I have no problem with the security measure of typing the URL (although a text cut & paste is faster and gets around the security flaw just as well. (And it's *really* fast in X-windows - one drag to do the select, and one click to do the paste.) What I have a problem with is that nobody in the Windows camp will remember that back when Microsoft first starting doing this stuff we unix-heads *said* it was a bad security idea and that there's no point since cutting & pasting is so trivially easy anyway. Nope. It was "That's lame!", and "get with the times!", and "but just clicking is easier!".
Now they won't remember that there were wiser heads, and that they were right.
they ignore the content-type header, favouring the file extension instead.
Now, wait a minute. Do they actually IGNORE the header, or do they merely have it take less precedence than the extension? Those aren't the same thing. (In other words, in cases where the file extension isn't helpful, do they drop back to the content-type?) If so, that's not google's fault. They're tring to archive the web as it is actually used in practice, by people who are on average, ignorant of the standards. There are a lot of files out there where the content-type is going to be some generic term that only tells you, "yup, it's a binary file alright", or worse, are actually downright wrong. Given that these mistakes are everywhere out there, it might be that google decided they would get a more accurate database if they let the file extensions take precedent, as wrong as that may be from a conceptual standpoint (and very unfair, too - if my file ends in.doc that doesn't have to mean it's a Word document, and google shouldn't archive it that way.)
But anyway, the choice to let extension take precedence might be their only option. If most of the internet sites out there are doing it the wrong way, google has to aquiese and go along with that in order to have a more accurate database.
that does unpack everything only to throw away all but the file you wanted,
Well, actually it only unpacks the stuff that comes before the file in the archive. If the file in question is near the top, most of the archive is not unpacked.
but that only works if the problem isn't a design flaw...
If the problem is a system-level design flaw, then redundnacy of that system in one craft wouldn't buy you anything *either*, so I fail to see the validity of this complaint. ("Hey, it seems our software managing the flash filesystem has a flaw - But hey lucky for us we're running three copies of it and getting the same error three times instead of just one. Isn't that a wonderful improvement?")
Redundancy is useful for handling hardware errors, but this problem was in the software that keeps track of the flash filesystem. So having three computers would just mean you get the same problem three times, and your expensive redundancy would have bought you nothing. (They are having to take steps to make sure Opportunity doesn't encounter the same problem.)
It was well established by that point that gollum had an amazing, inhuman climbing ability. And you didn't actually see him fall. The camera didn't follow him to the bottom. That's well-established cinema-speak for "He's out of this scene, but he survived."
There are good design reasons for not coding directly to the hardware level, and having things fall through several layers of abstraction along the way, right? I mean, like how the c library's printf talks to the more low level write() function, which talks to the filesystem driver, which talks to the device driver, etc... Well, the sort of thing that DirectX does would sit at a level above the/dev files. It's an additional abstraction layer on top of the device driver. (So, for example, a program would run a function call along the lines of "I want to know if the user is hitting a directional button on any of the input devices right now, be it keyboard or joystick or gamepad. Is the user pushing a button right now, and if so tell me what it is.", instead of talking directly to the gamepad's device driver, then the keyboard driver, then the joystick driver.)
Which is a useless measure since the users also have to open things like Word Documents as attachments, and they don't realize that there is a difference between running a program and opening a data file. So getting a message about "are you really sure you want to open this attachment" usually results in a thought of, "well, Duh, of course I do - that's what we do all the time here in my all-Microsoft company full of all-Microsoft products."
The problem is that Outlook (and MS apps in general) doesn't distinguish, in a way the average user understands, that there is a relevant difference between "This file is data to be viewed in some other program already installed on your computer (like a word document)", versus "this file is a program itself that will run when you open it." BOTH use the terminology "open the attachment", and BOTH run with mime setup doing the work of deciding the difference, and deciding what "opening" the file really actually means.
So, it is Microsoft's fault - their 'friendly' interface that keeps users ignorant of an important difference is what causes users to learn that the "right" behaviour is to open attachments. After all, that's what you do when someone sends you a Word attachment or an Excell attachment, right?
And the reason they don't ever fix it is that it's part of their strategy to make people think their software is better - "Hey, in Outlook, attachments just plain work! No hassles! All those other systems suck. In fact, I've even heard that one some systems you can't even look at these attachments. Dood! What losers!"
Comparing DirectX to OpenGL is like comparing an entire apple tree to a single orange. OpenGL is a graphics library ONLY. In that, it does very, very well and beats Direct3D. But DirectX is a graphics library (Direct3D) AND a whole lot of other stuff - a game input device system (for joysticks, pads, tactile feedback systems, and action keyboard drivers) - and a sound system.
