Except for the anoyance of clicking the prompt all the time. My problem is that even with ActiveX dissabled (not prompted; completely off) there are still many websites I visit that would pop up an anyoying 'this page may not look/work properly without ActiveX' warning on every single page that you are forced to click 'OK' to acknowledge. When I turn something off, I don't want to be harassed about it.
Of course Firefox doesn't have this problem.:-)
OFDM has nothing to do with mulitple antenae. You can send OFDM signals across a channel with a single RX and TX antena. OFDM is simply a convinient way to turn a freq-selective fading channel into multiple, independent flat fading channels by interpreting the data as the IFFT of the TX signal. Spreading the data out over frequency can be used to get good diversity and thus avoid bad fades.
Using multiple antenae is also a good way to get diversity in a way that complements OFDM (spatial vs. freqency). More diversity in an RF channel is almost always good. The exception is when it forces you to spread your energy over multiple sources too mutch. However multiple RX antena don't increase energy use, instead they just absorb more energy, and in a manner that is better than a single antena with twice the gain. This is because while one antena may be in a bad fade (a destructive multipath null), chances are much lower that both will be in a bad fade.
Multiple TX antenae can be used too, but the implementation (and decoding) gets mutch more complex as the article suggests: you need lots of computing power to grab the bits out of multiple simultaneous (but cooperating) TX sources. You also spread your energy with multiple TX antenae, which lowers overall SNR.
"They don't even have a decent state-wide model for carpool lanes: in SoCal, carpool lanes are carpool lanes 24 hours a day. In NorCal, carpool lanes are carpool lanes only during normal rush-hour traffic, and are normal lanes otherwise."
Pshyah. That is because normal rush-hour traffic is 24 hours a day in SoCal. I've run into gridlock in LA at 4 in the freakin morning.
I believe they've said several times on the show that he is much smarter than he puts on. He gets it, he just doesn't care. It's just his schtic. I mean the Asgard see something in him.
That, and I think that Anderson wanted to play a pollar opposite of his previous roll as he had been a bit type-cast.
/* devil's advocate */ What happens when someone brings extra, forged cards with them and slips them into the audit box?
Is there a simple way to make sure that there is only one card / voter without attatching reversable, unique IDs to each card (and therefore each voter)?
I was thinking some sort of pseduo random sequence per machine* where inserting a random number'd card would violate the sequence. Of course you wold then have to forget / not record the order in which voters used that machine*.
*or per polling place, if all the machines are networked, which would add some noise to the voter order. Still if you were the first to vote that day and the first 10 votes cast were all for candidate X, it would be failry obvious who you voted for.
You assume that the probability of being hit by an asteroid has a Poisson distribution and is therefore memoryless. In which case you are correct, given that nothing has hit us yet doesn't change the distribution function for the future.
But we don't know what causes asteroids to wander our way, only that it hapens on a semi periodic basis. Perhaps as we orbit the galaxy we come accros regions with more gravitational distortions that are more likely to send stuff hurtling inwards from the oort cloud. Perhaps there is a misterious 10th planet that goes through a dense part of the oort cloud. Perhaps....
Anything that makes the system non-memoryless (i.e. statefull) and makes the events more periodic than random allows us to say that given no events so far, the probability of an event in the near future is greater/has gone up. (Extreme example: We arrive in london at some random time and don't have a watch. The fact that Big Ben hasn't rung in the last 40 minutes allows us to state that it will ring 'soon' with greater certanty than the fact it hasn't rung in the last 10.)
Of couse the fact that an asteroid doesn't hit in just one year makes the already small probability change for the next year only by an infentesmal ammount. I.e. a change of 1/50000000 --> 1/49999999 or even smaller.
From the article, it sounds like they are only sending out to party loyalists / people already on distribution lists in order to create a 'buzz'. So you probably won't get one directly but it will be forwarded from a 'friend'.
It'll probably even say something allong the lines of "Forward this to 10 friends or you'll be cursed with 4 years of bad govornment."
But what is accomplished making the bar code hard to read if the plain text is right there just inches away on the same ballot? Or do you somehow cover only the text part before handing in the ballot?
In the one experience I've had with (human completed) scan-tron type ballots, you got to feed the ballot yourself, with an observer making sure you didn't put in more than one, but probably not close enough to make out text or barcode.
And with human-machine readable ballots you wouldn't even need to rely on open source for verification. The voter can just make sure the print-out matches his vote. It lessens the constraints placed on the ballot machine even further. You could have W. himself making a closed source version and you'd be ok, its basicaly just a glorified printer. *shudders thinking of the president trying to write code* (You still have to examine and trust the tallying machine or do it by hand.)
