I meant restricting the ping to the site that supplied the page the link is *on*, which can be different than the site the link is *to*. In any case, somebody else pointed out that this idea is suggested in the reference document already, though I think it will have to be the default for this to be accepted by enough people.
You are right the spec does suggest that browsers be able to limit the ping to only the site supplying the page. Missed that.
I would suggest that some setting like this be the default. Otherwise I suspect too many people will turn it off completley, making this entire ping thing useless.
Why not limit the ping to the server that made the current page? This should prevent people from embedding pings into blogs, and still allow the replacement of redirects for tracking where you go. I would think unless this is done, too many people will disable it for any real sites to use it, and it will *only* be used for nefarious purposes.
The history of computers is not "there were punch cards, and then Bill Gates (or Steve Jobs) invented Windows and we had modern computers".
Computer history should start with the transistorized computer, or even later, and explain that they really work almost the same as the first electronic machines, just much much larger and faster. Too many people way underestimate how old things are. Too many people do not realize that people were typing to computers and looking at graphical output on screens in the 1950s! I am shocked at how many Slashdot posters do not beloieve computers existed before 1990, including Microsoft fanboys who do not even seem to remember that Bill Gates was writing software for hobbyist computers. Please explain that everything we use today was invented before 1990, and that the computers and internet we use today was completley predictable in 1970 or earlier.
Explain that computers and software are not perfect gifts from the gods, that they are compromises and design mistakes, like cars would be if they were forced to look exactly like Model-T's because otherwise they would not work on the highways built for Model-T.s. Explain that the true genius today in the current systems is in figuring out innovative ways to remain compatable with 50 years of mistakes, and still make it better and useful. Explain that making the software "better" also means making it "different", unless you work REALLY HARD!
Explain that a search that does not find anything will take at least as long, and probably longer, than a search that does find something. Try to explain why. People seem to lack insight that many of us consider completly obvious, which is really why they cannot understand your instructions on how to use a computer.
I don't have the foggiest clue how laundry soap works
Yes you do: you know that you have to put some into the washer or the clothes will not get clean. You know that you have to put a certain amount in, that if you put too little in it won't work, and putting more than a certain amount will not make the clothes cleaner. You know that after some number of times putting the soap in, the box will be empty, and the only way to put more soap in is to buy a new box.
In fact you know a lot about laundry soap. The frustrating thing about teaching people about computers is that they don't know information as rudimentary as the above knowledge about soap.
Apprently WINE does not have this length==1 bug. It has the documented bug, which is "the next 4 bytes of this file are interpreted as a pointer to jump to if you abort printing", which is bad, but not exactly this.
I'm not really buying this guys explanation, however. Software errors can have very strange side effects. Probably the short length causes it to reuse (rather than overwrite) the contents of some buffer as the code pointer, and that buffer just happens to contain a pointer to the next record of the metafile, and the length is also considered an error by some other code and thus triggers an "abort". A length of zero is detected and skipped correctly, while lengths of 3 or 4 overwrite enough of the pointer so that it does not work, making this 1 case the only one.
The only sites that all windows machines access on a regular basis are Microsoft's. The employee would also have to have access to Microsoft's web site to exploit this reliably.
Huh? Well I guess that is true if your goal is to infect or attack literally all Windows machines. However it is not necessary if you just want to infect or attack lots of Windows machines, which I would believe is pretty interesting to some crackers.
It is possible to dupliate a disk, including the CSS encryption, using professional equipment for mass production. This can be done with zero understanding of CSS. So CSS does nothing to stop money-making pirates.
The true purpose of CSS was to prevent translation to different forms by unauthorized software. The good thing is that this makes it harder to copy over the internet (currently sending a CSS disk image over the internet is impractical, and all known compression schemes require decoding the CSS). The bad thing is that this allows region encoding and locking out the fast-forward and other things that the authorized software may enforce.
Are you seriously saying that if Microsoft did not exist, there would be NO progress in computers? You have got to be kidding. Either that or you Microsoft apologists are getting really pathetic.
Now I believe that there would be some other monopolist in their place, and Billy Gates would be here bitching about that one along with everybody else. This one may have been worse or better, it is hard to say, but we would probably be somewhat in the same place we are now.
But if there really was not a monopoly, for some reason, I'm quite certain that computers would cost about $50 and would be entirely intoperable in that you could communicate instantly with any other computer the moment you walked into the same room, and the GUI designs would be some fantastic design that is unimagined yet by Windows or KDE or Apple or anybody in todays world.
You probably were not even born in 1980 so you don't remember how fast things were advancing before Microsoft took over, though.
