranges is a life-sized bust of The King, that's an extremely unique orange. The uniqueness of the non-bust Elvis oranges is not undermined by discovering another instance because within the context of people's knowledge of oranges, two Elvis oranges would be rare enough as to be unique from an individual's point of view.
Absolute uniqueness is only a useful concept in abstract contexts like math and philosophy. In the real world we use the word loosely within implicit or explicit contexts which give the seemingly redundant expression 'very unique' a meaning which is different from other degrees of uniqueness.
People who fall into this trap often also argue that
* Black is not a color (it is not a hue, but it is a color)
* Zero is not a number (it is not a counting number, but it is a number)
* A logical statement must either be true, false, or a paradox (it depends on what you mean by 'logical', 'statement', 'true', etc.)
So yes, something can be 'very unique', or only 'slightly unique'.
"I can give the government half of what I create and they can squander it, or I can give them a quarter, invest the quarter I save, and in 100 years that quarter will have grown to about 1.05^100 (about 131) times its original size..."
The longer I have to enjoy them, the more I'll value the fruits of my labor.
Most people are just annoyed by assholes, as you say.
Anyone who genuinely believes cell phones and SUVs are a problem is probably just jealous. They think that life is more difficult for them because someone else has something they don't have. It's class envy motivated out of feelings of helplessness.
One could have said something similiar about automative and horse-drawn carriage interfaces around 90 years ago. It's not a flawless analogy, but your point is far from unassailable.
* terminal consoles HAVE changed a great deal in 30 years (tab completion, screen, mouse daemons, curses, whiptail, multi-byte support,...)
* Most GUI differences are superficial tweaks made to thwart lawsuits, or to convince potential customers that there's a difference between OS versions that's worth upgrading for.
* The people who are intimidated by either interface tend to just be intimidated by computers. The rest will use whatever is best for the job.
I prefer text interfaces because it suits the way I think, but my extremely intelligent girlfriend understands both and prefers GUIs because they match how she thinks.
It's about time we grow out of this kind of debate...
As you say, managing trust hierarchically is non-trivial on this scale.
Even if that weren't the case, I'm not comfortable with the idea that only certain entities have the power to decide who may or may not use a protocol publicly. The policy would have to be enforced to be useful, and enforcement would be a huge impingement on people's rights.
If you give certs away, there's no trust. If you restrict them there's no freedom.
If the enclure handled hot spares internally and used a third of its drives for redundancy, it would then have 10 levels of failure instead of one. Instead of being 100% or 0% working, it could be any of the nine values between.
This wouldn't be any help for anyone with space for two regular drives, but for a laptop it would be awesome.
Slashdot Mirror
Because the apartment building is in a college town.
Damn kids...
(This post isn't really about the parent or the article, I'm just getting on a soapbox).
..."
"If the law prevents others
Laws never prevent anything. Laws ascribe consequences to convictions. No matter what penal deterents you put in place to make a choice less attractive, it is still a choice.
It is possible to increase security, but all increases in security necessarily incur a usability cost. Adding a guard to a knife may protect the user's fingers, but will get in the way of a skilled craftsman. Adding metal to a car to make it sturdier may interfere with visibility and handling.
It is also possible to use physical force (handcuffs, walls) and threat of violence (guns) to lock people down, such as in a concentration camp or jail. This will prevent the prisoners from doing things outside the walls of their confines, but is otherwise hardly different from locking the rest of the world up, as far as 'preventing' things goes.
As long as we put responsibility for people's actions in the hands of other people, we can expect people to behave irresponsibly. It is a self-fulfilling prophecy. I don't claim to have the answer, I claim all answers are trade-offs.
http://en.wikipedia.org/wiki/Mycobacterium
thefreedictionary.com is a coat tail rider of Wikipedia. They give attribution in fine print at the bottom, but why not support the source?
_ dy namics
http://en.wikipedia.org/wiki/Modified_Newtonian
A) whitelist-only doesn't protect against spoofing
/. article is completly mis-directing. From the article:
B) whitelist-only doesn't protect against your friends giving you viruses
C) whitelist-only still allows stupid HTML mail with 30 charachters of text and a 10k 'stationary' background image.
