But Childs really lost context: It was not his network. He had no business trying to enforce anything. The SF IT department may run their networks as stupidly as they chose, and while this may lead to criminal and civil liability on their part, it does not lead to any accountability towards Childs.
If he had turned over the passwords on the conference call, and someone in the room full of unauthorized people had used them in an unauthorized manner, instead of 4 years with the possibility of parole, he'd be looking at 20 years in a federal pound-you-in-the-ass prison with no possibility of parole for "aiding people in unauthorized computer access", a federal crime.
It really was a no-win situation, particularly since the city's immediate reaction was to sequester him away from all authorized persons until the mayor could do his grandstanding in front of the press.
Bull shit he didn't booby trap it. The network was configured to run from RAM at his design and he was the only person with the configuration. Power to any site will go out and that site will be down until the "hero" comes to save the day.
So what exactly are the phone numbers for the House Intelligence Committee and the Senate Intelligence Committee, so that, you know, future whistle blowers can call them up, and not end up like Snowden?
Given that Atrix dock is cheaper than the target price and can be hacked (and we did just that at Google when Hexxeh interned there), why not just make a plastic case with a couple of connectors on it, and clip the Pi into it?
In general, German cars are known to exaggerate speed by up to 10% in order to guarantee compliance with European law (ECE-R39).
In the U.S., it's been historically common to "detune" speedometers in rental cars to exaggerate the speed, and therefore clock up additional miles which are then charged to the renter. It's also been historically common to roll back odometers prior to sales of cars coming from rental fleets to increase their market price as used cars. Both of these practices are illegal these days, but as shown in the articles above, you can get up to a 10% exaggeration in cars which are explicitly within manufacturer specifications, which translates into 10% more miles on your rental bill, if you rent a car from one of those manufacturers.
[...] but it's bad for potential employees fresh out of school. No employer wants to be the one to front the capital to train them. They would just rather poach someone that will be effective on day 1.
People just out of school should be capable of being effective, day 1, or they went to the wrong school, or they simply didn't avail themselves of the opportunities at the school they went to, and did the minimum effort trajectory from enrollment to degree, and therefore their degree is pretty much worthless anyway.
If you have a 4 year CS degree, and an employer STILL has to train you in how to solve problems using a computer writing code in the C language, then the problem is with you. The school you went to isn't at fault if it provided you with adequate
opportunity
to learn, although you could possibly fault them for having given you a diploma after you have failed to do so. I would argue that's more a function of accreditation standards than it is of the school, since pass/fail is mostly dictated by how accreditation is awarded to institutions, and is a function of them retaining it.
In any case the example used in the submission is silly. The speech pathologists is complaining that the articles to do the job costs $1000. I make less than a speech pathologist and I easily spend $1000 a year making sure that I am up to date so that I can keep my job. It is like a few percent of my income. Expenses have to be put in context. If you are billing $100 a patient to medicare, and seeing 10-15 patients a day, it is out of line to expect some of that to be used for professional development?
Depends. What if she only wants to access articles which are applicable to her private practice, and which don't suck? If the article, which she can't read until she pays for it, fails to meet either of those criteria, does she get a refund?
Preprint federally funded research should be available online for little or not cost.
No cost; the cost has already been borne by the tax paying public who paid for the research; what's happening with these journals is that the researcher is double-dipping: once at the public trough, and a second time at the journal trough.
But an alleged professional whining that they get charged for a valuable product when they charge large amounts for their services, that is just silly.
As is calling publicly funded research a "product" which can be sold for money beyond the public funding which has already funded the science and the creation of the article describing it. Again: double dipping. This is in effect defrauding of the public paying for the research.
A better example, and real problem, are those working in less developed countries in which the resources are actually taxed, and science, even medicine, is extremely strained because in some cases journal costs do actually provide a significant road block to possible innovation.
Here we disagree. Why should the countries willing to bear the costs of the research benefit for free the countries who are too busy oppressing their own people for all but the oppressors to benefit from said research?
Do you think there will be a sudden influx of HIV drugs in Haiti or general medical care in Angola, Liberia, Sierra Leone, Côte d'Ivoire, the Democratic Republic of the Congo, the Republic of Congo, or Zimbabwe for the people enslaved to mine conflict diamonds, or food in Ethiopia, where the corrupt government would rather let food rot on the docks than go to the majority of the countries people who oppose those governments?