It's not really a bug. It is in fact the documented function of the WMF files, and nobody (neither at Microsoft or WINE) noticed that it was in fact a security hole. Since it is documented there was no trouble replicating it's behavior.
There used to be several remotes with scroll and jog type controls but I don't see any anymore. I suspect some requirement for cheap manufacturing is the problem. If you just have a single plane of buttons, there are very few parts, just a membrane over the circuit card. A wheel will require a few more pieces and more depth to the unit. Apparently adding leds and little sliding covers and display screens are all much cheaper than any 3D component.
What's with this fanboy stuff? I scored 1620 on the SAT (including 900 on the math), which just proves I was a nerd. You sound like some sports fanatic quoting stats.
The EITC in the United States is characterized by a unique three-stage structure that consists of a phase-in range in which the credit increases as earnings increase, a plateau range in which the maximum credit has been reached and further earnings do not affect it, and a phase-out range in which the credit decreases as earnings increase. Currently, for a family with two dependent children, the credit is equal to 40 percent of the first $10,750 earned, plateaus at a maximum credit of $4,400, begins to phase-out when earnings increase beyond approximately $15,000, and reaches zero when earnings pass approximately $35,000. For a family with one dependent child, the structure is similar but has a phase-in rate of 34 percent and a maximum credit of $2,604. For those filing without dependents, there is a small credit of 7.65 percent of earnings with a maximum of $380. All dollar amounts are now indexed to inflation.
So there you have it, for all the naysayers and arguements on all sides. There really is a "negative tax rate". However note that a single person gets no more than $380, which is about the same amout as the Bush tax refund a few years ago. Not sure why it is arranged this way as that amount is uselessly small to effect anything. It looks more like the credit amount, ignoring this trivial $380, is approximately equal to the deduction for dependents. Thus it would be equivalent to instead refund a fixed amount per dependent, rather than deduct a fixed amount from income, to achieve the same result.
As others have pointed out the total handout is obviously smaller than what a family pays in sales and property taxes, so in no way does anybody get out of a tax burden.
The Wikipedia article goes on to say that the EITC is actually a really great and efficient welfare system. This may be true, I have no idea, and it is hard to tell if the writer has some agenda (either right or left). I would believe that simple things like this can be much more effective than a beuracracy.
You could see what you typed on a teletype. The rotating cylinder dropped after printing each letter so you could see it. All systems, including Unix, echoed each letter as it was typed.
There was a "backspace" though it was labelled delete or rubout. It's original purpose was to punch out all the holes in the teletype tape (there was a second button on the punch itself to back up, then you typed the rubout). However it sent a code to the computer so the computer could do anything, and Unix and almost all other systems used it to indicate that a character should be removed from the end of the current input line. Of course you could not erase the paper, the best ones sent the right number of backspaces, typed a # to erase the character, then went back to the clear area. I think Unix tended to type a backslash, each of the removed characters backwards, then another backslash. Unix actually had a lot more editing, such as the ability to cancel an entire line without backspacing the whole thing, and to delete words (with ^W?) than VMS or other systems I worked with.
Technically I feel that NeWS (from Sun) was even better. On NeXT you used a totally different interface (NextStep and/or system calls) to create and manage windows and events, than the interface you used to draw (DPS/PostScript). On NeWS *everything* was done in PostScript. NeWS also allowed arbitrarily shaped windows (including disconnected pieces and holes) and the interpreter in the first version was a good deal faster than DPS. It also appeared that running large amounts of the window toolkit as PostScript was a good idea, NeXT hides this but the fact is that most of the window borders and handing of events to them was done by PostScript routines on the server, not by ObjC code.
NeWS did suffer from not having any good communication between C and PostScript, making it easiest to write programs entirely in PostScript. Also Sun killed it by making it proprietary, while you could get the X11 source code for $100 and do anything you wanted with it.
It's probably more like 99.99% of Linux users do not hack on X source code.
However this change may allow that to be reduced to 99.9% by making it 10 times easier to hack on the source code for X and thus 10 times more popular.
Sure they will. What I meant is that the HD camcorder will not plug into a monitor that does not implement this DRM chip, whether or not the signal is protected.
No. It is highly likely that future devices will NOT play even non-protected content to a non-DRM display device. This is simply because the circuitry will not talk to the device unless it can negotiate it's DRM encryption. The original poster is quite correct that the designers expect to force every manufacturer to pay for their technology. If they were seriously interested in preventing piracy they would release a totally free design so everybody can build it, with some kind of registry of what keys are legit as opposed to fake keys built by hobbyists to try to circumvent.
