H.R. 811 is also very well considered in how it handles problems with the paper trail. I had concerns before I actually read the text of the bill when I heard that the paper trail was to be taken as the vote of record in the event of a discrepancy. However, the bill does consider the possibility of damage or other problems affecting the paper trail.
The only thing I find lacking in the bill (regarding the paper trail, anyway) is that it is somewhat lax about how the paper trail is handled: each machine's paper trail ballots should be considered and compared to that machine individually in any audit, up to and including a full recount of a state's ballots. By dividing up the ballots in such small parcels and comparing them to a "second opinion", i.e., the machine, which is actually good at counting things compared to people, inaccuracies can be further reduced during the recount process. The bill only requires the paper trail to remain at the polling place until needed, and it's not clear to me that the single-machine association is kept at all times and not just when some of the paper ballots are damaged.
Of course, I'm also disappointed that they didn't go with full open source. We don't need proprietary shenanigans, especially when I'm sure there are tons of open source developers who would be happy to develop software for a machine from a company that was only interested in producing the hardware.
Actually, it says that it can get the spacecraft to Mars in a week and can stop once it gets there. But it doesn't claim that anyone will survive the impact.
"Defective by design" would imply that Apple and/or AT&T intentionally designed the iPhone in this manner for some nefarious purpose, such as trapping their customers into massive fees for services they unwittingly used. The term normally refers to DRM, where the ability to use a product in a manner that one would otherwise be able to is intentionally restricted due to the whims of the distributor.
In this case, I think it's probably just poor design, but not intentionally so.
The more people that sign up for Schedules Direct, the lower their costs will be in the future (or something along those lines). They've already stated that their intention is to have a much lower fee for listings after the first quarter, but that the $15 cost for the first three months was necessary since their organization is just starting up.
Personally, I find the listings useful enough that I'd (just barely) pay the $5 a month, but I would hold them to a much higher QoS if I had to keep paying that much.
BTW, one of the SD guys mentioned that they found out that Tribune Media Services, and other TV schedule aggregators, do a lot more than just put together already-available data. They have to cross-reference syndication feeds with local schedules, they have to come up with episode descriptions, and in general, there's work and some original content arising from that. Even local stations have no idea what episode they're showing on a particular day - they just get the episodes from the distributor, possibly with promos, and then run them. TMS already knows how to do this stuff, and they had the infrastructure to distribute it already, but they weren't interested in managing the business of selling the listings. SD acts as a broker in that sense, paying TMS for the listings and collecting the individual fees from the users.
I've been totally confused as to whether this patent bill is a good thing or not, primarily because Howard Berman and Howard Coble, the Antichrist and the False Prophet to fair use and personal liberty involving legally purchased copyrighted works, have had a prominent role in bringing this bill through the House. On the other hand, Public Knowledge and other groups seem pretty upbeat on it (although I didn't sit and watch C-SPAN all day today to see what happened with amendments and such). So what am I to think? Is this a Good Thing (tm), or is there some hidden provision that will eventually screw us all yet again?
Wouldn't it make more sense to do your development on an isolated network, and then once you've figured out how to break into the botnet, release your results to the public so that everybody can have a field day?
I think there's a hair that needs splitting here. Access isn't a right, but once you've secured unfettered access, fair use of the material is a right, closely connected to free speech.
I had this weird dream about a week ago, where I ran into these mask-wearing aliens whose masks were apparently life support devices (despite them being more like theater masks than conventional breathing masks). The reason they needed the masks was because 67 million years ago, they had colonized Earth, whose atmosphere at the time could support their form of life, but they needed to do some terraforming. Unfortunately, the terraforming resulted in catastrophic changes to the atmosphere that made the planet inhospitable to them. Coincidentally, those atmospheric changes also killed off the dinosaurs.
Anyway, I'm thinking of starting my own cult based on that dream. Anybody here have any advice?
The only good part in that movie was when that one guy took a helicopter rotor blade to the chest. Even if the way it happened in the movie was actually impossible.