From scanning the article, it looks like it doesn't even use SMTP, POP, IMAP or anything else that would integrate with an email client.
In any case, as usual the
Hardware accepts dmail is not an "alternative to email" as the release claims, and cannot therefore eradicate spam. "I have a dmail account, but I could not do without email," he says. "I still get about 200 spams a day."
Think of it like that cell-phone service with the walkie-talkie feature. It's not a replacement, it's an addition.
They could even use taxes to enforce voting. When people fill in and submit their ballots, they could have a unique number given to them by the voting attendants which they could put on their tax form to give them an exemption from the "didn't vote tax". Those who supply an invalid number or a duplicate number get audited. Those who supply no number get 'fined' by paying the full tax. Those who don't submit taxes are already at risk for being audited.
This would have the added benefit of reminding people about the relationship between voting and the use of their taxes.
Bah. I checked the Arc FAQ, and it explicitly states they want nothing to do with Parrot.
Oh well.
Maybe he's waiting for Parrot to be done so he can host the first version on that?
I find the Parrot (perl6 vm) family of languages far more promising anyway. Arc is interesting in the same way Paul's essays are interesting: they get me thinking about things I like to think about. Then I go do something else.
Now I'm going to go do something else.
I agree with your premise, but good luck getting anything done about it.
Enforcing rules of consentual interaction in a prison is extremely expensive because the inmates quickly develop a culture for working around any enforcement scheme. For example, an inmate is welcome to complain about abuse from other inmates, and will probably be transferred to another prison in that event. They will be no better off at the new prison, and their 'tattling' will probably be 'leaked' at the new prison, where the inmate in question will be 'punished' for complaining.
To counter this, the prisons and procedures would have to be re-designed to not just keep prisoners in, but to keep them from getting at each other. Cells would have to have their own bathing facilities, and would have to not be shared. It would also be a huge hassle for the already stressed out staff.
I agree that it's a worthy goal, but you're going to have a hard time convincing the masses that 'guilty' prisoners deserve the expense of protection.
Schroedinger's Stock Market? The market is both Bull and Bear until you trade? Or the Market has an equal probability of being alive or dead until you open for business?
I'm confused.
Really all you need is a shared database of form filling information and a client-side module for generating it. Then you could generate new info every time you visit a site, which would have the added benefit of preventing them from even tracking users between visits. If the information submission module were integrated into a browser, it might be able to 'watch' someone fill out a form and learn from that as a means of populating the form database.
Even if they add 'captchas' to their registrations, this tool could reduce the 'registration' process to just identifying a captcha.
I think they're worried that one of the vehicals will "make a run for the border."
:%!tidy -q
These eleven keystrokes (including 'enter') mean 'replace all the lines in this file with the result of piping them through the command "tidy -q"'.
This should work for any version of vi, but vim really is the best.
I often begin an HTML doc with
Page description
And then ":%!tidy -iq -asxml" fills the rest in for me, indented and everything.
A user is only new to a tool once, but if the tool is any good the user will use the tool for the rest of her life, making the time spent learning the tool insignificant. Rather than making interfaces 'discoverable', I'd like to see interfaces made 'efficient' and augmented with a tutorial mode. There is no reason to combine tutorial mode with day-to-day usage mode. This applies equally well to software as to anything else.
> Hell, I could do this much easier.
> Take an ICBM, put a stewardess inside.
> Ta da' It's an airplane, not a missle.
Worst. Haiku. Ever.
What if the Xbox2 could be had for less than $150? The rumors say it's going to run on a PowerPC and have no hard drive. That sounds a lot like a GameCube to me. Throwing away all the PC crap in the X1 might allow them to drop the cost drastically. It might also simplify development, making the platform slightly more attractive to console game writers. Obviously this is pure speculation, but maybe the gambit they're making is that they can come out with something which is the cheapest, most beautiful, and easiest to develop for, and that more games will be ported to it if it's that accessible.
Or they could just be stupid.
As the devil's advocate, I must remind you that Those Who Favor This Sort Of Thing, contend that copying data is stealing. The argument they make is that the party who created the data has a claim to all potential revenue from all the copies of the data. Any copying which does not include revenue for the creator is stealing from that creator because the copier is getting something for nothing.
