OK, so we have this verifier. It swaps everything out but itself. It verifies the system, and it all comes up good.
Now what? It has to swap everything back IN. Including potential malware which made no attempt to evade the verifier. Sure, it prevented the malware from running for a time. But only at the cost of preventing anything else from running either. That's less than useful.
What evidence is there that it is a software bug? I mean, sure, a lot of us are computer programmers so we know how unreliable the software can be... but a lot can go wrong in mechanical systems as well. And so far, I haven't seen a bit of evidence which points to a software fault over a mechanical fault. (and some, like the pedal being physically stuck, which points the other way)
As for the addition bug found in the Pathfinder... I ran into a similar bug once. A simple addition was returning the wrong answer, in the middle of a decryption routine, which would then fail. It was, as was the Pathfinder bug, interrupt related. But, in this case, it was the CPU itself which would screw up the add (as confirmed by the manufacturer's errata sheet); hardware, not software. Sure, software's unreliable. But it's not the only thing which is.
It is ironic that we (the world's technologists - which is us, fellow Slashdaughters) are creating machines that will eventually be able to replace humans in the workplace at the same time that the world's population is entering the vertical section of a 'hockey stick' growth curve.
That's OK, our evil counterparts in the military-industrial complex are working on population-reduction measures as we speak.
The issues aren't separate. Once you ghettoize porn into an.xxx domain, it makes it much easier (both technically and politically) to block. In fact, it's likely the default would become to block the.xxx domain. Not just for corporations, but for ISPs (at the urging of the more prudish members of the community). Oh, and when the company filters websites at work it IS censorship, just on a smaller scale (and perhaps more justifiable).
...hanging around the water cooler, bitching about management, sexually harassing other co-workers, having non-work-related discussions... in short, we'll end up with Bender.
The things they found out aren't things most people have any reason to keep secret. OK, if you see that most of my Facebook friends went to Cowpie High or Mediocre State University, and you'll realize that I, too, probably went to Cowpie High and Mediocre State. So what? Mediocre State is on my (sometimes publicly available) resume, and it's not like its any secret that I went to Cowpie High either. (and yes, the school's actual nickname among the students was that)
Much more interesting would be if they could figure things which people are trying to keep private. Where they buried the bodies of their "missing" parents, if they're gay but in the closet (I think there already was an article about that over a year ago, though), membership in the Secret Order of Inquisitors and Torturers (friending Dick Cheney is the giveaway here), etc.
You can walk right through security (airport, border, corporate) with a microSD card in your pocket and nobody blinks an eye. Trying to "smuggle" a MicroSD card through is more likely to result in you getting caught and treated badly (even if it isn't even illegal). If the data on the MicroSD card is what you're trying to hide, a better spy device would be a trick card... say, which was internally partitioned into two cards with some very obscure way (SW or HW) of switching between them. Put innocuous data on one side, stick it in your camera, phone, music player, whatever. Even if the goons search the card, that's all they find. Short the right contacts or send the right command, and get access to the "evil" data.
...does this seem like absolutely nothing groundbreaking at all. OK, we can transmit information by pulsing LEDs. People have been doing that for years. The fact you don't put an optical fiber in front of it doesn't seem all that interesting.
The problem isn't this bill, which won't pass. The problem is that bad ideas like this, once introduced, have a life of their own. They keep getting reintroduced until they do pass. (good ideas, on the other hand, get shelved and are never heard from again).
They've already assaulted baked goods by banning trans-fats (certain baked goods need shortening for texture). Ruining everything else, even with a watered-down anti-salt bill, is now inevitable.
Put these same kids on an existing program that is a year late and already has a team of 20 programmers working on it. Get back to me in 6 months telling me just how fine things are.
In that case, I suspect firing the right 5-7 people (some of them programmers, but not all of them) would get the job done faster.
Re:Reminds me of broadband internet in the beginni
on
Gas Wants To Kill the Wind
·
· Score: 2, Insightful
Seems like it's starting to happen, but it's slow and you miss out on the water cooler conversations and therefore promotions. Is telecommuting ever going to happen for real?
If an employee can work from home, often enough he can also work from Bangalore. That's a serious limit to the growth of telecommuting.
There's a little thing in the EULA that states "If the media is faulty, then we'll replace that at your cost for about £15. There are no other warranties of fitness.".
A warranty of fitness isn't needed to cover the case of a product that doesn't work at all; that would be a warranty of merchantability. A warranty of fitness says that the product works for what the buyer thinks it does; a warranty of merchantability says the product works for what the seller says it does.
