Bypassing a technological measure is a crime in most cases, period.
Only in the USA and only because of the DMCA. We have a very bad habit in the USA of criminalizing acts which do not cause harm but may lead to harm at some later point. The anti-circumvention measures in the DMCA are a rather good example of that.
It exists as a patch. It was never merged because nobody actually wrote a driver that uses it.
Would it be possible to have an optimized API coexisting with the UDI one, and let the driver pick which one it wants to use?
Certainly. After all, the UDI layer for a particular platform will be implemented in terms of the underlying OS-specific (and in the case of Linux, volatile) API. But it would be difficult for a mfg to see value in writing both a UDI and native driver, unless some other operating system also supported UDI. It comes down to a chicken and egg scenario in that case, because UDI would have to be ported and available on more than one OS before it would become attractive to developers. It's not that it's a bad solution, it's just not compelling enough to divert OEMs away from writing native drivers.
You have to understand that the Microsoft world is one where double standards are not only accepted, they are embraced. It's useless to even point this out to one of those types, because double standards are considered acceptable in their reasoning system.
If you buy any piece of hardware at the store with plug and play support windows finds the right driver and it just works.
You mean, if you buy any piece of hardware at the store that Windows shipped with a driver for, then it just works. By the way, the same is true for Linux and the hotplug system.
If it doesn't have the driver, You put the CD in that came with it, and run the install program and it gets installed
You're a wishful thinker. It's actually been this easy about 50% of the time I've had to install some store-bought hardware on a home user's Windows machine. Observational bias is really rampant here, unfortunately.
This would truly be one of the holy grails of computing: the device manufacturer would only need to write and maintain a single driver, and everything from Windows to BeOS would be able to use it.
UDI has already been proposed, hyped up, and subsequently ignored. It turns out that getting the best performance on a given platform is more important to anyone involved than having driver source compatibility across platforms.
OS/2 was killed by glacial support from IBM and no Win32 support in WINOS2. It had absolutely nothing to do with the Windows 3.1 support built-in to it.
Customers who were looking forward had no choice but to go with Windows 95 as opposed to the dark horse with no marketing. If OS/2 had Win32 support and OEM preloads, it would have been a completely different ballgame. Microsoft was in a position to prevent both of those, and that ended up sealing their strategy to kill off OS/2.
I've heard people scream repeatedly that developers would simply target the Win3.1 API because it was included in OS/2, precluding OS/2 native applications and guaranteeing the failure of OS/2. First of all, I don't see how that guarantees the failure of OS/2 - it's a removal of a barrier to entry for the consumer, and the developers go where the consumer install base goes. Furthermore, if that assertion were true, then developers wouldn't have bothered writing Win32 native applications at all since Win3.1 compatibility was available - why lock out existing Win3.1 and OS/2 users?
The key here is who has a dominating market share and who has a forward path for compatibility with the dominating market share. Microsoft had a dominating market share, and IBM had no forward path for compatibility with it. OS/2 was doomed the moment Win95 was launched.
Engineers have had to deal with patents in every field for literally hundreds of years. Software development is no different. If you think it should be otherwise, I challenge you to examine the basis for that opinion.
You might like this to be true, but you're ignoring fundamental issues of software patents like:
They're too long. Software development isn't a 20 year product cycle like physical goods. It's a few years at most. The entire industry would have been held far, far back if AT&T and Xerox had patented everything they developed.
Most of them cover obvious inventions. The USPTO simply doesn't care enough to check prior art, because their paycheck doesn't depend on it. Furthermore, it seems any obvious idea somehow becomes novel to them by encoding it into a software process. Let the courts sort it out, sure, but that doesn't help the small-time outfit that's hit with a patent suit from a big corp or some slimy holding company.
They're vague. Software patents hardly ever describe a process (at least, the ones that bother actually describing a process and not just an idea) in a manner that could be translated into an implementation. This is a double win for the patent holder. Not only can they make the broadest possible claims to attempt to extract damages from a competitor, but the public gets absolutely nothing out of the deal once the 20 year legal monopoly expires.
A software patent policy discriminates against free software. The entry cost of developing software is much, much lower than the cost of engineering a physical product. Having to do a patent search on every technique used in a free program would absolutely dominate the cost of development.