Better that the oppressors not get as good medical care, the better that they die out faster, and that we embargo everything to those countries that the government embargos from their opposition.
I believe the intent... is that all healthcare practitioners do not have private practices, but are instead employed by large healthcare conglomerates like Connecticut Life, United Healthcare, etc., and that those conglomerates have online access to the journals from their networks.
As long as you do not hang out your own shingle, and remain a wage-slave to a large corporation, you will have no problem accessing the necessary publications.
"Try to get competent at it without breaking U.S. law. I believe the criminal trespass laws when into effect in 1984, and Dmitry Sklyarov was arrested under the DMCA when he went to DefCon after being granted a Visa for the purpose of attending the conference."
Not hard at all, you can EASILY set up a Cybersecurity lab in your basement with useless garbage class computers. Even Windows Server software and OS is free for you to use for 30 days, wipe the drive rand reinstall every 30.
If you cant figure out how to set up your own lab, then you have no business even thinking about Cyber Security or computers in general. I can get old Cisco gear for near nothing all over the place, same as computers, etc... so there is not excuse unless you are living under a dumpster.
Hi Lumpy; we worked together at Apple.
I think it's a lot easier to penetrate something you set up yourself than it is to penetrate something designed by someone else. Part of the skill set is, and must be, the ability to think about things sideways, and the ability to out-think your (implied/actual) opponent on the other side.
My favorite analogy for this is ice fishing, where you make system calls in unexpected ways or absurd combinations, and end up achieving an unexpected result. We used to use this to crash the campus mainframe any time the operator got close to monitoring our serial lines, but the principle is the same today. Look at you you might go about escaping an exclusion group on a Linux system using setegid(), and the fact that the cr_groups[0] entry gets displaced, rather than rotated. Now compare this to the Mac OS X credential code which I wrote.
My second favorite analogy is that you should think like Clark Fries in the Heinlein novel Podkayne of Mars. In the novel, there were cubic service robots with no access to the interior of the robot. Clark Fries war more or less a sociopath, but he was a clever sociopath, and he figured out how to reprogram the robots. The way he did it was by realizing that cubes have 6 sides, and going in through the bottom to access the interior of the robot, which is a non-obvious solution that is obvious in retrospect. This technique also used to work on Coke machines, since incompletely loaded six-packs were often placed in the bottom area of the machine with the refrigerated compressor. If the machine ripped you off, you just pulled it away from the wall, and accessed the compressor compartment via the back of the machine.
This type of thinking around the problem is generally not something that gets taught in CS programs these days, any more than you would see the "cookie monster" program able to annoy someone by using write permission to their tty on a UNIX system these days until they typed in the word "cookie" at a shell prompt.
I think a lot of what has to happen these days in order to learn this type of tradecraft is you have to engage in (minimally) "grey-hat" activities: violate license agreements against reverse engineering, violate the DMCA, if only in your basement, and engage in activity against an actual opponent, not just yourself. If your opponent isn't devious in ways in which you aren't, then you generally don't learn anything new, except, well, how to identify bad code using a disassembler, and how to fuzz system calls using a fuzzer, and not much else. If your opponent isn't a willing participant in the exercise, well, then you've moved from grey hat to black hat territory, unless all you ever do to learn is participate in hackathons (and most of the people who win those things have pre-prepared their attacks using grey or black hat techniques before the show).
Please give me a big list of other occupations which more than 24% of a random sample of kids are interested in, then I'll allow you to claim that too few youngsters are interested in cybersecurity.
Try to get competent at it without breaking U.S. law. I believe the criminal trespass laws when into effect in 1984, and Dmitry Sklyarov was arrested under the DMCA when he went to DefCon after being granted a Visa for the purpose of attending the conference.
I'm going to guess that most of them also don't want to become good pickpockets, good safecrackers, or anything else that could land them in jail just for visiting the U.S..
Miss. The majority of the homeless here (San Francisco, where you obviously are not) are mentally ill. Unlike our immigrants - legal or otherwise - they do not as rule have jobs. Furthermore, many of them were shipped here illegally by irresponsible jurisdictions in distant areas of the country.