My brain exploded all over my monitor after reading that question. Since this is usually the first place I go for legal advice, I ask you: how would you go about starting a personal injury suit against the guy who asked the question?
The summary sentences in many cases don't describe the whole beef, since all you did was take the first sentence from each numbered point. I encourage folks to read the first linked article.
1. Who cares. As long as the device functions properly from empirical evidence, who cares whether the code looks like shite or was tested in-house sufficiently.
2. The machine obviously doesn't correctly calculate the average, but it's not clear that the way it *does* combine these values is necessarily wrong. The sum of the values is still a weighted combination of the values where the total weight equals 1.
3. The summary line you quoted doesn't actually state the problem. I'm not an expert on this kind of device, though, so I can't tell whether the loss of precision they're talking about (supposedly down to 4 or 3 bits for one of the devices) has a substantial impact on the answer returned.
4. In the very unlikely event that the firmware on the device were somehow corrupted, it's possible that the processor would simply halt upon receiving an illegal instruction with no interrupt possible. This, of course, depends on the processor. On some processors, this would be counted as a double exception, which would cause a processor halt. In any case, the device would always fail completely, and likely would return no answer. Also, watchdog failure would likely result in a nonsensical answer or no answer at all.
5. This probably is bad design, but it's likely that in such a case, the device would provide a nonsensical reading if one of the devices failed. In fact, it would be more likely that a partial, undetectable failure of one of those devices would cause an in-range bad answer than for a total failure of one of those components to do so.
6. As long as the diagnostic routine calibrates the device so that spurious noise from one of its devices is normalized, this shouldn't be a problem ("arbitrary, favorable readings" is an arbitrary and favorable-for-the-defense term, when it actually just provides the saturated answer for an out-of-bounds measurement). If the device is providing spurious readings often enough to dramatically shift the calibration, then this should be flagged. It's unclear whether any device of this type returns occasional spurious readings, but even if they do, that doesn't necessarily mean they're broken, and some noise can be averaged away.
7. As long as the device is powered up under proper conditions (zero airflow in the tube, most likely, though this isn't discussed), this should be fine unless the device is used without powering it down for a very extended length of time (as in, long enough for wear and tear to cause alteration of the physical characteristics of the machine). It's not clear whether by "positive number" they mean zero, or greater than zero, and it's not clear that a spurious below-baseline airflow reading indicates anything beyond varying environmental conditions.
8. It's not mentioned whether a saturated reading from these instruments results in a failure of the device to calibrate later on.
9. WTF does this mean? The article didn't provide any more details than that, either.
10. This seems like a legitimate complaint, since a large number of spurious responses - but less than numerous consecutive failures - could indicate imminent failure of the device.
11. This section was either poorly described or just wrong. It sounds like the free-running timer loop signals the main loop to do something - whatever's next on the list - every 8.192 ms. But if I'm getting what they're saying, then it sounds like this is the right way to design the code. If you enter a tight sampling loop, then you can rely on the loop executing every 8.192 ms (or a known/calculable/measurable multiple thereof). If you cause the interrupt to occur 8.192 ms after the last iteration of that loop completed, then you also have to calculate the time to execute the loop's code and add it to the 8.192 ms. However, the device app
I suppose the question is whether high-dollar fingerprint scanners work any better than run-of-the-mill ones. They didn't exactly have a huge cash outlay for the one they used on that episode.
I saw a show not too long ago on Discovery or one of those channels talking about a home security system that used a fingerprint scanner to permit access to the secure portion of the house. (Brief article explaining this so I don't have to is here.)
Anyway, the guy claimed that the fingerprint scanner was able to reject a finger severed from the hand of the owner (although, TBH, once the crook has chopped off my finger to try it, whether or not it actually worked is the least of my worries).
The Ohio statute says "probable cause". An officer can arrest you on probable cause as well. In fact, an officer can detain you on lesser grounds than that (reasonable suspicion).
