In other words, some people somewhere think the term should be equivalent to the GPL or FreeBSD license. That's ridiculous. There is absolutely nothing about the phrase "open source", linguistically, which implies anything like what you have said. "Open" can simply mean that you are "allowed access", nothing more. What Apple billed as open source was, in fact, open source, because people were allowed access to the source.
No, "open source" means "the source code is available for you to see". If you disagree, please specify which part of the term implies anything more than I said.
"GPL" means you can take the code, port it to whatever you damn well want and fork it. All GPL applications are open source, but not all open source applications are GPL.
IPC is slooow, and fork takes ages compared to other systems.
Yeah what the hell is with that, anyway? I just finished an OS X app that made heavy use of IPC and fork calls, and it seems like the fork()/exec(...) parts always took at least 3 second a piece. Granted I'd never really worked IPC that much before, but that does seem pretty damn slow for process splitting and code overlaying.
Never had the pleasure, I'm afraid. I've got his books on computer organization and operating systems. The one on operating systems isn't bad, and truth be told, the computer organization one isn't bad as a reference text. But as a textbook for students, it's fucking terrible. Tanenbaum doesn't get that undergraduate students don't have PhD's in computer science, and I lost points on some of the reading question assignments because his wording of the questions was unclear.
If you have a computer science degree you have probably used at least one if not more of his textbooks.
His book on computer organization was one of the worst pieces of crap I've ever been forced to drop cash on. He frequently goes off on tangents and spends more time trying to be clever than clearly explaining the material. I fucking hate Tanenbaum, even though I like micro-kernels.
While it's true that you do have to have a shell code to make an exploit (at least of the buffer overflow variety, of which virtually every one of these fixes was), the simple fact is that shell code for both x86 and PowerPC / Mac OS X is readily available. Once you have that, it's a simple matter of making C do the requisite magic to insert the code into the vulnerable program. Unless a particular exploit demands a particular kind of shell code (i.e. the buffer is exceedingly short), you should be able to use one shell code for just about anything.
And crafting a code with no NULL bytes isn't any easier or harder on PowerPC than it is on x86. On PowerPC, it's a matter of finding out which bytes in an instruction are reserved and can be changed. Or just putting dummy operations in the code to avoid having to put a zero value in one of the registers. That kind of thing isn't very difficult for someone with x86 assembler experience to pick up. (Possible sources of confusion might be PowerPC's lack of a real program counter register or the fixed stack frame size, but that doesn't really factor in when it comes to writing a shell code, and it can even make inserting that code into the process at the right offset easier in some cases.)
The basic fact of the matter is that switching architectures doesn't magically create software vulnerabilities. It's still up to software writers to not do stupid things like copy a variable-sized byte stream to a fixed-size buffer with no bounds checking or run the process as root when it doesn't need to be. A buffer overflow is still a buffer overflow, regardless of the architecture it's on.
I always got a kick out of that exchange. Because the sad fact is that politicians can't be made to agree on anything. If you put a bunch of them in a room and told them to come to a consensus on a certain issue in an hour or else the room would be flooded with nerve gas, they'd die bickering about whose side God is on and who is the best approximation of Hitler.
Actually, he did offer a hilariously stupid legal defense for his illegal wire-tapping. He claimed that, when Congress authorized him to invade Afghanistan, they also authorized him to do whatever he wanted in the name of fighting terrorism at home. (This idiotic interpretation of the resolution to invade Afghanistan has since been quietly dropped in favor of the "I'm King George, and you're going to do what I say" theory of American government.)
Unfortunately, it's difficult to judge just how well-designed QuickTime is because the documentation detailing its design is non-existent. For all we know, it could be the best-designed API in the world. But without documentation telling us how to use it, that doesn't do anyone much good.
That said, there still exists a whole lot of crufty code in QuickTime's core. It is, after all, about 15 years old. I wouldn't be surprised if Apple's been working on a massive overhaul for the past few years on it. With the Intel transition, they have to make sure it flies on two architectures. I'm betting that QuickTime 7's changes (QTMovieView, proper CoreAudio integration, a CoreVideo pipeline) were mere face-lifts compared to what Apple has in store for QuickTime 8.
QuickTime isn't "horribly written". It happens to be extremely fast for many operations. (Large portions of the API are actually macros rather than function calls.) What QuickTime is is horribly documented. A lot of the QuickTime docs on Apple's site are from the early 90's and simply don't apply to a video pipeline that's being run through the GPU. Apple's released a new Cocoa API for easily adding QuickTime views to applications, but writing plug-ins is still a pain in the ass. I have heard rumblings that QuickTime is due for a major overhaul in Leopard, though. This would make sense, since Apple's already provided a new high-level API for people to use.