Which is notepad. That is the ONLY program on Windows I have ever seen that does not understand bare LF. I believe this is done on purpose to break double-clicking on files stored on Unix. Running any other program such as textedit or MSWord or the VC++ IDE will show that they all work just fine.
Not a joke. There has to be a huge contingent of intelligent Microsoft programmers who are itching to get rid of some of the 1950's technology legacy, in particular the CR+LF line breaks and the drive letters. These are undeniably defects in Windows compared to Unix, no questions or arguments are possible, and intelligent non-zealots at Microsoft certainly realize this and really, really want to fix it.
Changing stored dates to be GMT would is also an obvious improvement but it is less visible. Everything else about the file system (how protection is done, case-dependent filenames, reserved characters in filenames, the directory names used, etc) are all debatable and you can make intelligent arguments to say the way Windows does it is better. Interestingly enough, these more debatable things are also much less of an impediment to cross-platform code.
Fixing the newlines is pretty easy: change the system and libraries so that "text/binary" is ignored while writing, but still there while reading. And fix stupid Notepad, which is the only program on Windows that does not understand LF.
Drive letters should be fixed by adding "/My Computer/A/" and so on to the namespace. The names should EXACTLY match what the user sees in Explorer (if they changed this for Vista it should be changed to match). Any other solution is bogus.
You seem to be confused. Yes there is still a kernel thing that talks to the hardware. Even in non-kernel X there is something in the Linux kernel that grants this process access to certain hardware. However you have to realize that this thing is TINY compared to a graphics server. Likely the difference in size is three or four orders of magnitude. Assumming bugs are evenly spread (which is probably false, there are probably fewer bugs than that in the hardware-talking layer), what NT (and X) has done is move 99.99% of the bugs out of the kernel!
Parent is right. The official "Windows" method is Ctrl+Insert for copy, Shift+Delete to cut, and Shift+Insert to paste. This is also the offical CDE/Unix method. These actually still work in many applications on both Windows and Unix, try them.
Both Windows and Unix quickly added the xcv settings because it was obvious from the Mac that they are easier. So the poster is bascially saying the orignal Mac is the best interface.
Slashdot Mirror
I meant restricting the ping to the site that supplied the page the link is *on*, which can be different than the site the link is *to*. In any case, somebody else pointed out that this idea is suggested in the reference document already, though I think it will have to be the default for this to be accepted by enough people.
You are right the spec does suggest that browsers be able to limit the ping to only the site supplying the page. Missed that.
I would suggest that some setting like this be the default. Otherwise I suspect too many people will turn it off completley, making this entire ping thing useless.
Why not limit the ping to the server that made the current page? This should prevent people from embedding pings into blogs, and still allow the replacement of redirects for tracking where you go. I would think unless this is done, too many people will disable it for any real sites to use it, and it will *only* be used for nefarious purposes.
The history of computers is not "there were punch cards, and then Bill Gates (or Steve Jobs) invented Windows and we had modern computers".
Computer history should start with the transistorized computer, or even later, and explain that they really work almost the same as the first electronic machines, just much much larger and faster. Too many people way underestimate how old things are. Too many people do not realize that people were typing to computers and looking at graphical output on screens in the 1950s! I am shocked at how many Slashdot posters do not beloieve computers existed before 1990, including Microsoft fanboys who do not even seem to remember that Bill Gates was writing software for hobbyist computers. Please explain that everything we use today was invented before 1990, and that the computers and internet we use today was completley predictable in 1970 or earlier.
Explain that computers and software are not perfect gifts from the gods, that they are compromises and design mistakes, like cars would be if they were forced to look exactly like Model-T's because otherwise they would not work on the highways built for Model-T.s. Explain that the true genius today in the current systems is in figuring out innovative ways to remain compatable with 50 years of mistakes, and still make it better and useful. Explain that making the software "better" also means making it "different", unless you work REALLY HARD!
Explain that a search that does not find anything will take at least as long, and probably longer, than a search that does find something. Try to explain why. People seem to lack insight that many of us consider completly obvious, which is really why they cannot understand your instructions on how to use a computer.
I don't have the foggiest clue how laundry soap works
Yes you do: you know that you have to put some into the washer or the clothes will not get clean. You know that you have to put a certain amount in, that if you put too little in it won't work, and putting more than a certain amount will not make the clothes cleaner. You know that after some number of times putting the soap in, the box will be empty, and the only way to put more soap in is to buy a new box.
In fact you know a lot about laundry soap. The frustrating thing about teaching people about computers is that they don't know information as rudimentary as the above knowledge about soap.