It's possible to go on and on about the ramifications of software patents, but I doubt you really care about what anyone has to say on the topic.
Legal measures may not stop piracy as a trend (since it is a problem involving subjective morality, and legislation cannot change people's will), but to an individual, the financial risk and freedom risk is much greater in breaking the law than in sinking time and effort into breaking a rights-protection scheme. (Which would be a prerequisite in the former case anyway; do you think rights holders would abandon technological schemes simply because they have legislation on their side? Your ability to invoke your fair use rights is irrelevant to them.)
Do you want to play the black lottery with your entire future, just because you wanted to record a TV show for later viewing against the wishes of the rights holders?
Bringing countries outside the scope of the intended legislation into the debate is a red herring.
I believe those who bash "zealots" are simply saddened by their own inability to have passion and mindlessly lash out at those who can.
I don't think you can generalize quite like that. Passionate people get bashed as zealots when they cause inconvenience or harm to others, and they are unwilling to negotiate or even acknowledge that harm, because they think that everyone else _should_ share their worldview. A milder form of zealot is one who simply presumes that everyone else subscribes to the same moral axioms that he does (not necessarily attempting to convert them), and argues (annoyingly and unconvincingly) based on those false premises. Finally, there are the wishful thinkers, that ignore evidence contrary to their views and give a free pass to claims supporting their views regardless of the strength of the evidence.
An example of the first would be an Islamic fundamentalist (causing harm), or a marketdroid (causing inconvenience). An example of the second is a Linux zealot pushing Linux on a home user because "it's open source" or "it's free as in freedom", not because "it's a more solid product" or "it has a very high performance/price ratio" or "it gives you more control over what your kids can do on the computer". The last can be found in Microsoft defenders who insist that there are no security problems in Windows and that all software has hidden devastating security flaws anyway.
My guess is that the proponents of the broadcast flag are willing to lose it because it only serves to strengthen the fortifications for their next attempt at plugging the HDTV hole.
Which involves less risk to bypass: a technological measure, or a legal measure?
The DMCA prohibits circumventing access controls on copyrighted works or trafficking in circumvention devices. It was created specifically for the purpose of ensuring that once measures like the broadcast flag were put into place, consumers would have to become criminals in order to circumvent those measures.
Profit motive isn't the problem. Crooked politicians are.
Go to opensecrets.org. Look for the politicians that have a high correlation between their campaign supporters and the legislation they promote. Don't vote for those fuckers anymore, and convince everyone you know not to. If scouring opensecrets takes too much time, just vote for politicians who refuse campaign contributions as part of their platform.
The latter is literally the only failsafe we have if we expect to be able to trust elected officials not to make decisions contrary to the interests of their constituents. The alternative is to keep an eye on opensecrets and determine who we can and cannot trust according to their performance record.
The information is there. Why do we continue to vote crooks into office? As we continue to do this, more crooks are attracted to public office because they believe they are wanted there. It's a self-perpetuating problem. End it with your knowledge of each candidate and your vote.
Good point. And one that I would like the average Randian Libertarian/.er to explain. If they are so against the government regulating industry, why would they be for industry regulating government?
You're nuts. No "Randian Libertarian" would give government the power to prevent free enterprise at the request of other free enterprise. I think you have us confused with Big-Business Republicans.
And if corporations could do it cheaper, what is wrong with letting government do it, and then if they are correct the gov't won't be able to compete?
This statement makes no sense, unless you assume that because the government does it, it costs nothing. It is subsidized through taxes or inflation, take your pick. When that happens, people no longer have their money to choose the private enterprise instead of the government. Since they've already paid for one service, they will be inclined to use that one instead of pay for a different one, putting the private enterprise at a disadvantage. See the whole public school vouchers debate for why this strategy is a bad idea if you want quality and progress.
The choice is between having a security hole in a deployed piece of software, and running the risk of breaking applications that depend on that security hole. It's your choice whether or not to install security updates. How is the community supposed to regression test against your buggy closed source in-house software? Obviously, they can't. That's one of the responsibilities that you took upon yourself by standardizing on a poorly-supported proprietary application in-house.
It's ridiculous to blame the community for not having a magic wand to detect how every deployed site is using the software internally. If you want to do more regression testing than the community is able to do, then you are free to do it yourself before you deploy the fix.