Except they purposely obfuscate the barcode. Asside from failing spot checks, how do you (easily) verify/trust the barcode encodes for the english printed part of the ballot? The stated purpose is to prevent anyone from reading the votes easily, yet it's on the same piece of paper as the human readable vote.
Make the ballot use human-computer readable text. So what the voter looks at to verify is actualy what the machine reads when tallying votes.
Uh. that was exacly my point with #2. You and a machine can read the account number on your checks. I've never seen this with the alphabet, but it probably already exists / wouldn't be too much trouble.
> Should businessmen be penalized and disallowed to vote simply because they were out of the country at the time? What about the military?
Voting should be important enough to businessmen for them to schedule buisness trips arround voting days, which only happen 1-2 days a year. Or rather if voting is important to them (it seems it isn't to 1/2 the eligible voters) then they are free to shedule their trips arround those days.
Granted, this isn't as possible for those in the military or people who are leaving for extended periods (months-years). But there is no reason why electronic voting is supperior to balloting by paper. The only difference is the speed at which results can be transmited/tallied. I, for one, do not wish to sacrifice security / trustworthiness of something as important as an election just for expediency.
This has been my exact idea too. The problem is we are asking the computer voting machines to do too much. The whole rationale behind touch screens was to simplify the creation of the ballot: many mechanical systems were too difficult, too error prone, or available in only one language. The whole Florida debackle was due to a mechanical punch that failed to properly mark the ballot.
The solution? "Lets use computers." Yet some how the assumption leaked in that the computers used to do the balloting had to be the same machines that tallied the votes as well. This is a paradigm that should be abollished as soon as possible. While we fix one problem (ballots not reflecting voter preferences properly), we introduce another (allowing increased access to the device doing the tallying).
The solution as the original comment said is to split the process. Use computers to create 'standardized' ballots, and to simplify/error check each voters choice making. And let the voter see a human readable ballot that they can confirm and turn in at a different part of the polling station.
Tallying can be done in a sepparate process, much the way scan-tron type ballots are counted today quickly and accurately.
Some thoughts / added benefits:
1. The paper trail. Voters see their votes correctly printed on paper. And the ballot machines can be used as a double check to make sure no ballots were destroyed. (added reassurance against tampering, since now it requires a coordinated attack both physical and electronic).
2. If you make the ballot human & computer readable (just like your account number on checks) you can verrify the ballot and not have to assume that the bar code the ballot machine produces matches the text.
3. If the ballot form is standardized then the voting equipment becomes commoditized. States / localities can choose the balloting and tallying equipment manufacturers to buy from independently and no one is tied to a given manufacturer for either device. They can even purchace from more than one vendor for the balloting devices. This will drive down prices, as well as letting the govornment take trustworthiness into account when purchacing equipment.
It would be great if legislatures could demand this type of system instead of letting each district try to 'roll your own' and get unknown results in terms of reliability / trustworthiness. It would also mean we wouldn't have to put anywhere near as much trust in the makers of the machines.
And when your heating / cooling bill goes through the roof because you lined your house with a great thermal conductor?
I'd rather spend the cash on aerogel insulation in the walls.
Uh, except now you have match smoke in the same air you are trying to breath. Also, why would anyone on the ISS have a match?
Better use something that is light, but harder to inhale, like smallish bits of paper. Or tissue.
Oh, and better remember to turn off any air circulating equipment. With a leak that slow, normal air scrubbers are probably going to have more effect on air currents than the leak.
> and any object far enough out from the earth to require a year in order to complete an orbit would passed the instability limit, where it would be captured away by the sun's gravity, and would no longer orbit earth
L1-L3 are on the earth-sun line but aren't stable. L4 and L5, which lie in earth orbit +/- 60 dgrees, are! And all of them orbit earth in a year. (of course, the centers of the orbits aren't earth, they're close to the sun, but they still take a circular path arround us every year).
NASA is currently using L1 (SOHO, ACE) with plans to use L2 for deep space observations.
Of course none of these are practical for a stationary elivator because the earth spins faster than 1/year (unless the end 'dangles' over the equator and moves east-west). And they are too far away.
> The company isn't giving Rick a loan. Rather, he's agreed to work for a paycheck.
Yes. He's entering into a deal where they have to give him money on a regular schedule. If anyone should be giving anyone else a credit report, the company should be giving one to Rick. No one ever credit checks their bank before borrowing a home loan.
"How can I be sure you'll pay me back for my 2weeks/month worth of work if you have poor credit?"
>... This is often done in an interactive fashion.
There is a flaw that makes this not a true turring test: The competition is set up so that your program only plays itself. Your program does not get a chance to play other programs or real humans. So it is sorta like a turring test where the computer is only allowed to talk with itself.