What would be nice about a DirectX equivilent for Linux would be that it is handy to encapsulate all that game-related driver stuff into one package, with one version number - so you can say "To run this program you need gamelib release 7 installed" - instead of "You need release 4.1.1 of foo, and release 1.0.1 of bar, and release 0.5.1.23 of baz, and..."
It is *always* possible to make biased reports like these if you have the deception skills of Microsoft PR. (Or, more importantly the unethical willingness to do so.) So changing the system so the specific issue the biased report is focusing on is taken care of doesn't *really* help. You can work toward that one small issue, but leave behind the millions more that could be exploited. It's like trying to move a puddle using a rake.
The real problem is that the marketplace is populated by people who can't see the flaws in the reports. If the average consumer isn't making informed buying decisions, then all other consumers collectively suffer because of it. (Which is one flaw in the notion that a free market will automatically be a corruption-free place in which everyone gets only what they deserve. The premise that your buying decisions affect only yourself isn't true.)
File and Print have never been a strong spot for Unix
For printing, yes. For file sharing, no. For file sharing, which OS is best is fully dependant on which OS is on the client machines. Trying to use a Windows machine as a file server for a unix client sucks even worse than trying to use a unix machine as a file server for a Windows client. So the 'winner' has nothing to do with the actual file server. Whichever server is identical to the clients and therefore stores all the same kinds of file permissions, flags, and whatnot, ends up winning the contest. Regardless of whether it is better or worse overall, it is a closer match to what the client machines need.
So, if you report "Most clients are Windows machines" as a seperate item from "Windows file servers work better", you are being deceptive. They are in fact the same thing. The second is merely a direct consequence of the first.
Second, he is artfully avoiding the reporter's question:.
Not necessarily. If he is actually brain damaged enough to believe what he said with all honesty, namely that a negative result from the survey would be impossible, then it is a perfectly valid response to the question to point out that it assumes facts not in evidence. It's like answering the question "have you stopped beating your wife", when you haven't even started. Technically, the answer is 'no', but you know that will be interpreted badly, so you attack the question itself. It is not necessarily dodging to point out that a question is unfair and thus you refuse to answer it.
Now, I don't for a minute believe he actually does think there was no chance of the results being negative, as he claimed. It was in *that* part that he was being dodgy and, to put it bluntly, lying, and that's where the deception lay, everything after that (including the 'dodge' of the question by pointing out it assumes facts not in evidence) was consequential to that first lie.
Bill Gates is the man who made computers accessible to the common people.
No, that was Compaq, back when they broke the stranglehold on the IBM PC and started that phrase "PC-compatable" that brought some actual competition to the hardware for once. The PC sucked compared to it's conteporaries like Atari ST and Amiga, in every techincal measure it was lacking - BUT what it had was actual competition from component makers, and actual choices of where you could buy one from. In the end, that level of openness in the PC, although it was unintended and even fought by it's parent company IBM, was strong enough to overcome any technical failings. It drove the prices down. It made it so anyone could make hardware without needing to pay huge licensing fees. Open stuff advances faster, even if it starts out behind it eventually catches up, and that's what happened with the PC and that's why it took off and everyone had one.
Microsoft was just lucky enough to be the little guy making the OS that ran on this open archetecture, and they were dragged along for the ride in a day an age when people compared the hardware first, and the OS second (because they were considered one in the same).
I have no qualms with admitting that Bill Gates' contribution to commerce is exactly on par with the likes of his fellow knights Mick Jagger and Elton John.
Let's say I produce a list of phone numbers and I *DO* offer it publicly. Who has to prove where I got it from? Me, or the accusing phone company that says I took the data from their phone book? If it's data that's publically available, then it's impossible to prove where you got it from. If I take every phone number, and list them by name alphabetically, then *in parallel* I will have created a work extremely similar to the phone company's phone book, whether I derived it from them or not. There are some laws that make the assumption that whomever uses an idea second must have copied it from whomever publicly took credit for it first (Patent law, for example), and totally ignore the ugly issue of trying to prove whether you really copied it or if it was a case of people with similar problems generating similar solutions in parallel.
I would think people with such high technical smarts would have a little more common sense than to ride a slippery slope all the way to the bottom on any issue regarding the protection of one's efforts.
That's pretty hypocritical coming from someone who promotes CommieWatch.org in his sig - which is a site FILLED with claims that are essentially cases of following the slippery-slope to the bottom.