You can't make an argument about the wisdom of large nations accepting immigrants on a net basis, by referring to a particular public health issue in a single city.
Recent studies indicate that 30% of chronically homeless persons are mentally ill, while 50% of homeless persons are substance abusers. there is some overlap in these two groups: http://homeless.samhsa.gov/ResourceFiles/hrc_factsheet.pdf
According to this 2010 article, the last survey that linked homelessness and illegal immigration occurred in 2005; according to the article, when people who have "fallen off" unemployment roles are considered, the U.S. is at an average 16.9% unenployment, far higher than the current figures, which consider only those receiving unemployment, would have you believe. The article claims that illegal immigration contributes to homelessness not through arriving and subsequently being homeless themselves, but by providing a cheaper "under the counter" labor force which displaces unemployed legal residents from obtaining those jobs. Here's the article: http://www.examiner.com/article/illegal-immigration-contributing-to-homeless-crisis
The things we can do now adays in medicine are shocking...
The article claims the thing can last "up to 13 years" before having to be replaced.
In the past, we used Pu-238 RTGs called "Plutonium cells", and the pacemakers never had to be replaced.
I guess this step backwards, towards treating pacemakers as a treatment, rather than a cure, guarantees a recurring revenue stream. One wonders, given the industry that surrounds it, whether we will ever get a cure for anything that started out with just a treatment, such as diabetes, when there's so much money tied up in "recurring revenue streams" and so little in "pay for it once". The whole SAS field itself is based on it.
Why didn't your state setup their own exchange like my own, New York? It worked great. The fed site redirected me right to new york's site.
Easy Peasy.
I guess if the state you live in just couldn't get the job done themselves, and NEEDS TO RELY ON THE FEDERAL GOVERNMENT TO DO IT FOR THEM, well, beggers can't be choosers, can they?
Are you sure the domain name for your state's site should end in.ru?
Most cheapo DVI-HDMI converters fail to perform correct proxy EDID because they are simply electrical signal conversion blobs. This means that they fail to identify capabilities of the devices on either end of the cable to the other end of the cable.
This is why when you jam a cheap cable between your XBox and a Samsung TV, unless you are using the "default input used when there is no input signal on my plethora of inputs" connector, you have to wait for the TV to see it as a valid input so that it doesn't switch away from it back to the default, and then quickly unplug and replug the cable so that the EDID negotiation can take place over the electrically converting cable.
A decent (read: "more expensive") cable would act as an idle input at all times because it would be electrically active, and therefore willing to push "placeholder" negotiations down the line so that there would be "something there" as far as the TV is concerned, and then redo the EDID negotiation on behalf the the XBox (something it can't do, when it's not the primary TV input source, since the Samsung TV will only negotiate EDID on its currently active input source) as if a plug event had occurred.
Frankly, the standards for HDMI and DVI are pretty crappy for multi-input devices, and are squishy around when/if negotiations must take place for ports which are not the current active input. The workaround is smart cable so that both ends can think that, as far as the other end is concerned, they are the only device in the universe.
A secondary issue for AMD is that a lot of these cables fail to indicate which one of the combinations of DVI that the DVI device supports; it's generally not clear because they fail to differentiate analog vs. digital vs. with USB vs. without USB vs. with audio vs. without audio. Again, it's because they are dumb cables converting between a standard with a lot of optional-to-implement (DVI) and a standard that doesn't expect options to be left out (HDMI). The result is that some cheap cables will work for some things and others won't.
The fix, again, is smarter cables which identify themselves via EDID information that indicate and correctly negotiate their capabilities, in the asymmetric case of the output device and the input device not both being present and able to negotiate at the same time.
Yeah, it's a pain that such workarounds are needed; take it up with the standards committee and the device manufacturers.
PS: Linux has the same problem; which could easily be fixed by repeating attempts at EDID negotiation until it got an answer, since this wouldn't disrupt the ability to continue outputting in whatever default format while the negotiations take place; this is why the Dell 1600x1200 monitors have such a bitch of a time when you plug in Linux boxes.