There are really two fully separate questions raised by this case. One, did the store have the right to detain the shopper? And two, did the shopper subsequently violate a law by refusing to show his identification to an officer?
Unfortunately, right now, we only have the shopper's story to go by. Of course, we have courts to suss these things out, so eventually, more sides of this situation will come out.
But for the moment, let's go by his version of events:
1) Did the store have the right to detain the shopper? Ohio state law grants to store owners the right to detain someone whom they have probable cause to suspect of stealing from their establishment, if they do so for the purpose of calling the cops. According to this law, they do not have the right, by themselves, to search someone suspected of shoplifting. The question arises whether or not a refusal to consent to a search is sufficient to establish probable cause, which is better answered by actual lawyers who understand the tons of case law most likely established on this question.
If the store was in the wrong by attempting to prevent him from leaving, then it's possible that they committed a crime in preventing him from leaving (unlawful restraint, a third-degree misdemeanor).
2) Did the shopper violate the law by refusing to provide his driver's license to the responding officer? In this case, Ohio state law specifically requires a person to provide their name, address, and date of birth to an officer when they are a witness or a suspect in a crime. In this case, there were two possibilities for crimes under investigation - the possible shoplifting or the unlawful restraint - which would have caused the shopper to become responsible under this statute. The question is then whether the case law predicating this statute (Hiibel v. 6th Circuit...) implicitly allows an officer to require that the person present a driver's license. That's again one for the courts.
(Note that lying to the officer by stating you don't have a driver's license, or don't have yours on you, when you actually do, is likely to make things worse.)
Slashdot Mirror
You make it sound like Americans are the only ones eligible for the Darwin Award.
I thought Russia owned the whole damn thing.
I'm sure the Northwest Passage Cruise Line vacation scam spams will begin soon.
On the other hand, when's the last time that SCO tried a legal strategy that did work?
They finally listened to his comments! Now with wireless and *more* space than a Nomad.
Plus, it has paradox-absorbing crumple zones.
H.R. 811 is also very well considered in how it handles problems with the paper trail. I had concerns before I actually read the text of the bill when I heard that the paper trail was to be taken as the vote of record in the event of a discrepancy. However, the bill does consider the possibility of damage or other problems affecting the paper trail.
The only thing I find lacking in the bill (regarding the paper trail, anyway) is that it is somewhat lax about how the paper trail is handled: each machine's paper trail ballots should be considered and compared to that machine individually in any audit, up to and including a full recount of a state's ballots. By dividing up the ballots in such small parcels and comparing them to a "second opinion", i.e., the machine, which is actually good at counting things compared to people, inaccuracies can be further reduced during the recount process. The bill only requires the paper trail to remain at the polling place until needed, and it's not clear to me that the single-machine association is kept at all times and not just when some of the paper ballots are damaged.
Of course, I'm also disappointed that they didn't go with full open source. We don't need proprietary shenanigans, especially when I'm sure there are tons of open source developers who would be happy to develop software for a machine from a company that was only interested in producing the hardware.
Actually, it says that it can get the spacecraft to Mars in a week and can stop once it gets there. But it doesn't claim that anyone will survive the impact.
"Defective by design" would imply that Apple and/or AT&T intentionally designed the iPhone in this manner for some nefarious purpose, such as trapping their customers into massive fees for services they unwittingly used. The term normally refers to DRM, where the ability to use a product in a manner that one would otherwise be able to is intentionally restricted due to the whims of the distributor.
In this case, I think it's probably just poor design, but not intentionally so.
The more people that sign up for Schedules Direct, the lower their costs will be in the future (or something along those lines). They've already stated that their intention is to have a much lower fee for listings after the first quarter, but that the $15 cost for the first three months was necessary since their organization is just starting up.
Personally, I find the listings useful enough that I'd (just barely) pay the $5 a month, but I would hold them to a much higher QoS if I had to keep paying that much.