You've never been sledding or skiiing or even hiking in extreme cold, have you?
False. I have (hiking and sledding, but not skiing)
It's easy to be overheated on the torso and have icy cold hands. That leads to its own problems - sweating, then chilling and then hypothermia.
Any analysis based on self-reporting is suspect because your body is fooling you as to what part of the body is losing the most heat. It will *always* feel like the fingers are the coldest part, even if the heat is leaking somewhere else.
Actually, it's not just as difficult. It's easier to prove something is existant than something is non-existant. And that's WHY what you suggest is the way things should be. One side, if correct, has the possiblity of proving it. The other, if correct, does not. The one who *can* prove it gets the burden of proof. To say otherwise is to assume the one who can't is automatically guilty.
The phone book fits the defintion. Is it now a copyright violation to write somone'es phone number down from the book?
And, the memory of my computer right now contains a database of individual running processes' context records. Is it a copyright violation to post the output of a 'ps' command?
Let me give you a couple examples from history. Roget's Thesaurus was basically the dictionary, re-indexed.
Which would have been illegal if a database that is a collection of definitions was considered copyrightable. Consider DeCSS. Even though it was reverse-engineered from scratch, it was still considered a copyright violation agaisnt the original CSS descrambling code. So, NO - you don't have to actually be caught copying the original work to be considered in violation. If your work is similar in content it is assumed to be the case that you copied it. That's what's so evil about copyrighting collections of public data. If someone else collects it the same way you did and puts it together into the same format you did, that's going to become illegal. And *THAT* is going to be a hell of a mess. (Consider, for example, that Google and Yahoo are doing pretty much the same service, creating pretty much the same kind of database out of the same exact source material, and presenting it to the user in pretty much the same way, and the data they use is coming from the same source. That should be perfectly legal.)
I remember vaguely some study that found that men who wear loose fitting pants have higher sperm counts on average than men who wear tighter, (presumably, "sexier") pants - because the tight pants keep the testicles too warm.
Slashdot Mirror
I have no problem with the security measure of typing the URL (although a text cut & paste is faster and gets around the security flaw just as well. (And it's *really* fast in X-windows - one drag to do the select, and one click to do the paste.) What I have a problem with is that nobody in the Windows camp will remember that back when Microsoft first starting doing this stuff we unix-heads *said* it was a bad security idea and that there's no point since cutting & pasting is so trivially easy anyway. Nope. It was "That's lame!", and "get with the times!", and "but just clicking is easier!".
Now they won't remember that there were wiser heads, and that they were right.
they ignore the content-type header, favouring the file extension instead.
Now, wait a minute. Do they actually IGNORE the header, or do they merely have it take less precedence than the extension? Those aren't the same thing. (In other words, in cases where the file extension isn't helpful, do they drop back to the content-type?) If so, that's not google's fault. They're tring to archive the web as it is actually used in practice, by people who are on average, ignorant of the standards. There are a lot of files out there where the content-type is going to be some generic term that only tells you, "yup, it's a binary file alright", or worse, are actually downright wrong. Given that these mistakes are everywhere out there, it might be that google decided they would get a more accurate database if they let the file extensions take precedent, as wrong as that may be from a conceptual standpoint (and very unfair, too - if my file ends in
But anyway, the choice to let extension take precedence might be their only option. If most of the internet sites out there are doing it the wrong way, google has to aquiese and go along with that in order to have a more accurate database.
that does unpack everything only to throw away all but the file you wanted,
Well, actually it only unpacks the stuff that comes before the file in the archive. If the file in question is near the top, most of the archive is not unpacked.
Directory services are not necessary outside of Windows. It's a tool that solves a non-existant problem in the unix world.
The point is, we're not talking about someone with a reasonable skeptical filter in place. It's GW Bush.
but that only works if the problem isn't a design flaw...
If the problem is a system-level design flaw, then redundnacy of that system in one craft wouldn't buy you anything *either*, so I fail to see the validity of this complaint. ("Hey, it seems our software managing the flash filesystem has a flaw - But hey lucky for us we're running three copies of it and getting the same error three times instead of just one. Isn't that a wonderful improvement?")
Redundancy is useful for handling hardware errors, but this problem was in the software that keeps track of the flash filesystem. So having three computers would just mean you get the same problem three times, and your expensive redundancy would have bought you nothing. (They are having to take steps to make sure Opportunity doesn't encounter the same problem.)