Read up on monopsony and monopsony-like forces and how they distort willingness in the job market. It's not just for the deemed-unskilled, it's for everyone that is not an employer(until they get their regulatory comeuppance).
The putative "distortion of willingness" is proportional to the ability of individuals to make pareto optimal decisions based on external pressures. Thankfully, the people involved are not as mechanistic as you appear to be trying to imply they are.
Bad decisions about career path are the individual's responsibility, as is the decision to major in a field which society at large values less than some other field. So is the decision to rack up a large amount of debt, rather than working ones way through college, rather than taking on that debt, if one is incapable of obtaining third party funding in the form of stipends, grants, or scholarships. When you combine the two, and rack up a large debt to specialize in something society deems has little economic value, you've made a mistake.
And it's exactly the same thing why we still have things like Wal-Mart: their competitive advantage is that their size allows them to profit from economies of scale.
Except in areas where protectionist zoning doesn't permit them to operate at all. And it's in fact reasonable to do this, with regard to Walmarts, whose major business model is to open a store in an area, lower prices to the point everyone else is run out of business from their inability to compete on economies of scale, and then raises prices.
This is why almost all the Walmarts in the bay area, except the one in San Jose, are in the East Bay, and not on the peninsula (and you could argue about the one in San Jose not being on the Peninsula either).
Slashdot Mirror
But Childs really lost context: It was not his network. He had no business trying to enforce anything. The SF IT department may run their networks as stupidly as they chose, and while this may lead to criminal and civil liability on their part, it does not lead to any accountability towards Childs.
If he had turned over the passwords on the conference call, and someone in the room full of unauthorized people had used them in an unauthorized manner, instead of 4 years with the possibility of parole, he'd be looking at 20 years in a federal pound-you-in-the-ass prison with no possibility of parole for "aiding people in unauthorized computer access", a federal crime.
It really was a no-win situation, particularly since the city's immediate reaction was to sequester him away from all authorized persons until the mayor could do his grandstanding in front of the press.
Bull shit he didn't booby trap it. The network was configured to run from RAM at his design and he was the only person with the configuration. Power to any site will go out and that site will be down until the "hero" comes to save the day.
Citation needed
Feel free to cite any reputable news outlet.
So what exactly are the phone numbers for the House Intelligence Committee and the Senate Intelligence Committee, so that, you know, future whistle blowers can call them up, and not end up like Snowden?
Given that Atrix dock is cheaper than the target price and can be hacked (and we did just that at Google when Hexxeh interned there), why not just make a plastic case with a couple of connectors on it, and clip the Pi into it?
There are about 8 comparable telescopes...
http://en.wikipedia.org/wiki/List_of_neutrino_experiments
Which makes me wonder why this one is more likely than any of the others to detect a supernova.
Source please?
http://www.caranddriver.com/features/speedometer-scandal
http://online.wsj.com/news/articles/SB123119286106955181
http://www.theglobeandmail.com/globe-drive/car-tips/why-you-may-not-be-driving-as-fast-as-you-think/article11487709/
In general, German cars are known to exaggerate speed by up to 10% in order to guarantee compliance with European law (ECE-R39).
In the U.S., it's been historically common to "detune" speedometers in rental cars to exaggerate the speed, and therefore clock up additional miles which are then charged to the renter. It's also been historically common to roll back odometers prior to sales of cars coming from rental fleets to increase their market price as used cars. Both of these practices are illegal these days, but as shown in the articles above, you can get up to a 10% exaggeration in cars which are explicitly within manufacturer specifications, which translates into 10% more miles on your rental bill, if you rent a car from one of those manufacturers.
[...] but it's bad for potential employees fresh out of school. No employer wants to be the one to front the capital to train them. They would just rather poach someone that will be effective on day 1.
People just out of school should be capable of being effective, day 1, or they went to the wrong school, or they simply didn't avail themselves of the opportunities at the school they went to, and did the minimum effort trajectory from enrollment to degree, and therefore their degree is pretty much worthless anyway.
If you have a 4 year CS degree, and an employer STILL has to train you in how to solve problems using a computer writing code in the C language, then the problem is with you. The school you went to isn't at fault if it provided you with adequate
to learn, although you could possibly fault them for having given you a diploma after you have failed to do so. I would argue that's more a function of accreditation standards than it is of the school, since pass/fail is mostly dictated by how accreditation is awarded to institutions, and is a function of them retaining it.