BTW, one of the SD guys mentioned that they found out that Tribune Media Services, and other TV schedule aggregators, do a lot more than just put together already-available data. They have to cross-reference syndication feeds with local schedules, they have to come up with episode descriptions, and in general, there's work and some original content arising from that. Even local stations have no idea what episode they're showing on a particular day - they just get the episodes from the distributor, possibly with promos, and then run them. TMS already knows how to do this stuff, and they had the infrastructure to distribute it already, but they weren't interested in managing the business of selling the listings. SD acts as a broker in that sense, paying TMS for the listings and collecting the individual fees from the users.
How responsive would a real dragon be to being whacked on the head with a blunt object?
Seems to me you'd only get one chance to find out.
Harvey Birdman: I'll take the case!
I've been totally confused as to whether this patent bill is a good thing or not, primarily because Howard Berman and Howard Coble, the Antichrist and the False Prophet to fair use and personal liberty involving legally purchased copyrighted works, have had a prominent role in bringing this bill through the House. On the other hand, Public Knowledge and other groups seem pretty upbeat on it (although I didn't sit and watch C-SPAN all day today to see what happened with amendments and such). So what am I to think? Is this a Good Thing (tm), or is there some hidden provision that will eventually screw us all yet again?
Wouldn't it make more sense to do your development on an isolated network, and then once you've figured out how to break into the botnet, release your results to the public so that everybody can have a field day?
I think there's a hair that needs splitting here. Access isn't a right, but once you've secured unfettered access, fair use of the material is a right, closely connected to free speech.
Zapp Brannigan: Double yes. Guilty! I will now carry out the punishment. Kif, my gun.
I had this weird dream about a week ago, where I ran into these mask-wearing aliens whose masks were apparently life support devices (despite them being more like theater masks than conventional breathing masks). The reason they needed the masks was because 67 million years ago, they had colonized Earth, whose atmosphere at the time could support their form of life, but they needed to do some terraforming. Unfortunately, the terraforming resulted in catastrophic changes to the atmosphere that made the planet inhospitable to them. Coincidentally, those atmospheric changes also killed off the dinosaurs.
Anyway, I'm thinking of starting my own cult based on that dream. Anybody here have any advice?
The only good part in that movie was when that one guy took a helicopter rotor blade to the chest. Even if the way it happened in the movie was actually impossible.
Dear Slashdot,
My brain exploded all over my monitor after reading that question. Since this is usually the first place I go for legal advice, I ask you: how would you go about starting a personal injury suit against the guy who asked the question?
Thanks!
The summary sentences in many cases don't describe the whole beef, since all you did was take the first sentence from each numbered point. I encourage folks to read the first linked article.
1. Who cares. As long as the device functions properly from empirical evidence, who cares whether the code looks like shite or was tested in-house sufficiently.
2. The machine obviously doesn't correctly calculate the average, but it's not clear that the way it *does* combine these values is necessarily wrong. The sum of the values is still a weighted combination of the values where the total weight equals 1.
3. The summary line you quoted doesn't actually state the problem. I'm not an expert on this kind of device, though, so I can't tell whether the loss of precision they're talking about (supposedly down to 4 or 3 bits for one of the devices) has a substantial impact on the answer returned.
4. In the very unlikely event that the firmware on the device were somehow corrupted, it's possible that the processor would simply halt upon receiving an illegal instruction with no interrupt possible. This, of course, depends on the processor. On some processors, this would be counted as a double exception, which would cause a processor halt. In any case, the device would always fail completely, and likely would return no answer. Also, watchdog failure would likely result in a nonsensical answer or no answer at all.
5. This probably is bad design, but it's likely that in such a case, the device would provide a nonsensical reading if one of the devices failed. In fact, it would be more likely that a partial, undetectable failure of one of those devices would cause an in-range bad answer than for a total failure of one of those components to do so.