It was well established by that point that gollum had an amazing, inhuman climbing ability. And you didn't actually see him fall. The camera didn't follow him to the bottom. That's well-established cinema-speak for "He's out of this scene, but he survived."
There are good design reasons for not coding directly to the hardware level, and having things fall through several layers of abstraction along the way, right? I mean, like how the c library's printf talks to the more low level write() function, which talks to the filesystem driver, which talks to the device driver, etc... Well, the sort of thing that DirectX does would sit at a level above the /dev files. It's an additional abstraction layer on top of the device driver. (So, for example, a program would run a function call along the lines of "I want to know if the user is hitting a directional button on any of the input devices right now, be it keyboard or joystick or gamepad. Is the user pushing a button right now, and if so tell me what it is.", instead of talking directly to the gamepad's device driver, then the keyboard driver, then the joystick driver.)
it does, however, ask your permission first.
Which is a useless measure since the users also have to open things like Word Documents as attachments, and they don't realize that there is a difference between running a program and opening a data file. So getting a message about "are you really sure you want to open this attachment" usually results in a thought of, "well, Duh, of course I do - that's what we do all the time here in my all-Microsoft company full of all-Microsoft products."
The problem is that Outlook (and MS apps in general) doesn't distinguish, in a way the average user understands, that there is a relevant difference between "This file is data to be viewed in some other program already installed on your computer (like a word document)", versus "this file is a program itself that will run when you open it." BOTH use the terminology "open the attachment", and BOTH run with mime setup doing the work of deciding the difference, and deciding what "opening" the file really actually means.
So, it is Microsoft's fault - their 'friendly' interface that keeps users ignorant of an important difference is what causes users to learn that the "right" behaviour is to open attachments. After all, that's what you do when someone sends you a Word attachment or an Excell attachment, right?
And the reason they don't ever fix it is that it's part of their strategy to make people think their software is better - "Hey, in Outlook, attachments just plain work! No hassles! All those other systems suck. In fact, I've even heard that one some systems you can't even look at these attachments. Dood! What losers!"
Comparing DirectX to OpenGL is like comparing an entire apple tree to a single orange. OpenGL is a graphics library ONLY. In that, it does very, very well and beats Direct3D. But DirectX is a graphics library (Direct3D) AND a whole lot of other stuff - a game input device system (for joysticks, pads, tactile feedback systems, and action keyboard drivers) - and a sound system.
..."
What would be nice about a DirectX equivilent for Linux would be that it is handy to encapsulate all that game-related driver stuff into one package, with one version number - so you can say "To run this program you need gamelib release 7 installed" - instead of "You need release 4.1.1 of foo, and release 1.0.1 of bar, and release 0.5.1.23 of baz, and
It is *always* possible to make biased reports like these if you have the deception skills of Microsoft PR. (Or, more importantly the unethical willingness to do so.) So changing the system so the specific issue the biased report is focusing on is taken care of doesn't *really* help. You can work toward that one small issue, but leave behind the millions more that could be exploited. It's like trying to move a puddle using a rake.
The real problem is that the marketplace is populated by people who can't see the flaws in the reports. If the average consumer isn't making informed buying decisions, then all other consumers collectively suffer because of it. (Which is one flaw in the notion that a free market will automatically be a corruption-free place in which everyone gets only what they deserve. The premise that your buying decisions affect only yourself isn't true.)
File and Print have never been a strong spot for Unix
For printing, yes. For file sharing, no. For file sharing, which OS is best is fully dependant on which OS is on the client machines. Trying to use a Windows machine as a file server for a unix client sucks even worse than trying to use a unix machine as a file server for a Windows client. So the 'winner' has nothing to do with the actual file server. Whichever server is identical to the clients and therefore stores all the same kinds of file permissions, flags, and whatnot, ends up winning the contest. Regardless of whether it is better or worse overall, it is a closer match to what the client machines need.
So, if you report "Most clients are Windows machines" as a seperate item from "Windows file servers work better", you are being deceptive. They are in fact the same thing. The second is merely a direct consequence of the first.
Second, he is artfully avoiding the reporter's question:.
Not necessarily. If he is actually brain damaged enough to believe what he said with all honesty, namely that a negative result from the survey would be impossible, then it is a perfectly valid response to the question to point out that it assumes facts not in evidence. It's like answering the question "have you stopped beating your wife", when you haven't even started. Technically, the answer is 'no', but you know that will be interpreted badly, so you attack the question itself. It is not necessarily dodging to point out that a question is unfair and thus you refuse to answer it.