Personality, with a few exceptions, is the big factor. Being able to work well with others is more important than all the experience in the world. If you can work well with others, you'll get the experience, and the rest of the team you work with will be more productive as a result. Like it or not, the corporate world is all about teamwork, whether you're a programmer, marketing person or graphic designer. Unless you're an actuary or pure numbers runner, you're simply going to have to work with others.
Now, as for these traits being assessed by an online personality quiz... that I haven't heard of. At least not outside the context of Best Buy or some place like that.
Meanwhile, users (though they don't quite yet know it) are offered virtually every function as a web application, at least for ninety percent, and in many ways the new applications surpass the old resident application paradigm for convenience, service, and ease of use and maintenance.
Yes, and those "web applications" only run on Internet Explorer 6. Fat lot of good that does.
I didn't realize that insults were part of a rational discussion.
Why couldn't they be? The only constraints governing a rational conversation is whether or not the participants are, in fact, being rational. Being rational says nothing about being civil.
No matter. I think you are confusing the issue. I think some of these creationists of whom you speak are pointing out that some structures, such as the eye and the digestive system (I question the eye argument myself) would have to make multiple improvements simultaneously to be an advantage.
Then they don't understand evolution. What's so terribly difficult to fathom about a clump of light-sensitive cells growing more complex as a result of natural selection? Hell, we see eyeballs of varying degrees of complexity and capability all around us. Is it really difficult to see an evolutionary progression and deduce common ancestry?
For example, a stomach is useless without HCl to digest food and the HCl would dissolve your tissues without the specialized lining found in the stomach.
Evolution does not make useless mutations impossible. There are, in fact, plenty of mutations which grant nothing and stick around because the organisms which had those useless mutations also had advantageous ones. Evolution is not a perfect process. Our bodies are basically cobbled together from parts of our ancestors, and it shows. Beyond that, there has been plenty of study done on the evolution of digestive systems. Even if there are some things that evolution cannot explain, that does not automatically grant credit to the idea that an invisible man waved his magic wand and created everything exactly as we see it today.
The bill requires consumers to be told if their privacy has been violated because of a breach.
So phone companies will be required to notify customers if the NSA illegally wire-taps their line under a Holy Decree -- er, I mean "executive order" from King George?
No, I was insulting them. There's a difference. Don't throw around terms you don't understand.
Straw man. No creationist claims that a creature must evolve overnight.
They most certainly do. Personally, I've had many creationists e-mail me demanding that I prove that creatures can just spontaneously morph into a higher life form. In fact, it's so common among creationists that AnswersInGenesis thinks that creationists should refrain from using it, since it's so incredibly stupid.
Micro-evolution and macro-evolution are two things pretty much dreamed up by creationists to overcome their own cognitive dissonance and idiotic misconceptions about evolution. On the one hand, they insist that no one has observed evolution. When someone points out that we have, they need to have a retort ready. So they said, "Well that's just micro-evolution! I want to see a fish change into a deer overnight!" In reality, if you want to classify micro-evolution as occurring when members of a certain population with a certain mutation can still breed with the rest of the population, that's fine. But it follows that, after a sufficient number of such changes, macro-evolution will occur. Meaning that after enough changes, you'll have speciation.
Micro-evolution and macro-evolution both have the exact mechanism. Creationists are simply incapable of making the trivial mental leap from "lots of small changes" to "one big change".
Slashdot Mirror
www.m-w.com
Credible enough for you?
Clever Acronyms Ignite Thinking. (CLIT)
In other words, some people somewhere think the term should be equivalent to the GPL or FreeBSD license. That's ridiculous. There is absolutely nothing about the phrase "open source", linguistically, which implies anything like what you have said. "Open" can simply mean that you are "allowed access", nothing more. What Apple billed as open source was, in fact, open source, because people were allowed access to the source.
No, "open source" means "the source code is available for you to see". If you disagree, please specify which part of the term implies anything more than I said.
"GPL" means you can take the code, port it to whatever you damn well want and fork it. All GPL applications are open source, but not all open source applications are GPL.
ICBM == Intel Chip-Based Mac.
Never had the pleasure, I'm afraid. I've got his books on computer organization and operating systems. The one on operating systems isn't bad, and truth be told, the computer organization one isn't bad as a reference text. But as a textbook for students, it's fucking terrible. Tanenbaum doesn't get that undergraduate students don't have PhD's in computer science, and I lost points on some of the reading question assignments because his wording of the questions was unclear.