Of course we all know what the fallacy is here. It's the idea that the creator can tell when someone has 'stolen' from them. With normal property the creator no longer has the thing which was stollen from them. With data this is not the case.
The reason I bring this up is that some of us will get into arguments with people who don't understand this distinction. If we merely tell them that noone is hurt buy copying, they will think we are stupid or ignorant because CLEARLY the author is being cheated out of earned revenue. Rather than basing our arguments on these contentious notions, we must start from positions we can agree with them on.
Unfortunately, I don' know of a way to state our position in a way they can understand. "They can't tell it's been copied" won't work because the people we are arguing with will say, "sure they can - they didn't get money for the copy." This sounds ludicrous to us because we know you can't test for things not happening if you don't have a reason to expect them to happen, but they don't think that way.
In any case, it's important that we not descend into a "does not / does too" argument with these people.
It says two exceptions, not two women. They're probably aliens, or dogs, or elder gods or something.
Every object in the universe is unique, for if two objects were identical in every way, they would be the same object. Furthermore, no objects are unique because they all have something in common with every other object: they are all unique.
Degrees of uniqueness exist because there are different contexts for evaluating an object's differences from its peers. Every evaluation has a context within which it has more than two degrees.
If one in a billion oranges looks like Elvis, that's pretty unique. If one in a billion of those Elvis-shaped oranges is a life-sized bust of The King, that's an extremely unique orange. The uniqueness of the non-bust Elvis oranges is not undermined by discovering another instance because within the context of people's knowledge of oranges, two Elvis oranges would be rare enough as to be unique from an individual's point of view.
Absolute uniqueness is only a useful concept in abstract contexts like math and philosophy. In the real world we use the word loosely within implicit or explicit contexts which give the seemingly redundant expression 'very unique' a meaning which is different from other degrees of uniqueness.
People who fall into this trap often also argue that
* Black is not a color (it is not a hue, but it is a color)
* Zero is not a number (it is not a counting number, but it is a number)
* A logical statement must either be true, false, or a paradox (it depends on what you mean by 'logical', 'statement', 'true', etc.)
So yes, something can be 'very unique', or only 'slightly unique'.
"I can give the government half of what I create and they can squander it, or I can give them a quarter, invest the quarter I save, and in 100 years that quarter will have grown to about 1.05^100 (about 131) times its original size..."
The longer I have to enjoy them, the more I'll value the fruits of my labor.
Most people are just annoyed by assholes, as you say.
Anyone who genuinely believes cell phones and SUVs are a problem is probably just jealous. They think that life is more difficult for them because someone else has something they don't have. It's class envy motivated out of feelings of helplessness.
One could have said something similiar about automative and horse-drawn carriage interfaces around 90 years ago. It's not a flawless analogy, but your point is far from unassailable.
...)
* terminal consoles HAVE changed a great deal in 30 years (tab completion, screen, mouse daemons, curses, whiptail, multi-byte support,
* Most GUI differences are superficial tweaks made to thwart lawsuits, or to convince potential customers that there's a difference between OS versions that's worth upgrading for.
* The people who are intimidated by either interface tend to just be intimidated by computers. The rest will use whatever is best for the job.
I prefer text interfaces because it suits the way I think, but my extremely intelligent girlfriend understands both and prefers GUIs because they match how she thinks.
It's about time we grow out of this kind of debate...
As you say, managing trust hierarchically is non-trivial on this scale.
Even if that weren't the case, I'm not comfortable with the idea that only certain entities have the power to decide who may or may not use a protocol publicly. The policy would have to be enforced to be useful, and enforcement would be a huge impingement on people's rights.
If you give certs away, there's no trust.
If you restrict them there's no freedom.
lose-lose situation.
If the enclure handled hot spares internally and used a third of its drives for redundancy, it would then have 10 levels of failure instead of one. Instead of being 100% or 0% working, it could be any of the nine values between.
This wouldn't be any help for anyone with space for two regular drives, but for a laptop it would be awesome.
http://www.wsdot.wa.gov/PugetSoundTraffic/cameras/