Alternately... use the alternative audio and run speech recognition on it to solve the captcha.
No one thinks outside the box any more...
In any case the example used in the submission is silly. The speech pathologists is complaining that the articles to do the job costs $1000. I make less than a speech pathologist and I easily spend $1000 a year making sure that I am up to date so that I can keep my job. It is like a few percent of my income. Expenses have to be put in context. If you are billing $100 a patient to medicare, and seeing 10-15 patients a day, it is out of line to expect some of that to be used for professional development?
Depends. What if she only wants to access articles which are applicable to her private practice, and which don't suck? If the article, which she can't read until she pays for it, fails to meet either of those criteria, does she get a refund?
Preprint federally funded research should be available online for little or not cost.
No cost; the cost has already been borne by the tax paying public who paid for the research; what's happening with these journals is that the researcher is double-dipping: once at the public trough, and a second time at the journal trough.
But an alleged professional whining that they get charged for a valuable product when they charge large amounts for their services, that is just silly.
As is calling publicly funded research a "product" which can be sold for money beyond the public funding which has already funded the science and the creation of the article describing it. Again: double dipping. This is in effect defrauding of the public paying for the research.
A better example, and real problem, are those working in less developed countries in which the resources are actually taxed, and science, even medicine, is extremely strained because in some cases journal costs do actually provide a significant road block to possible innovation.
Here we disagree. Why should the countries willing to bear the costs of the research benefit for free the countries who are too busy oppressing their own people for all but the oppressors to benefit from said research?
Do you think there will be a sudden influx of HIV drugs in Haiti or general medical care in Angola, Liberia, Sierra Leone, Côte d'Ivoire, the Democratic Republic of the Congo, the Republic of Congo, or Zimbabwe for the people enslaved to mine conflict diamonds, or food in Ethiopia, where the corrupt government would rather let food rot on the docks than go to the majority of the countries people who oppose those governments?
Better that the oppressors not get as good medical care, the better that they die out faster, and that we embargo everything to those countries that the government embargos from their opposition.
I believe the intent... is that all healthcare practitioners do not have private practices, but are instead employed by large healthcare conglomerates like Connecticut Life, United Healthcare, etc., and that those conglomerates have online access to the journals from their networks.
As long as you do not hang out your own shingle, and remain a wage-slave to a large corporation, you will have no problem accessing the necessary publications.
"Try to get competent at it without breaking U.S. law. I believe the criminal trespass laws when into effect in 1984, and Dmitry Sklyarov was arrested under the DMCA when he went to DefCon after being granted a Visa for the purpose of attending the conference."
Not hard at all, you can EASILY set up a Cybersecurity lab in your basement with useless garbage class computers. Even Windows Server software and OS is free for you to use for 30 days, wipe the drive rand reinstall every 30.
If you cant figure out how to set up your own lab, then you have no business even thinking about Cyber Security or computers in general. I can get old Cisco gear for near nothing all over the place, same as computers, etc... so there is not excuse unless you are living under a dumpster.
Hi Lumpy; we worked together at Apple.
I think it's a lot easier to penetrate something you set up yourself than it is to penetrate something designed by someone else. Part of the skill set is, and must be, the ability to think about things sideways, and the ability to out-think your (implied/actual) opponent on the other side.
My favorite analogy for this is ice fishing, where you make system calls in unexpected ways or absurd combinations, and end up achieving an unexpected result. We used to use this to crash the campus mainframe any time the operator got close to monitoring our serial lines, but the principle is the same today. Look at you you might go about escaping an exclusion group on a Linux system using setegid(), and the fact that the cr_groups[0] entry gets displaced, rather than rotated. Now compare this to the Mac OS X credential code which I wrote.
My second favorite analogy is that you should think like Clark Fries in the Heinlein novel Podkayne of Mars. In the novel, there were cubic service robots with no access to the interior of the robot. Clark Fries war more or less a sociopath, but he was a clever sociopath, and he figured out how to reprogram the robots. The way he did it was by realizing that cubes have 6 sides, and going in through the bottom to access the interior of the robot, which is a non-obvious solution that is obvious in retrospect. This technique also used to work on Coke machines, since incompletely loaded six-packs were often placed in the bottom area of the machine with the refrigerated compressor. If the machine ripped you off, you just pulled it away from the wall, and accessed the compressor compartment via the back of the machine.