What about all those who signed the Gator/Gain network EULA which prohibits the removal of said spyware/adware from PCs?
You mean zero? I don't know of anyone who has ever signed a EULA. This, combined with the fact that you are not allowed to see them until after the (non-refundable) purchase, is why their legality is dubious.
Assuming the proprietary software I chose was good (as I specified in my original statement) why is this a problem at all?
Because vendors change their priorities faster than you do. If you'd been around the industry long enough, you'd note that this is one of the only constants that has been around as long as commercial software has been around. Proprietary software is fine. Proprietary file formats and protocols are not fine, because your data and in-house applications are locked into those. Unless you are in a lucky minority, eventually the vendor will see a more lucrative opportunity elsewhere, and you'll be stuck.
Are you dense? He referred to patents and the DMCA. A program that uses a patented idea, even if it was discovered independently of the patent, cannot be legally distributed without the patent holder's permission. And with a program that circumvents the CSS protection on DVDs in order to play a DVD on Linux, you are trafficking in a "circumvention device" as defined by the DMCA and you can't legally distribute this either. Neither of these has anything to do with the self-created software recreating some closed-source proprietary software.
No offense. But it sounds like people are searching for things to dismiss this study.
Yeah. That's usually what you do when a study contradicts all previous studies and known experience. Does that imply some sort of bias? Only if you count being biased towards evidence as bias.
And, yes, Linux is NOT perfect, it is NOT flawless, and it IS full of security holes like anything else.
You mean it's probably full of security holes, based on the premise that it is difficult to create secure software. If it were actually as a fact full of security holes, it's likely we would know about them by now.
Nobody should take their operating systems so personally that they feel attacked when Linux is criticized.
Criticizing Linux is no problem. What is a problem is when Microsoft uses a study like this to promote their own agenda, which includes making it as difficult as possible to produce open source software that competes with their products. If this study is not absolutely solid, why would anyone with an interest in open source development let it go and not "search for things to dismiss it"? Eventually it will be used contrary to one's own interests if it is not shot down.
The problems with this study are numerous. The lead researcher was previously paid by Microsoft for a study. The statement by the Linux "advocate" about Microsoft producing more secure code is nonsense, since the researchers did not audit that code. The methodology was not precisely revealed. For example, they may have included trivial Apache vulnerabilities in external modules that aren't even installed or enabled in RHEL3. Without being given the precise methodology to reproduce the results, this is little more than a marketing exercise with a carefully chosen metric.
Oh, we got more oil after we invaded Iraq. I see. That explains these wonderfully low gasoline costs!
Gasoline prices at the pump have far more to do with refinery capacity and price fixing than they have to do with the price of a barrel of crude at the moment.
Downloading a copy of a song you have no right to have IS infringing a copyright. Whether you keep it or not.
Maybe, maybe not. Technically, it depends where the court would consider the reproduction to be taking place. Is it your computer that is creating the reproduction, or is it the server that is serving the file to you (who also happens to be in violation of the distribution privilege)?
Slashdot Mirror
We were talking about DOS programming. Where else would you use INT 19? :) Well, I guess you could use the GNU assembler under DOS, but ....
You have to understand that the Microsoft world is one where double standards are not only accepted, they are embraced. It's useless to even point this out to one of those types, because double standards are considered acceptable in their reasoning system.
Customers who were looking forward had no choice but to go with Windows 95 as opposed to the dark horse with no marketing. If OS/2 had Win32 support and OEM preloads, it would have been a completely different ballgame. Microsoft was in a position to prevent both of those, and that ended up sealing their strategy to kill off OS/2.
I've heard people scream repeatedly that developers would simply target the Win3.1 API because it was included in OS/2, precluding OS/2 native applications and guaranteeing the failure of OS/2. First of all, I don't see how that guarantees the failure of OS/2 - it's a removal of a barrier to entry for the consumer, and the developers go where the consumer install base goes. Furthermore, if that assertion were true, then developers wouldn't have bothered writing Win32 native applications at all since Win3.1 compatibility was available - why lock out existing Win3.1 and OS/2 users?
The key here is who has a dominating market share and who has a forward path for compatibility with the dominating market share. Microsoft had a dominating market share, and IBM had no forward path for compatibility with it. OS/2 was doomed the moment Win95 was launched.