Apprently WINE does not have this length==1 bug. It has the documented bug, which is "the next 4 bytes of this file are interpreted as a pointer to jump to if you abort printing", which is bad, but not exactly this.
I'm not really buying this guys explanation, however. Software errors can have very strange side effects. Probably the short length causes it to reuse (rather than overwrite) the contents of some buffer as the code pointer, and that buffer just happens to contain a pointer to the next record of the metafile, and the length is also considered an error by some other code and thus triggers an "abort". A length of zero is detected and skipped correctly, while lengths of 3 or 4 overwrite enough of the pointer so that it does not work, making this 1 case the only one.
The only sites that all windows machines access on a regular basis are Microsoft's. The employee would also have to have access to Microsoft's web site to exploit this reliably.
Huh? Well I guess that is true if your goal is to infect or attack literally all Windows machines. However it is not necessary if you just want to infect or attack lots of Windows machines, which I would believe is pretty interesting to some crackers.
It is possible to dupliate a disk, including the CSS encryption, using professional equipment for mass production. This can be done with zero understanding of CSS. So CSS does nothing to stop money-making pirates.
The true purpose of CSS was to prevent translation to different forms by unauthorized software. The good thing is that this makes it harder to copy over the internet (currently sending a CSS disk image over the internet is impractical, and all known compression schemes require decoding the CSS). The bad thing is that this allows region encoding and locking out the fast-forward and other things that the authorized software may enforce.
Are you seriously saying that if Microsoft did not exist, there would be NO progress in computers? You have got to be kidding. Either that or you Microsoft apologists are getting really pathetic.
Now I believe that there would be some other monopolist in their place, and Billy Gates would be here bitching about that one along with everybody else. This one may have been worse or better, it is hard to say, but we would probably be somewhat in the same place we are now.
But if there really was not a monopoly, for some reason, I'm quite certain that computers would cost about $50 and would be entirely intoperable in that you could communicate instantly with any other computer the moment you walked into the same room, and the GUI designs would be some fantastic design that is unimagined yet by Windows or KDE or Apple or anybody in todays world.
You probably were not even born in 1980 so you don't remember how fast things were advancing before Microsoft took over, though.
It's not really a bug. It is in fact the documented function of the WMF files, and nobody (neither at Microsoft or WINE) noticed that it was in fact a security hole. Since it is documented there was no trouble replicating it's behavior.
I would agree, but including OS/X and Solaris as "Linux" is equivalent to including all bugs in WINE and FreeDos as "Windows" bugs.
There used to be several remotes with scroll and jog type controls but I don't see any anymore. I suspect some requirement for cheap manufacturing is the problem. If you just have a single plane of buttons, there are very few parts, just a membrane over the circuit card. A wheel will require a few more pieces and more depth to the unit. Apparently adding leds and little sliding covers and display screens are all much cheaper than any 3D component.
What's with this fanboy stuff? I scored 1620 on the SAT (including 900 on the math), which just proves I was a nerd. You sound like some sports fanatic quoting stats.
It would be fair if they counted all bugs in Wine and FreeDos and every other Windows emulator as being in the Windows catgagory.
From Wikipedia:
The EITC in the United States is characterized by a unique three-stage structure that consists of a phase-in range in which the credit increases as earnings increase, a plateau range in which the maximum credit has been reached and further earnings do not affect it, and a phase-out range in which the credit decreases as earnings increase. Currently, for a family with two dependent children, the credit is equal to 40 percent of the first $10,750 earned, plateaus at a maximum credit of $4,400, begins to phase-out when earnings increase beyond approximately $15,000, and reaches zero when earnings pass approximately $35,000. For a family with one dependent child, the structure is similar but has a phase-in rate of 34 percent and a maximum credit of $2,604. For those filing without dependents, there is a small credit of 7.65 percent of earnings with a maximum of $380. All dollar amounts are now indexed to inflation.
So there you have it, for all the naysayers and arguements on all sides. There really is a "negative tax rate". However note that a single person gets no more than $380, which is about the same amout as the Bush tax refund a few years ago. Not sure why it is arranged this way as that amount is uselessly small to effect anything. It looks more like the credit amount, ignoring this trivial $380, is approximately equal to the deduction for dependents. Thus it would be equivalent to instead refund a fixed amount per dependent, rather than deduct a fixed amount from income, to achieve the same result.
As others have pointed out the total handout is obviously smaller than what a family pays in sales and property taxes, so in no way does anybody get out of a tax burden.
The Wikipedia article goes on to say that the EITC is actually a really great and efficient welfare system. This may be true, I have no idea, and it is hard to tell if the writer has some agenda (either right or left). I would believe that simple things like this can be much more effective than a beuracracy.