This type of thinking around the problem is generally not something that gets taught in CS programs these days, any more than you would see the "cookie monster" program able to annoy someone by using write permission to their tty on a UNIX system these days until they typed in the word "cookie" at a shell prompt.
I think a lot of what has to happen these days in order to learn this type of tradecraft is you have to engage in (minimally) "grey-hat" activities: violate license agreements against reverse engineering, violate the DMCA, if only in your basement, and engage in activity against an actual opponent, not just yourself. If your opponent isn't devious in ways in which you aren't, then you generally don't learn anything new, except, well, how to identify bad code using a disassembler, and how to fuzz system calls using a fuzzer, and not much else. If your opponent isn't a willing participant in the exercise, well, then you've moved from grey hat to black hat territory, unless all you ever do to learn is participate in hackathons (and most of the people who win those things have pre-prepared their attacks using grey or black hat techniques before the show).
Please give me a big list of other occupations which more than 24% of a random sample of kids are interested in, then I'll allow you to claim that too few youngsters are interested in cybersecurity.
Try to get competent at it without breaking U.S. law. I believe the criminal trespass laws when into effect in 1984, and Dmitry Sklyarov was arrested under the DMCA when he went to DefCon after being granted a Visa for the purpose of attending the conference.
I'm going to guess that most of them also don't want to become good pickpockets, good safecrackers, or anything else that could land them in jail just for visiting the U.S..
Miss. The majority of the homeless here (San Francisco, where you obviously are not) are mentally ill. Unlike our immigrants - legal or otherwise - they do not as rule have jobs. Furthermore, many of them were shipped here illegally by irresponsible jurisdictions in distant areas of the country.
You can't make an argument about the wisdom of large nations accepting immigrants on a net basis, by referring to a particular public health issue in a single city.
Recent studies indicate that 30% of chronically homeless persons are mentally ill, while 50% of homeless persons are substance abusers. there is some overlap in these two groups: http://homeless.samhsa.gov/ResourceFiles/hrc_factsheet.pdf
According to this 2010 article, the last survey that linked homelessness and illegal immigration occurred in 2005; according to the article, when people who have "fallen off" unemployment roles are considered, the U.S. is at an average 16.9% unenployment, far higher than the current figures, which consider only those receiving unemployment, would have you believe. The article claims that illegal immigration contributes to homelessness not through arriving and subsequently being homeless themselves, but by providing a cheaper "under the counter" labor force which displaces unemployed legal residents from obtaining those jobs. Here's the article: http://www.examiner.com/article/illegal-immigration-contributing-to-homeless-crisis
The things we can do now adays in medicine are shocking...
The article claims the thing can last "up to 13 years" before having to be replaced.
In the past, we used Pu-238 RTGs called "Plutonium cells", and the pacemakers never had to be replaced.
I guess this step backwards, towards treating pacemakers as a treatment, rather than a cure, guarantees a recurring revenue stream. One wonders, given the industry that surrounds it, whether we will ever get a cure for anything that started out with just a treatment, such as diabetes, when there's so much money tied up in "recurring revenue streams" and so little in "pay for it once". The whole SAS field itself is based on it.
Support Debian's build system, keep Greg employed!
That is all...
> The default cipher list for Java 7 was updated, but Android is stuck using JDK 6 and a default cipher list over a decade old.
The Android platform did not upgrade. How is that Oracle's fault? Next we will be blaming vendors for vulnerabilities that were patched years ago.
I fully understand your position. You're from the universe where Spock has a beard, right?
THIS.
Why didn't your state setup their own exchange like my own, New York? It worked great. The fed site redirected me right to new york's site.
Easy Peasy.
I guess if the state you live in just couldn't get the job done themselves, and NEEDS TO RELY ON THE FEDERAL GOVERNMENT TO DO IT FOR THEM, well, beggers can't be choosers, can they?
Are you sure the domain name for your state's site should end in .ru?