They're too long. Software development isn't a 20 year product cycle like physical goods. It's a few years at most. The entire industry would have been held far, far back if AT&T and Xerox had patented everything they developed.
Most of them cover obvious inventions. The USPTO simply doesn't care enough to check prior art, because their paycheck doesn't depend on it. Furthermore, it seems any obvious idea somehow becomes novel to them by encoding it into a software process. Let the courts sort it out, sure, but that doesn't help the small-time outfit that's hit with a patent suit from a big corp or some slimy holding company.
They're vague. Software patents hardly ever describe a process (at least, the ones that bother actually describing a process and not just an idea) in a manner that could be translated into an implementation. This is a double win for the patent holder. Not only can they make the broadest possible claims to attempt to extract damages from a competitor, but the public gets absolutely nothing out of the deal once the 20 year legal monopoly expires.
A software patent policy discriminates against free software. The entry cost of developing software is much, much lower than the cost of engineering a physical product. Having to do a patent search on every technique used in a free program would absolutely dominate the cost of development.
It's possible to go on and on about the ramifications of software patents, but I doubt you really care about what anyone has to say on the topic.
Do you want to play the black lottery with your entire future, just because you wanted to record a TV show for later viewing against the wishes of the rights holders?
Bringing countries outside the scope of the intended legislation into the debate is a red herring.
It'd be asm("int 0x19"), and anyway, you can just replace the RETF opcode at the end of the main function with CD19 and you're good to go.
An example of the first would be an Islamic fundamentalist (causing harm), or a marketdroid (causing inconvenience). An example of the second is a Linux zealot pushing Linux on a home user because "it's open source" or "it's free as in freedom", not because "it's a more solid product" or "it has a very high performance/price ratio" or "it gives you more control over what your kids can do on the computer". The last can be found in Microsoft defenders who insist that there are no security problems in Windows and that all software has hidden devastating security flaws anyway.
The DMCA prohibits circumventing access controls on copyrighted works or trafficking in circumvention devices. It was created specifically for the purpose of ensuring that once measures like the broadcast flag were put into place, consumers would have to become criminals in order to circumvent those measures.
Go to opensecrets.org. Look for the politicians that have a high correlation between their campaign supporters and the legislation they promote. Don't vote for those fuckers anymore, and convince everyone you know not to. If scouring opensecrets takes too much time, just vote for politicians who refuse campaign contributions as part of their platform.
The latter is literally the only failsafe we have if we expect to be able to trust elected officials not to make decisions contrary to the interests of their constituents. The alternative is to keep an eye on opensecrets and determine who we can and cannot trust according to their performance record.
The information is there. Why do we continue to vote crooks into office? As we continue to do this, more crooks are attracted to public office because they believe they are wanted there. It's a self-perpetuating problem. End it with your knowledge of each candidate and your vote.
Please visit the following link for more information.
The choice is between having a security hole in a deployed piece of software, and running the risk of breaking applications that depend on that security hole. It's your choice whether or not to install security updates. How is the community supposed to regression test against your buggy closed source in-house software? Obviously, they can't. That's one of the responsibilities that you took upon yourself by standardizing on a poorly-supported proprietary application in-house.
It's ridiculous to blame the community for not having a magic wand to detect how every deployed site is using the software internally. If you want to do more regression testing than the community is able to do, then you are free to do it yourself before you deploy the fix.
Only on Slashdot would a PC be considered a "powerful server".
Are you dense? He referred to patents and the DMCA. A program that uses a patented idea, even if it was discovered independently of the patent, cannot be legally distributed without the patent holder's permission. And with a program that circumvents the CSS protection on DVDs in order to play a DVD on Linux, you are trafficking in a "circumvention device" as defined by the DMCA and you can't legally distribute this either. Neither of these has anything to do with the self-created software recreating some closed-source proprietary software.
The problems with this study are numerous. The lead researcher was previously paid by Microsoft for a study. The statement by the Linux "advocate" about Microsoft producing more secure code is nonsense, since the researchers did not audit that code. The methodology was not precisely revealed. For example, they may have included trivial Apache vulnerabilities in external modules that aren't even installed or enabled in RHEL3. Without being given the precise methodology to reproduce the results, this is little more than a marketing exercise with a carefully chosen metric.