Agreed. While it is still technically possible that plan all along was to immediately terminate the relationship and claim ignorance, this seems exceedingly unlikely.
And more importantly, as you point out, the non-inclusion of this in the submitter's post (whether intentional or not) is a glaring oversight. Ironically, the OP failed in the same way (a little more research) that Dean (or more likely, Dean's campaign crew) failed in.
A link near the bottom of the "technical details" page indicates that Dean was responsible. The page goes on to imply that it was foolish/irresponsible but unintentional.
After the Dean campaign was presented with clear cut evidence as to the nature of emailresponse.net, they investigated promptly and terminated their relationship with the company that same day.
I think this emphasizes the conclusion of the article even more. Make benchmarks on current games and current applications. If manufactures make changes to "cheat" against these benchmarks, then they are doing us a service. This is as opposed to when they "cheat" on benchmarks for older software... 160 fps vs. 100 fps in Quake III isn't so useful.
Actually, this reminds me of the "teaching to the test" controversy. This should lead to good results when the test is carefully designed, but may lead to poor results otherwise.
This kind of idea applies to a system where intrusions and failures are acceptable in order to learn the weak points (and then fix them). On development systems in a controlled environment, this may be appropriate. On fully functional systems, it may not be.
I see people comment daily on the faults of security through obscurity... to me, obscurity can be one part of a total security package.
How many of you have email addresses partly designed to avoid random spam? How do you feel about having one nuclear disaster to learn how to prevent a similar disaster from happening in the future?
I don't think this detracts from your argument, but probabilities do not add -- they multiply.
So, for 100 parts, each good 99% of the time, the chance that no part fails is about 36.6%. Thus, one could expect an average of.634 failures.
For 100 parts, each good 99.99% of the time, the chance that no part fails is about 99.0%. Thus, one could expect an average of.01 failures. Your simplified analysis got this one right due to the laws of small numbers.
Genetic algorithms do not imply "true to life" algorithms. They are merely a general method used to search for good solutions in difficult optimization problems.
A google search led me to this article at abcnews.go.com from the end of July. I personally disagree with many decisions Bush has made, but I am satisfied with him as president.
More broadly, 69 percent of Americans in a new ABCNEWS/Washington Post poll approve of Bush's overall job performance. That peaked at a record 92 percent in an ABCNEWS poll last October and has inched down in 10 out of 12 polls since.
Bush's rating is closer to his pre-Sept. 11 level (14 points higher) than to his post-Sept. 11 high (23 points lower). But 69 percent approval isn't chopped liver: It matches Bill Clinton's career high, and falls just four points short of Ronald Reagan's.
I don't think either candidate (Bush/Gore) was an overwhelmingly poor choice.
Granted, this does assume (as you pointed out in a different response to me) that some candidate will be available that is not a terrible choice. However, when this is not the case, the issue about which voting system to use seems less important overall.
Issue (A) is designed to prevent issue (B). As a side effect, (A) also encourages moderate candidates. Think of it this way... if candidate 1 takes a liberal view on most issues, and candidate 2 takes a slightly liberal view, candidate 2 will collect almost all conservative votes and most moderate votes, winning the election. Of course, due to multiple issues of interest, things aren't *quite* this simple, but the concept still applies.
This mathematical impossibility is important to keep in mind. The article is not particularly scientific... early on it states that the plurality system is the cause of paradoxes... and thus implying that other systems may not have such issues. Later, the following quote:
"While Arrow's theorem shows that no system is flawless, many capture voter preferences more effectively than plurality voting does."
Voting will never be mathematically successful in determining the collective will of the people (assuming such a thing exists). Voting merely gives the people the ability to collectively reject a poor candidate. Plurality voting (along with many other reasonable systems) accomplishes this task.
"On average, a package takes about twice as long to download in source form than in binary form. Also, source takes about as much time again to build. So all up, you're looking at about 4 times as long to install a given Gentoo package as the same package on Debian."
No. You are looking at about two times as long to install the package.
Except in rare circumstances (with degenerate solutions), you need to be within range of at least three receiver stations for any attempt at triangulation. This is due to the need to determine the (x, y, t) source of the cell phone signal. For simplicity, one can assume the signal originates from some univeral ground level assigned to each receiver station. In practice, this is often not a safe assumption.
Assuming you are in an area where you are visible to at least three receiver stations, there is an additional problem with multipath. Signals may reflect off of objects, increasing the distance (and thus time) to the receiver station. Due to this, estimations with *five* receiver stations (and non-trivial estimation theory) can still be grossly inaccurate.
That was the quote that stuck out most to me, too. However, I am not entirely sure I agree with it.
First of all, it implies some absolute sense of right and wrong... while this may be indeed be the case, it is somewhat subjective.
I would then argue that there are laws which were appropriate for the time they were created, but have since been outdated. It seems that technology could certainly be a (if not the) major factor in the process of laws becoming outdated.
I'd like to think others can come up with some good examples of this, but for starters, consider child labor laws. This link explains why these laws were "good" in the 1930s and then proceeds to discuss the implications for today.