WABC-DT transmits from the Empire State Building on VHF channel 7. Almost every Cablevision subscriber should at least have a shot at getting it -- and the other NY majors --- with not much of an antenna if they have a decent (5th generation or newer) tuner. And I don't know about Cablevision, but the quality will be a LOT better than what RCN (which has not been cut off) is putting out.
The court clearly didn't understand the language. They accepted the idea that because Echostar refers to PID filtering as "parsing", they were admitting that PID filtering was "parsing video and audio data". Which is just wrong; the video and audio data are at a different level of abstraction from the container which was being parsed. The court simply doesn't understand the word "parsing" and uses it as if it has some universal meaning outside the context of what is being parsed.
It doesn't help that what TiVo actually meant by "parsing...from the broadcast data" appears to be "separating" -- that is, they did mean the stream filtering process, though they stated that badly.
Basically with the court's very broad reading of the patent, there's no way to build a DVR without infringing the software patents, as TiVo can always "map" your implementation onto their (extremely broad) claims. Which is another slap in the face for the patent apologists who like to claim that you can always "invent around" a patent and that patents cover specific implementations, not ideas.
The "correct" techniques were developed for keyboards which were basically like this: Old manual typewriter
Note the huge vertical spacing between the rows. Note the long travel of the keys, and consider the need to strike them consistently (or typing quality suffered). Consider the amount of force it took to strike a key. Is your computer keyboard much like that? I suspect not. So while there may be good techniques for computer keyboards, they're unlikely to be similar to the tried and true "correct" techniques a typing instructor will torture you with. If you're at 110wpm, you're fine for anything but a speed typing contest. Even allowing for some degree of bragging.
With all those references to MPEG in the patent, couldn't they simply use Ogg Theora (or some other codec) to circumvent the patent?
Probably not as a practical matter. For ATSC (and DVB), the stream comes in as MPEG, which would mean you'd have to transcode to use another codec.
Interestingly, MythTV doesn't seem to infringe this patent. The essential claims have steps where the streams are disassembled into video and audio, stored, and re-assembled on playback. MythTV doesn't work this way; it stores the program streams with the video and audio still interleaved, and disassembles on playback. I'm not sure why EchoStar couldn't use a similar technique; unfortunately, it's possible that the courts interpret the patent broadly and figure that difference doesn't matter.
No, the problem with Myth is that it's just awful. Configuration is a PITA, the various UIs all have largely the same faults, you need to hunt around for documentation to find out simple stuff like what keys do what, you get cryptic error messages, and my experience of the developers thus far (albeit it was one developer, but apparently he was typical of the group) was extremely negative, with the team being defensive and, actually, rather proud of the fact that the product is barely usable.
Yes, configuration is a pain. That's hard to avoid given the diversity of hardware MythTV supports. The UI seems perfectly competent for use as a DVR, though, once you have a remote control set up (which is also a pain, but it's a "do it once" kind of thing). My non-techie wife can use it, anyway. And I haven't seen cryptic error messages in quite a while.
You're right about the team, though. Bug reports (complete with detailed information on how to reproduce) get closed/no action, and feature requests just get a "no" (the latest one I submitted was one for a jump point -- something which works anywhere in the frontend-- to return to idle, so the system will shut off. It's less than a 10 line change; IIRC two lines of code and a row in the database. Rejected because there's a kludgy (read: less reliable) way to do it with irexec).
Actually, I'd say they have almost a 100% response rate. They ask the copyright holder, "May I please have a copy of your content?" and in most cases, they receive a response within 500 milliseconds saying, "Sure! Here it is!"
But then to store the content, that's another copy, not the same as the bits transmitted over the wire. There may be other copies made in the process of making that copy, which may or may not count. Copyright is completely and irretrievably broken where computers and the web are concerned; either you have to make up an implicit license theory out of whole cloth (and then finding the boundaries of it is just a guessing came), or using the web involves wholesale copyright violation even if only legitimate sites are visited.
I don't think there's an effective way of throttling I/O load per user in OS X. But if he's always accessing the same designer's machine, why not just fool him? Change the name of the designer's actual machine. Set up another machine with the old name. Set up whatever sync processes you need between the two. Kid will log into the replacement machine and not even know he's doing anything different.
Now, next time perhaps lawyers will think twice before defending those nasty pirates against the might of the RIAA.