The true guts of the turring test come from telling if an unknown entity is a human or machine through its interaction with a known human, not its interaction with a copy of itself or another machine.
(sometimes the ineracting human decides, and sometimes an over the sholder observer tries to decide. depends on the setup)
To realy make it work, you need programs that can be input an arbitrary sequence of moves that can then decide the next move. Programs and humans can then be run against humans (and even other programs) randomly. The detectors job would be to decide wich players were human instead of which sets of games were human.
Slashdot Mirror
Except for the anoyance of clicking the prompt all the time. My problem is that even with ActiveX dissabled (not prompted; completely off) there are still many websites I visit that would pop up an anyoying 'this page may not look/work properly without ActiveX' warning on every single page that you are forced to click 'OK' to acknowledge. When I turn something off, I don't want to be harassed about it. Of course Firefox doesn't have this problem. :-)
OFDM has nothing to do with mulitple antenae. You can send OFDM signals across a channel with a single RX and TX antena. OFDM is simply a convinient way to turn a freq-selective fading channel into multiple, independent flat fading channels by interpreting the data as the IFFT of the TX signal. Spreading the data out over frequency can be used to get good diversity and thus avoid bad fades.
Using multiple antenae is also a good way to get diversity in a way that complements OFDM (spatial vs. freqency). More diversity in an RF channel is almost always good. The exception is when it forces you to spread your energy over multiple sources too mutch. However multiple RX antena don't increase energy use, instead they just absorb more energy, and in a manner that is better than a single antena with twice the gain. This is because while one antena may be in a bad fade (a destructive multipath null), chances are much lower that both will be in a bad fade.
Multiple TX antenae can be used too, but the implementation (and decoding) gets mutch more complex as the article suggests: you need lots of computing power to grab the bits out of multiple simultaneous (but cooperating) TX sources. You also spread your energy with multiple TX antenae, which lowers overall SNR.
Based on the posters, I thought it was a movie about soap.
That's not a Half Life 2 case mod! That's a Half Life case mod with 'Half Life' crossed out and 'Half Life 2' written in in crayon!
LAPD...we treat you like a King.
"They don't even have a decent state-wide model for carpool lanes: in SoCal, carpool lanes are carpool lanes 24 hours a day. In NorCal, carpool lanes are carpool lanes only during normal rush-hour traffic, and are normal lanes otherwise."
Pshyah. That is because normal rush-hour traffic is 24 hours a day in SoCal. I've run into gridlock in LA at 4 in the freakin morning.
I believe they've said several times on the show that he is much smarter than he puts on. He gets it, he just doesn't care. It's just his schtic. I mean the Asgard see something in him.
That, and I think that Anderson wanted to play a pollar opposite of his previous roll as he had been a bit type-cast.
/* devil's advocate */ What happens when someone brings extra, forged cards with them and slips them into the audit box? Is there a simple way to make sure that there is only one card / voter without attatching reversable, unique IDs to each card (and therefore each voter)? I was thinking some sort of pseduo random sequence per machine* where inserting a random number'd card would violate the sequence. Of course you wold then have to forget / not record the order in which voters used that machine*. *or per polling place, if all the machines are networked, which would add some noise to the voter order. Still if you were the first to vote that day and the first 10 votes cast were all for candidate X, it would be failry obvious who you voted for.
a scotsman and a blind man with a hair clip hurredly fleeing from the scene.
More like the 50s style news pieces / commercials spliced into Starship Troopers (the movie not the book which are two completely different stories):
"Would you like to learn more?"
But we don't know what causes asteroids to wander our way, only that it hapens on a semi periodic basis. Perhaps as we orbit the galaxy we come accros regions with more gravitational distortions that are more likely to send stuff hurtling inwards from the oort cloud. Perhaps there is a misterious 10th planet that goes through a dense part of the oort cloud. Perhaps....
Anything that makes the system non-memoryless (i.e. statefull) and makes the events more periodic than random allows us to say that given no events so far, the probability of an event in the near future is greater/has gone up. (Extreme example: We arrive in london at some random time and don't have a watch. The fact that Big Ben hasn't rung in the last 40 minutes allows us to state that it will ring 'soon' with greater certanty than the fact it hasn't rung in the last 10.)
Of couse the fact that an asteroid doesn't hit in just one year makes the already small probability change for the next year only by an infentesmal ammount. I.e. a change of 1/50000000 --> 1/49999999 or even smaller.
It'll probably even say something allong the lines of "Forward this to 10 friends or you'll be cursed with 4 years of bad govornment."
In the one experience I've had with (human completed) scan-tron type ballots, you got to feed the ballot yourself, with an observer making sure you didn't put in more than one, but probably not close enough to make out text or barcode.