While it's true that you do have to have a shell code to make an exploit (at least of the buffer overflow variety, of which virtually every one of these fixes was), the simple fact is that shell code for both x86 and PowerPC / Mac OS X is readily available. Once you have that, it's a simple matter of making C do the requisite magic to insert the code into the vulnerable program. Unless a particular exploit demands a particular kind of shell code (i.e. the buffer is exceedingly short), you should be able to use one shell code for just about anything.
And crafting a code with no NULL bytes isn't any easier or harder on PowerPC than it is on x86. On PowerPC, it's a matter of finding out which bytes in an instruction are reserved and can be changed. Or just putting dummy operations in the code to avoid having to put a zero value in one of the registers. That kind of thing isn't very difficult for someone with x86 assembler experience to pick up. (Possible sources of confusion might be PowerPC's lack of a real program counter register or the fixed stack frame size, but that doesn't really factor in when it comes to writing a shell code, and it can even make inserting that code into the process at the right offset easier in some cases.)
The basic fact of the matter is that switching architectures doesn't magically create software vulnerabilities. It's still up to software writers to not do stupid things like copy a variable-sized byte stream to a fixed-size buffer with no bounds checking or run the process as root when it doesn't need to be. A buffer overflow is still a buffer overflow, regardless of the architecture it's on.
I always got a kick out of that exchange. Because the sad fact is that politicians can't be made to agree on anything. If you put a bunch of them in a room and told them to come to a consensus on a certain issue in an hour or else the room would be flooded with nerve gas, they'd die bickering about whose side God is on and who is the best approximation of Hitler.
Mac OS Rumors is, as usual, very, very high.
Actually, he did offer a hilariously stupid legal defense for his illegal wire-tapping. He claimed that, when Congress authorized him to invade Afghanistan, they also authorized him to do whatever he wanted in the name of fighting terrorism at home. (This idiotic interpretation of the resolution to invade Afghanistan has since been quietly dropped in favor of the "I'm King George, and you're going to do what I say" theory of American government.)
Unfortunately, it's difficult to judge just how well-designed QuickTime is because the documentation detailing its design is non-existent. For all we know, it could be the best-designed API in the world. But without documentation telling us how to use it, that doesn't do anyone much good.
That said, there still exists a whole lot of crufty code in QuickTime's core. It is, after all, about 15 years old. I wouldn't be surprised if Apple's been working on a massive overhaul for the past few years on it. With the Intel transition, they have to make sure it flies on two architectures. I'm betting that QuickTime 7's changes (QTMovieView, proper CoreAudio integration, a CoreVideo pipeline) were mere face-lifts compared to what Apple has in store for QuickTime 8.
QuickTime isn't "horribly written". It happens to be extremely fast for many operations. (Large portions of the API are actually macros rather than function calls.) What QuickTime is is horribly documented. A lot of the QuickTime docs on Apple's site are from the early 90's and simply don't apply to a video pipeline that's being run through the GPU. Apple's released a new Cocoa API for easily adding QuickTime views to applications, but writing plug-ins is still a pain in the ass. I have heard rumblings that QuickTime is due for a major overhaul in Leopard, though. This would make sense, since Apple's already provided a new high-level API for people to use.
Multiply that by another 10 times if it will actually display the full names of directories instead of "MYDOCUM~1".
Personality, with a few exceptions, is the big factor. Being able to work well with others is more important than all the experience in the world. If you can work well with others, you'll get the experience, and the rest of the team you work with will be more productive as a result. Like it or not, the corporate world is all about teamwork, whether you're a programmer, marketing person or graphic designer. Unless you're an actuary or pure numbers runner, you're simply going to have to work with others.
... that I haven't heard of. At least not outside the context of Best Buy or some place like that.
Now, as for these traits being assessed by an online personality quiz
Yes, in the country which elected George W. Bush to the presidency. Twice.
Need I remind you that Bill O'Reilly still has his own show?
No. See here.
Micro-evolution and macro-evolution are two things pretty much dreamed up by creationists to overcome their own cognitive dissonance and idiotic misconceptions about evolution. On the one hand, they insist that no one has observed evolution. When someone points out that we have, they need to have a retort ready. So they said, "Well that's just micro-evolution! I want to see a fish change into a deer overnight!" In reality, if you want to classify micro-evolution as occurring when members of a certain population with a certain mutation can still breed with the rest of the population, that's fine. But it follows that, after a sufficient number of such changes, macro-evolution will occur. Meaning that after enough changes, you'll have speciation.
Micro-evolution and macro-evolution both have the exact mechanism. Creationists are simply incapable of making the trivial mental leap from "lots of small changes" to "one big change".
Ugh ... it's "Mac", not "MAC". MAC is an acronym for "Media Access Control". "Mac" is shorthand for "Macintosh".