Now, I don't for a minute believe he actually does think there was no chance of the results being negative, as he claimed. It was in *that* part that he was being dodgy and, to put it bluntly, lying, and that's where the deception lay, everything after that (including the 'dodge' of the question by pointing out it assumes facts not in evidence) was consequential to that first lie.
The whole thing where you think that Gollum was killed, then he comes back right at the climax - cheesy.
That would have been cheesy. Good thing then that it never happened in this movie. Where were we supposed to think Gollum was killed?
Why plant a flag when you're going to be leaving the rover itself behind? It can *be* the flag.
Bill Gates is the man who made computers accessible to the common people.
No, that was Compaq, back when they broke the stranglehold on the IBM PC and started that phrase "PC-compatable" that brought some actual competition to the hardware for once. The PC sucked compared to it's conteporaries like Atari ST and Amiga, in every techincal measure it was lacking - BUT what it had was actual competition from component makers, and actual choices of where you could buy one from. In the end, that level of openness in the PC, although it was unintended and even fought by it's parent company IBM, was strong enough to overcome any technical failings. It drove the prices down. It made it so anyone could make hardware without needing to pay huge licensing fees. Open stuff advances faster, even if it starts out behind it eventually catches up, and that's what happened with the PC and that's why it took off and everyone had one.
Microsoft was just lucky enough to be the little guy making the OS that ran on this open archetecture, and they were dragged along for the ride in a day an age when people compared the hardware first, and the OS second (because they were considered one in the same).
I have no qualms with admitting that Bill Gates' contribution to commerce is exactly on par with the likes of his fellow knights Mick Jagger and Elton John.
Let's say I produce a list of phone numbers and I *DO* offer it publicly. Who has to prove where I got it from? Me, or the accusing phone company that says I took the data from their phone book? If it's data that's publically available, then it's impossible to prove where you got it from. If I take every phone number, and list them by name alphabetically, then *in parallel* I will have created a work extremely similar to the phone company's phone book, whether I derived it from them or not. There are some laws that make the assumption that whomever uses an idea second must have copied it from whomever publicly took credit for it first (Patent law, for example), and totally ignore the ugly issue of trying to prove whether you really copied it or if it was a case of people with similar problems generating similar solutions in parallel.
I would think people with such high technical smarts would have a little more common sense than to ride a slippery slope all the way to the bottom on any issue regarding the protection of one's efforts.
That's pretty hypocritical coming from someone who promotes CommieWatch.org in his sig - which is a site FILLED with claims that are essentially cases of following the slippery-slope to the bottom.
You've never been sledding or skiiing or even hiking in extreme cold, have you?
False. I have (hiking and sledding, but not skiing)
It's easy to be overheated on the torso and have icy cold hands. That leads to its own problems - sweating, then chilling and then hypothermia.
Any analysis based on self-reporting is suspect because your body is fooling you as to what part of the body is losing the most heat. It will *always* feel like the fingers are the coldest part, even if the heat is leaking somewhere else.
Actually, it's not just as difficult. It's easier to prove something is existant than something is non-existant. And that's WHY what you suggest is the way things should be. One side, if correct, has the possiblity of proving it. The other, if correct, does not. The one who *can* prove it gets the burden of proof. To say otherwise is to assume the one who can't is automatically guilty.
The phone book fits the defintion. Is it now a copyright violation to write somone'es phone number down from the book?
And, the memory of my computer right now contains a database of individual running processes' context records. Is it a copyright violation to post the output of a 'ps' command?
Let me give you a couple examples from history. Roget's Thesaurus was basically the dictionary, re-indexed.
Which would have been illegal if a database that is a collection of definitions was considered copyrightable. Consider DeCSS. Even though it was reverse-engineered from scratch, it was still considered a copyright violation agaisnt the original CSS descrambling code. So, NO - you don't have to actually be caught copying the original work to be considered in violation. If your work is similar in content it is assumed to be the case that you copied it. That's what's so evil about copyrighting collections of public data. If someone else collects it the same way you did and puts it together into the same format you did, that's going to become illegal. And *THAT* is going to be a hell of a mess. (Consider, for example, that Google and Yahoo are doing pretty much the same service, creating pretty much the same kind of database out of the same exact source material, and presenting it to the user in pretty much the same way, and the data they use is coming from the same source. That should be perfectly legal.)
I remember vaguely some study that found that men who wear loose fitting pants have higher sperm counts on average than men who wear tighter, (presumably, "sexier") pants - because the tight pants keep the testicles too warm.