And with human-machine readable ballots you wouldn't even need to rely on open source for verification. The voter can just make sure the print-out matches his vote. It lessens the constraints placed on the ballot machine even further. You could have W. himself making a closed source version and you'd be ok, its basicaly just a glorified printer. *shudders thinking of the president trying to write code* (You still have to examine and trust the tallying machine or do it by hand.)
Make the ballot use human-computer readable text. So what the voter looks at to verify is actualy what the machine reads when tallying votes.
Uh. that was exacly my point with #2. You and a machine can read the account number on your checks. I've never seen this with the alphabet, but it probably already exists / wouldn't be too much trouble.
Voting should be important enough to businessmen for them to schedule buisness trips arround voting days, which only happen 1-2 days a year. Or rather if voting is important to them (it seems it isn't to 1/2 the eligible voters) then they are free to shedule their trips arround those days.
Granted, this isn't as possible for those in the military or people who are leaving for extended periods (months-years). But there is no reason why electronic voting is supperior to balloting by paper. The only difference is the speed at which results can be transmited/tallied. I, for one, do not wish to sacrifice security / trustworthiness of something as important as an election just for expediency.
The solution? "Lets use computers." Yet some how the assumption leaked in that the computers used to do the balloting had to be the same machines that tallied the votes as well. This is a paradigm that should be abollished as soon as possible. While we fix one problem (ballots not reflecting voter preferences properly), we introduce another (allowing increased access to the device doing the tallying).
The solution as the original comment said is to split the process. Use computers to create 'standardized' ballots, and to simplify/error check each voters choice making. And let the voter see a human readable ballot that they can confirm and turn in at a different part of the polling station.
Tallying can be done in a sepparate process, much the way scan-tron type ballots are counted today quickly and accurately.
Some thoughts / added benefits:
1. The paper trail. Voters see their votes correctly printed on paper. And the ballot machines can be used as a double check to make sure no ballots were destroyed. (added reassurance against tampering, since now it requires a coordinated attack both physical and electronic).
2. If you make the ballot human & computer readable (just like your account number on checks) you can verrify the ballot and not have to assume that the bar code the ballot machine produces matches the text.
3. If the ballot form is standardized then the voting equipment becomes commoditized. States / localities can choose the balloting and tallying equipment manufacturers to buy from independently and no one is tied to a given manufacturer for either device. They can even purchace from more than one vendor for the balloting devices. This will drive down prices, as well as letting the govornment take trustworthiness into account when purchacing equipment.
It would be great if legislatures could demand this type of system instead of letting each district try to 'roll your own' and get unknown results in terms of reliability / trustworthiness. It would also mean we wouldn't have to put anywhere near as much trust in the makers of the machines.
And when your heating / cooling bill goes through the roof because you lined your house with a great thermal conductor? I'd rather spend the cash on aerogel insulation in the walls.
Yes, but will his evidence stand up in court?
Better use something that is light, but harder to inhale, like smallish bits of paper. Or tissue.
Oh, and better remember to turn off any air circulating equipment. With a leak that slow, normal air scrubbers are probably going to have more effect on air currents than the leak.
They are called the Lagrangian points.
L1-L3 are on the earth-sun line but aren't stable. L4 and L5, which lie in earth orbit +/- 60 dgrees, are! And all of them orbit earth in a year. (of course, the centers of the orbits aren't earth, they're close to the sun, but they still take a circular path arround us every year).
NASA is currently using L1 (SOHO, ACE) with plans to use L2 for deep space observations.
Of course none of these are practical for a stationary elivator because the earth spins faster than 1/year (unless the end 'dangles' over the equator and moves east-west). And they are too far away.
No, but I cun relieve you of them thar bags of money. *ch-chick*
The article says he got this on the first day of his job after he was hired.
Yes. He's entering into a deal where they have to give him money on a regular schedule. If anyone should be giving anyone else a credit report, the company should be giving one to Rick. No one ever credit checks their bank before borrowing a home loan.
"How can I be sure you'll pay me back for my 2weeks/month worth of work if you have poor credit?"
There is a flaw that makes this not a true turring test: The competition is set up so that your program only plays itself. Your program does not get a chance to play other programs or real humans. So it is sorta like a turring test where the computer is only allowed to talk with itself.
The true guts of the turring test come from telling if an unknown entity is a human or machine through its interaction with a known human, not its interaction with a copy of itself or another machine.
(sometimes the ineracting human decides, and sometimes an over the sholder observer tries to decide. depends on the setup)
To realy make it work, you need programs that can be input an arbitrary sequence of moves that can then decide the next move. Programs and humans can then be run against humans (and even other programs) randomly. The detectors job would be to decide wich players were human instead of which sets of games were human.