Well, for starters, most people do not live in a big city.
It should be "watching TV"... not "looking TV." Of course, I would be pleased if I could write Swedish as well as you write English. For reference sake, I can speak a little Hungarian... a terribly useful skill here in New York.
If I recall correctly, the average American with at least one TV in the household watches an average of two hours of TV per day. Of course, I'm sure that number varies wildly from study to study. I would trust it about as much as "The average American eats two donuts per day."
And have you tried Phoenix yet? Using Mozilla may put you amongst the "masses" in a few years.
The novel part is the queing system. Rather than queing inefficiently at the bathroom, a virtual queue would be diffused over a large area.
-- jetlag --
Backing up the CD for games you buy is generally not necessary if you have access to high speed internet. Just go online and download it... legally!
On the other hand, keeping a CD with your saved-games for lots of different programs might not be a bad idea. Never know when you may want to show some little cousin your super city in SimCity or your total destruction of some StarCraft level.
To argue that a five year (or *any* fixed year) term for a (stock-like) investment is weak. It may be that the stock market will be low ten, twenty, and thirty years from now.
A better plan may be to invest with the intentions of cashing it only when you will be happy with the outcome. Granted this plan must be modified to address the case of something continually slipping down... when do you jump off the boat?
This plan is based on the concept of investing money that you have no immediate or long-term plans to use. Once you do develop usage plans, you should immediately think about cashing in when you would be (again) happy with the outcome. Don't wait until the need for the cash is upon you.
Not being in the crypto world, I may be incorrect on this, but I would say that AES is not mathematically secure. In fact, anything that can be brute forced is not mathematically secure.
Rather, it is (by current standards) practically secure.
Companies are distrustful for Microsoft? I don't think so. I suspect that most companies are thankful and appreciative of Microsoft. Maybe the IT folk at companies that get hit by major security exploits are upset at Microsoft... but usually (I suspect) the blame lies equally with the IT folk.
As a disclaimer, I am not in IT, and I am well aware of many responsible people in it.
Actually, the grenade itself may have a constant chance of blowing up at any given moment of fiddling. I think you meant to say that the longer you fiddle with it, the more likely you are to blow up.
Slashdot Mirror
Agreed. While it is still technically possible that plan all along was to immediately terminate the relationship and claim ignorance, this seems exceedingly unlikely.
And more importantly, as you point out, the non-inclusion of this in the submitter's post (whether intentional or not) is a glaring oversight. Ironically, the OP failed in the same way (a little more research) that Dean (or more likely, Dean's campaign crew) failed in.
-- jetlag --
A link near the bottom of the "technical details" page indicates that Dean was responsible. The page goes on to imply that it was foolish/irresponsible but unintentional.
After the Dean campaign was presented with clear cut evidence as to the nature of emailresponse.net, they investigated promptly and terminated their relationship with the company that same day.
-- jetlag --
I think this emphasizes the conclusion of the article even more. Make benchmarks on current games and current applications. If manufactures make changes to "cheat" against these benchmarks, then they are doing us a service. This is as opposed to when they "cheat" on benchmarks for older software ... 160 fps vs. 100 fps in Quake III isn't so useful.
Actually, this reminds me of the "teaching to the test" controversy. This should lead to good results when the test is carefully designed, but may lead to poor results otherwise.
-- jetlag --
Wow, I suppose this isn't too surprising, but it isn't every day you see an established company have their stock increase by 17% in one day.
-- jetlag --
This kind of idea applies to a system where intrusions and failures are acceptable in order to learn the weak points (and then fix them). On development systems in a controlled environment, this may be appropriate. On fully functional systems, it may not be.
... to me, obscurity can be one part of a total security package.
I see people comment daily on the faults of security through obscurity
How many of you have email addresses partly designed to avoid random spam? How do you feel about having one nuclear disaster to learn how to prevent a similar disaster from happening in the future?
-- jetlag --
An increase from 48X to 52X only represents an 8.33 percent increase in speed. Am I the only one not impressed by this?
-- jetlag --
I don't think this detracts from your argument, but probabilities do not add -- they multiply.
.634 failures.
.01 failures. Your simplified analysis got this one right due to the laws of small numbers.
So, for 100 parts, each good 99% of the time, the chance that no part fails is about 36.6%. Thus, one could expect an average of
For 100 parts, each good 99.99% of the time, the chance that no part fails is about 99.0%. Thus, one could expect an average of
-- jetlag --
I see the terminology 'runway incursion' and 'airspace incursion' in the publications I read.
-- jetlag --
"True to life" AI, does not imply good "Game" AI.
Genetic algorithms do not imply "true to life" algorithms. They are merely a general method used to search for good solutions in difficult optimization problems.
-- jetlag --
Bush's rating is closer to his pre-Sept. 11 level (14 points higher) than to his post-Sept. 11 high (23 points lower). But 69 percent approval isn't chopped liver: It matches Bill Clinton's career high, and falls just four points short of Ronald Reagan's.
-- jetlag --
I don't think either candidate (Bush/Gore) was an overwhelmingly poor choice.