This is like freezing the funds or or prosecuting mob lawyers; yeah, it's really tempting to do, but it's not so much a slippery slope as a sheer drop into injustice. It effectively makes it far more difficult for future defendants for similar actions to obtain legal counsel.
It appears Apple's problem with the apps isn't with what they do but with how they do it; namely, using non-public frameworks. There probably isn't a way to do it using public frameworks, though (on Mac OS X, you need to use the private Apple80211.framework, not sure about iPhone OS X).
There are "bright spots" visible by satellite within North Korea. They are believed to correspond to enclaves of the wealthy elites. What's that you ask? How can they have "wealthy elites" within a Marxist utopian state? You may be nearing an answer to your own question.
Slashdot Mirror
OK, so we have this verifier. It swaps everything out but itself. It verifies the system, and it all comes up good.
Now what? It has to swap everything back IN. Including potential malware which made no attempt to evade the verifier. Sure, it prevented the malware from running for a time. But only at the cost of preventing anything else from running either. That's less than useful.
What evidence is there that it is a software bug? I mean, sure, a lot of us are computer programmers so we know how unreliable the software can be... but a lot can go wrong in mechanical systems as well. And so far, I haven't seen a bit of evidence which points to a software fault over a mechanical fault. (and some, like the pedal being physically stuck, which points the other way)
As for the addition bug found in the Pathfinder... I ran into a similar bug once. A simple addition was returning the wrong answer, in the middle of a decryption routine, which would then fail. It was, as was the Pathfinder bug, interrupt related. But, in this case, it was the CPU itself which would screw up the add (as confirmed by the manufacturer's errata sheet); hardware, not software. Sure, software's unreliable. But it's not the only thing which is.
That's OK, our evil counterparts in the military-industrial complex are working on population-reduction measures as we speak.
The issues aren't separate. Once you ghettoize porn into an .xxx domain, it makes it much easier (both technically and politically) to block. In fact, it's likely the default would become to block the .xxx domain. Not just for corporations, but for ISPs (at the urging of the more prudish members of the community). Oh, and when the company filters websites at work it IS censorship, just on a smaller scale (and perhaps more justifiable).
...hanging around the water cooler, bitching about management, sexually harassing other co-workers, having non-work-related discussions... in short, we'll end up with Bender.
The things they found out aren't things most people have any reason to keep secret. OK, if you see that most of my Facebook friends went to Cowpie High or Mediocre State University, and you'll realize that I, too, probably went to Cowpie High and Mediocre State. So what? Mediocre State is on my (sometimes publicly available) resume, and it's not like its any secret that I went to Cowpie High either. (and yes, the school's actual nickname among the students was that)
Much more interesting would be if they could figure things which people are trying to keep private. Where they buried the bodies of their "missing" parents, if they're gay but in the closet (I think there already was an article about that over a year ago, though), membership in the Secret Order of Inquisitors and Torturers (friending Dick Cheney is the giveaway here), etc.
You can walk right through security (airport, border, corporate) with a microSD card in your pocket and nobody blinks an eye. Trying to "smuggle" a MicroSD card through is more likely to result in you getting caught and treated badly (even if it isn't even illegal). If the data on the MicroSD card is what you're trying to hide, a better spy device would be a trick card... say, which was internally partitioned into two cards with some very obscure way (SW or HW) of switching between them. Put innocuous data on one side, stick it in your camera, phone, music player, whatever. Even if the goons search the card, that's all they find. Short the right contacts or send the right command, and get access to the "evil" data.
...does this seem like absolutely nothing groundbreaking at all. OK, we can transmit information by pulsing LEDs. People have been doing that for years. The fact you don't put an optical fiber in front of it doesn't seem all that interesting.
The problem isn't this bill, which won't pass. The problem is that bad ideas like this, once introduced, have a life of their own. They keep getting reintroduced until they do pass. (good ideas, on the other hand, get shelved and are never heard from again).
They've already assaulted baked goods by banning trans-fats (certain baked goods need shortening for texture). Ruining everything else, even with a watered-down anti-salt bill, is now inevitable.
In that case, I suspect firing the right 5-7 people (some of them programmers, but not all of them) would get the job done faster.
If an employee can work from home, often enough he can also work from Bangalore. That's a serious limit to the growth of telecommuting.
A warranty of fitness isn't needed to cover the case of a product that doesn't work at all; that would be a warranty of merchantability. A warranty of fitness says that the product works for what the buyer thinks it does; a warranty of merchantability says the product works for what the seller says it does.