6. As long as the diagnostic routine calibrates the device so that spurious noise from one of its devices is normalized, this shouldn't be a problem ("arbitrary, favorable readings" is an arbitrary and favorable-for-the-defense term, when it actually just provides the saturated answer for an out-of-bounds measurement). If the device is providing spurious readings often enough to dramatically shift the calibration, then this should be flagged. It's unclear whether any device of this type returns occasional spurious readings, but even if they do, that doesn't necessarily mean they're broken, and some noise can be averaged away.
7. As long as the device is powered up under proper conditions (zero airflow in the tube, most likely, though this isn't discussed), this should be fine unless the device is used without powering it down for a very extended length of time (as in, long enough for wear and tear to cause alteration of the physical characteristics of the machine). It's not clear whether by "positive number" they mean zero, or greater than zero, and it's not clear that a spurious below-baseline airflow reading indicates anything beyond varying environmental conditions.
8. It's not mentioned whether a saturated reading from these instruments results in a failure of the device to calibrate later on.
9. WTF does this mean? The article didn't provide any more details than that, either.
10. This seems like a legitimate complaint, since a large number of spurious responses - but less than numerous consecutive failures - could indicate imminent failure of the device.
11. This section was either poorly described or just wrong. It sounds like the free-running timer loop signals the main loop to do something - whatever's next on the list - every 8.192 ms. But if I'm getting what they're saying, then it sounds like this is the right way to design the code. If you enter a tight sampling loop, then you can rely on the loop executing every 8.192 ms (or a known/calculable/measurable multiple thereof). If you cause the interrupt to occur 8.192 ms after the last iteration of that loop completed, then you also have to calculate the time to execute the loop's code and add it to the 8.192 ms. However, the device app
Forget it. Eliot Spitzer has too much ambition to stay where he's needed most.
I suppose the question is whether high-dollar fingerprint scanners work any better than run-of-the-mill ones. They didn't exactly have a huge cash outlay for the one they used on that episode.
I saw a show not too long ago on Discovery or one of those channels talking about a home security system that used a fingerprint scanner to permit access to the secure portion of the house. (Brief article explaining this so I don't have to is here.)
Anyway, the guy claimed that the fingerprint scanner was able to reject a finger severed from the hand of the owner (although, TBH, once the crook has chopped off my finger to try it, whether or not it actually worked is the least of my worries).
Finally, a virtual place that kids can keep their virtual weed when they're at school.
The Ohio statute says "probable cause". An officer can arrest you on probable cause as well. In fact, an officer can detain you on lesser grounds than that (reasonable suspicion).
There are really two fully separate questions raised by this case. One, did the store have the right to detain the shopper? And two, did the shopper subsequently violate a law by refusing to show his identification to an officer?
Unfortunately, right now, we only have the shopper's story to go by. Of course, we have courts to suss these things out, so eventually, more sides of this situation will come out.
But for the moment, let's go by his version of events:
1) Did the store have the right to detain the shopper? Ohio state law grants to store owners the right to detain someone whom they have probable cause to suspect of stealing from their establishment, if they do so for the purpose of calling the cops. According to this law, they do not have the right, by themselves, to search someone suspected of shoplifting. The question arises whether or not a refusal to consent to a search is sufficient to establish probable cause, which is better answered by actual lawyers who understand the tons of case law most likely established on this question.
If the store was in the wrong by attempting to prevent him from leaving, then it's possible that they committed a crime in preventing him from leaving (unlawful restraint, a third-degree misdemeanor).
2) Did the shopper violate the law by refusing to provide his driver's license to the responding officer? In this case, Ohio state law specifically requires a person to provide their name, address, and date of birth to an officer when they are a witness or a suspect in a crime. In this case, there were two possibilities for crimes under investigation - the possible shoplifting or the unlawful restraint - which would have caused the shopper to become responsible under this statute. The question is then whether the case law predicating this statute (Hiibel v. 6th Circuit...) implicitly allows an officer to require that the person present a driver's license. That's again one for the courts.
(Note that lying to the officer by stating you don't have a driver's license, or don't have yours on you, when you actually do, is likely to make things worse.)