Who else thinks it'd be fun to sneak onto the roof ... and paint the roof like a giant Stargate?
I have no idea what people are bitching about... I just signed up for Obamacare through http://zombo.com/ .
Remember: all things are possible, with http://zombo.com/ !
Seriously, AMD, Why?
Most cheapo DVI-HDMI converters fail to perform correct proxy EDID because they are simply electrical signal conversion blobs. This means that they fail to identify capabilities of the devices on either end of the cable to the other end of the cable.
This is why when you jam a cheap cable between your XBox and a Samsung TV, unless you are using the "default input used when there is no input signal on my plethora of inputs" connector, you have to wait for the TV to see it as a valid input so that it doesn't switch away from it back to the default, and then quickly unplug and replug the cable so that the EDID negotiation can take place over the electrically converting cable.
A decent (read: "more expensive") cable would act as an idle input at all times because it would be electrically active, and therefore willing to push "placeholder" negotiations down the line so that there would be "something there" as far as the TV is concerned, and then redo the EDID negotiation on behalf the the XBox (something it can't do, when it's not the primary TV input source, since the Samsung TV will only negotiate EDID on its currently active input source) as if a plug event had occurred.
Frankly, the standards for HDMI and DVI are pretty crappy for multi-input devices, and are squishy around when/if negotiations must take place for ports which are not the current active input. The workaround is smart cable so that both ends can think that, as far as the other end is concerned, they are the only device in the universe.
A secondary issue for AMD is that a lot of these cables fail to indicate which one of the combinations of DVI that the DVI device supports; it's generally not clear because they fail to differentiate analog vs. digital vs. with USB vs. without USB vs. with audio vs. without audio. Again, it's because they are dumb cables converting between a standard with a lot of optional-to-implement (DVI) and a standard that doesn't expect options to be left out (HDMI). The result is that some cheap cables will work for some things and others won't.
The fix, again, is smarter cables which identify themselves via EDID information that indicate and correctly negotiate their capabilities, in the asymmetric case of the output device and the input device not both being present and able to negotiate at the same time.
Yeah, it's a pain that such workarounds are needed; take it up with the standards committee and the device manufacturers.
PS: Linux has the same problem; which could easily be fixed by repeating attempts at EDID negotiation until it got an answer, since this wouldn't disrupt the ability to continue outputting in whatever default format while the negotiations take place; this is why the Dell 1600x1200 monitors have such a bitch of a time when you plug in Linux boxes.
Read up on monopsony and monopsony-like forces and how they distort willingness in the job market. It's not just for the deemed-unskilled, it's for everyone that is not an employer(until they get their regulatory comeuppance).
The putative "distortion of willingness" is proportional to the ability of individuals to make pareto optimal decisions based on external pressures. Thankfully, the people involved are not as mechanistic as you appear to be trying to imply they are.
Bad decisions about career path are the individual's responsibility, as is the decision to major in a field which society at large values less than some other field. So is the decision to rack up a large amount of debt, rather than working ones way through college, rather than taking on that debt, if one is incapable of obtaining third party funding in the form of stipends, grants, or scholarships. When you combine the two, and rack up a large debt to specialize in something society deems has little economic value, you've made a mistake.
And it's exactly the same thing why we still have things like Wal-Mart: their competitive advantage is that their size allows them to profit from economies of scale.
Except in areas where protectionist zoning doesn't permit them to operate at all. And it's in fact reasonable to do this, with regard to Walmarts, whose major business model is to open a store in an area, lower prices to the point everyone else is run out of business from their inability to compete on economies of scale, and then raises prices.
This is why almost all the Walmarts in the bay area, except the one in San Jose, are in the East Bay, and not on the peninsula (and you could argue about the one in San Jose not being on the Peninsula either).
Yet despite all the discussion... MasterCard remains about as relevant as Diners Club.
I'd settle for making it compile with the 3.10 kernels... so I don't end up without a FUCKING GUI thanks NVIDIA.
I think this comes down to:
(1) When you change an API and break our software, you are an asshole
(2) When we change an API and break your software and you don't scramble to use the new API, we are forward thinkers contributing to progress
In the second case, it's clear that your code has "bitrotted", and it has nothing to do with the fact that we changed the API.