You could see what you typed on a teletype. The rotating cylinder dropped after printing each letter so you could see it. All systems, including Unix, echoed each letter as it was typed.
There was a "backspace" though it was labelled delete or rubout. It's original purpose was to punch out all the holes in the teletype tape (there was a second button on the punch itself to back up, then you typed the rubout). However it sent a code to the computer so the computer could do anything, and Unix and almost all other systems used it to indicate that a character should be removed from the end of the current input line. Of course you could not erase the paper, the best ones sent the right number of backspaces, typed a # to erase the character, then went back to the clear area. I think Unix tended to type a backslash, each of the removed characters backwards, then another backslash. Unix actually had a lot more editing, such as the ability to cancel an entire line without backspacing the whole thing, and to delete words (with ^W?) than VMS or other systems I worked with.
Technically I feel that NeWS (from Sun) was even better. On NeXT you used a totally different interface (NextStep and/or system calls) to create and manage windows and events, than the interface you used to draw (DPS/PostScript). On NeWS *everything* was done in PostScript. NeWS also allowed arbitrarily shaped windows (including disconnected pieces and holes) and the interpreter in the first version was a good deal faster than DPS. It also appeared that running large amounts of the window toolkit as PostScript was a good idea, NeXT hides this but the fact is that most of the window borders and handing of events to them was done by PostScript routines on the server, not by ObjC code.
NeWS did suffer from not having any good communication between C and PostScript, making it easiest to write programs entirely in PostScript. Also Sun killed it by making it proprietary, while you could get the X11 source code for $100 and do anything you wanted with it.
It's probably more like 99.99% of Linux users do not hack on X source code.
However this change may allow that to be reduced to 99.9% by making it 10 times easier to hack on the source code for X and thus 10 times more popular.
Sure they will. What I meant is that the HD camcorder will not plug into a monitor that does not implement this DRM chip, whether or not the signal is protected.
No. It is highly likely that future devices will NOT play even non-protected content to a non-DRM display device. This is simply because the circuitry will not talk to the device unless it can negotiate it's DRM encryption. The original poster is quite correct that the designers expect to force every manufacturer to pay for their technology. If they were seriously interested in preventing piracy they would release a totally free design so everybody can build it, with some kind of registry of what keys are legit as opposed to fake keys built by hobbyists to try to circumvent.
Microsoft and Firefox agree on standard "dupe" icon!
Which is notepad. That is the ONLY program on Windows I have ever seen that does not understand bare LF. I believe this is done on purpose to break double-clicking on files stored on Unix. Running any other program such as textedit or MSWord or the VC++ IDE will show that they all work just fine.
Not a joke. There has to be a huge contingent of intelligent Microsoft programmers who are itching to get rid of some of the 1950's technology legacy, in particular the CR+LF line breaks and the drive letters. These are undeniably defects in Windows compared to Unix, no questions or arguments are possible, and intelligent non-zealots at Microsoft certainly realize this and really, really want to fix it.
Changing stored dates to be GMT would is also an obvious improvement but it is less visible. Everything else about the file system (how protection is done, case-dependent filenames, reserved characters in filenames, the directory names used, etc) are all debatable and you can make intelligent arguments to say the way Windows does it is better. Interestingly enough, these more debatable things are also much less of an impediment to cross-platform code.
Fixing the newlines is pretty easy: change the system and libraries so that "text/binary" is ignored while writing, but still there while reading. And fix stupid Notepad, which is the only program on Windows that does not understand LF.
Drive letters should be fixed by adding "/My Computer/A/" and so on to the namespace. The names should EXACTLY match what the user sees in Explorer (if they changed this for Vista it should be changed to match). Any other solution is bogus.
You seem to be confused. Yes there is still a kernel thing that talks to the hardware. Even in non-kernel X there is something in the Linux kernel that grants this process access to certain hardware. However you have to realize that this thing is TINY compared to a graphics server. Likely the difference in size is three or four orders of magnitude. Assumming bugs are evenly spread (which is probably false, there are probably fewer bugs than that in the hardware-talking layer), what NT (and X) has done is move 99.99% of the bugs out of the kernel!
Parent is right. The official "Windows" method is Ctrl+Insert for copy, Shift+Delete to cut, and Shift+Insert to paste. This is also the offical CDE/Unix method. These actually still work in many applications on both Windows and Unix, try them.
Both Windows and Unix quickly added the xcv settings because it was obvious from the Mac that they are easier. So the poster is bascially saying the orignal Mac is the best interface.