Granted, this does assume (as you pointed out in a different response to me) that some candidate will be available that is not a terrible choice. However, when this is not the case, the issue about which voting system to use seems less important overall.
-- jetlag --
- A) the two party system
Issue (A) is designed to prevent issue (B). As a side effect, (A) also encourages moderate candidates. Think of it this wayB) Nader 'stealing' Gore's votes
-- jetlag --
This mathematical impossibility is important to keep in mind. The article is not particularly scientific ... early on it states that the plurality system is the cause of paradoxes ... and thus implying that other systems may not have such issues. Later, the following quote:
"While Arrow's theorem shows that no system is flawless, many capture voter preferences more effectively than plurality voting does."
Voting will never be mathematically successful in determining the collective will of the people (assuming such a thing exists). Voting merely gives the people the ability to collectively reject a poor candidate. Plurality voting (along with many other reasonable systems) accomplishes this task.
-- jetlag --
"On average, a package takes about twice as long to download in source form than in binary form. Also, source takes about as much time again to build. So all up, you're looking at about 4 times as long to install a given Gentoo package as the same package on Debian."
No. You are looking at about two times as long to install the package.
2*dwn_ld + 2*src_build = 2(dwn_ld + src_build)
-- jetlag --
Except in rare circumstances (with degenerate solutions), you need to be within range of at least three receiver stations for any attempt at triangulation. This is due to the need to determine the (x, y, t) source of the cell phone signal. For simplicity, one can assume the signal originates from some univeral ground level assigned to each receiver station. In practice, this is often not a safe assumption.
Assuming you are in an area where you are visible to at least three receiver stations, there is an additional problem with multipath. Signals may reflect off of objects, increasing the distance (and thus time) to the receiver station. Due to this, estimations with *five* receiver stations (and non-trivial estimation theory) can still be grossly inaccurate.
-- jetlag --
That was the quote that stuck out most to me, too. However, I am not entirely sure I agree with it.
... while this may be indeed be the case, it is somewhat subjective.
First of all, it implies some absolute sense of right and wrong
I would then argue that there are laws which were appropriate for the time they were created, but have since been outdated. It seems that technology could certainly be a (if not the) major factor in the process of laws becoming outdated.
I'd like to think others can come up with some good examples of this, but for starters, consider child labor laws. This link explains why these laws were "good" in the 1930s and then proceeds to discuss the implications for today.
-- jetlag --
"American colleges are a hotbed of socialism and left thinking. Everything is a slippery slope to them, and they'll fight back."
Student-wise? Yes.
Professor-wise? Maybe.
Administration-wise? I don't think so.
-- jetlag --
Well, for starters, most people do not live in a big city.
... not "looking TV." Of course, I would be pleased if I could write Swedish as well as you write English. For reference sake, I can speak a little Hungarian ... a terribly useful skill here in New York.
It should be "watching TV"
If I recall correctly, the average American with at least one TV in the household watches an average of two hours of TV per day. Of course, I'm sure that number varies wildly from study to study. I would trust it about as much as "The average American eats two donuts per day."
And have you tried Phoenix yet? Using Mozilla may put you amongst the "masses" in a few years.
-- jetlag --
The novel part is the queing system. Rather than queing inefficiently at the bathroom, a virtual queue would be diffused over a large area. -- jetlag --
Backing up the CD for games you buy is generally not necessary if you have access to high speed internet. Just go online and download it ... legally!
On the other hand, keeping a CD with your saved-games for lots of different programs might not be a bad idea. Never know when you may want to show some little cousin your super city in SimCity or your total destruction of some StarCraft level.
-- jetlag --
To argue that a five year (or *any* fixed year) term for a (stock-like) investment is weak. It may be that the stock market will be low ten, twenty, and thirty years from now.
... when do you jump off the boat?
A better plan may be to invest with the intentions of cashing it only when you will be happy with the outcome. Granted this plan must be modified to address the case of something continually slipping down
This plan is based on the concept of investing money that you have no immediate or long-term plans to use. Once you do develop usage plans, you should immediately think about cashing in when you would be (again) happy with the outcome. Don't wait until the need for the cash is upon you.
-- jetlag --
Not being in the crypto world, I may be incorrect on this, but I would say that AES is not mathematically secure. In fact, anything that can be brute forced is not mathematically secure.
Rather, it is (by current standards) practically secure.
-- jetlag --
Companies are distrustful for Microsoft? I don't think so. I suspect that most companies are thankful and appreciative of Microsoft. Maybe the IT folk at companies that get hit by major security exploits are upset at Microsoft ... but usually (I suspect) the blame lies equally with the IT folk.
As a disclaimer, I am not in IT, and I am well aware of many responsible people in it.
-- jetlag --
The linked article is very weak. His homepage is here:
nd.edu/~alb/
The specific article is here:
nd.edu/~networks/PDF/NatureImmunol%202002.pdf
Hopefully Notre Dame can handle the traffic.
-- jetlag --
Actually, the grenade itself may have a constant chance of blowing up at any given moment of fiddling. I think you meant to say that the longer you fiddle with it, the more likely you are to blow up.