WABC-DT transmits from the Empire State Building on VHF channel 7. Almost every Cablevision subscriber should at least have a shot at getting it -- and the other NY majors --- with not much of an antenna if they have a decent (5th generation or newer) tuner. And I don't know about Cablevision, but the quality will be a LOT better than what RCN (which has not been cut off) is putting out.
The court clearly didn't understand the language. They accepted the idea that because Echostar refers to PID filtering as "parsing", they were admitting that PID filtering was "parsing video and audio data". Which is just wrong; the video and audio data are at a different level of abstraction from the container which was being parsed. The court simply doesn't understand the word "parsing" and uses it as if it has some universal meaning outside the context of what is being parsed.
It doesn't help that what TiVo actually meant by "parsing...from the broadcast data" appears to be "separating" -- that is, they did mean the stream filtering process, though they stated that badly.
Basically with the court's very broad reading of the patent, there's no way to build a DVR without infringing the software patents, as TiVo can always "map" your implementation onto their (extremely broad) claims. Which is another slap in the face for the patent apologists who like to claim that you can always "invent around" a patent and that patents cover specific implementations, not ideas.
The "correct" techniques were developed for keyboards which were basically like this:
Old manual typewriter
Note the huge vertical spacing between the rows. Note the long travel of the keys, and consider the need to strike them consistently (or typing quality suffered). Consider the amount of force it took to strike a key. Is your computer keyboard much like that? I suspect not. So while there may be good techniques for computer keyboards, they're unlikely to be similar to the tried and true "correct" techniques a typing instructor will torture you with. If you're at 110wpm, you're fine for anything but a speed typing contest. Even allowing for some degree of bragging.
Probably not as a practical matter. For ATSC (and DVB), the stream comes in as MPEG, which would mean you'd have to transcode to use another codec.
Interestingly, MythTV doesn't seem to infringe this patent. The essential claims have steps where the streams are disassembled into video and audio, stored, and re-assembled on playback. MythTV doesn't work this way; it stores the program streams with the video and audio still interleaved, and disassembles on playback. I'm not sure why EchoStar couldn't use a similar technique; unfortunately, it's possible that the courts interpret the patent broadly and figure that difference doesn't matter.
Yes, configuration is a pain. That's hard to avoid given the diversity of hardware MythTV supports. The UI seems perfectly competent for use as a DVR, though, once you have a remote control set up (which is also a pain, but it's a "do it once" kind of thing). My non-techie wife can use it, anyway. And I haven't seen cryptic error messages in quite a while. You're right about the team, though. Bug reports (complete with detailed information on how to reproduce) get closed/no action, and feature requests just get a "no" (the latest one I submitted was one for a jump point -- something which works anywhere in the frontend-- to return to idle, so the system will shut off. It's less than a 10 line change; IIRC two lines of code and a row in the database. Rejected because there's a kludgy (read: less reliable) way to do it with irexec).
But then to store the content, that's another copy, not the same as the bits transmitted over the wire. There may be other copies made in the process of making that copy, which may or may not count. Copyright is completely and irretrievably broken where computers and the web are concerned; either you have to make up an implicit license theory out of whole cloth (and then finding the boundaries of it is just a guessing came), or using the web involves wholesale copyright violation even if only legitimate sites are visited.
It's only a tautology if you make that particular assumption, which I do not.
I don't think there's an effective way of throttling I/O load per user in OS X. But if he's always accessing the same designer's machine, why not just fool him? Change the name of the designer's actual machine. Set up another machine with the old name. Set up whatever sync processes you need between the two. Kid will log into the replacement machine and not even know he's doing anything different.
Now, next time perhaps lawyers will think twice before defending those nasty pirates against the might of the RIAA.
This is like freezing the funds or or prosecuting mob lawyers; yeah, it's really tempting to do, but it's not so much a slippery slope as a sheer drop into injustice. It effectively makes it far more difficult for future defendants for similar actions to obtain legal counsel.
It appears Apple's problem with the apps isn't with what they do but with how they do it; namely, using non-public frameworks. There probably isn't a way to do it using public frameworks, though (on Mac OS X, you need to use the private Apple80211.framework, not sure about iPhone OS X).
That's the Korean East Sea to you, Comrade.
A simple "yes" would have been fine.
Certicom has a bunch of patents on ECC, though. RSA is unencumbered as the patents on modular exponentiation in a cryptosystem actually expired.