I suspect 90% of modems sold today are onboard modems which the buyer doesn't need anwyay. (In my case it is 100%)
Re:Parent point valid despite foul language
on
Worst Linux Annoyances?
·
· Score: 4, Insightful
In the case of a disk eject, the OS needs to forcibly unmount the disk and allow the user to eject, and it should be the responsibility of any programs to gracefully fail, or even better, handle the error, if they really needed to access that disk.
I don't completely agree, but something similar to what you describe would be a nice feature. (As long as we don't force it upon anybody, choice is the answer). I don't like the Windows way of handling removable media. I don't like the Linux way of handling removable media. I don't like the Machintosh way of handling removable media. I don't like the IRIX way of handling removable media. And I don't like the SunOS way of handling removable media. AmigaOS got it almost right at first attempt. Now if somebody will please tell me how to detect the eject button in software, I will try to make an AmigaOS-like implementation for Linux. I also need to know how to detect that a disc was inserted.
I don't care if I get an I/O error. Just give me the damned
disk.
echo 0 >/proc/sys/dev/cdrom/lock
Re:So we can delete Windows even faster than befor
on
Windows 95 in 4.47MB
·
· Score: 1
Not much good for me; that would delete my nice squeaky-clean karma-free Linux/boot partition
Well, who knows who have what on which partition? I usually put a small/boot partition first on the disk, then a 2GB FAT partition (just in case I eventually need to run DOS, it can sit empty for years, because I never found any reason to install DOS).
However it turned out the computer I wrote this on actually have a 32MB FAT partition before/boot. Lucky me, that I only wrote that command on slashdot and non in my root shell.
In this particular case Win just means junk. Winmodems are known to be cripled hardware, that to save a few cents on the hardware let the CPU do an awfull lot of work with the driver code. In addition that it also means, that driver bugs can cause serious violations of standards. Finally I have heard about cases of attempts to hide the information you need to know to write the driver correctly. (Like if that would cause fewer broken drivers violating standards.)
Ever spent hours trying (and failing) to get a printer driver to work on Linux?
No, I never tried that. OTOH after my mother had spent months trying (and failing) to get a printer to work on Windows, I got the same printer working on Linux in five minutes.
Re:So we can delete Windows even faster than befor
on
Windows 95 in 4.47MB
·
· Score: 1
And...some things won't work on a conventional turning machine, at least not in polynomial time.
You are wrong about that. Anything that can be done on a conventional computer (or a RAM if you want a theoretical model) in polynomial time can also be done on a TM in polynomial time. However it will be a different polynomial. AFAIK at most you will add one or two to the degree of the polynomial. But in the real world it is less than that which seperates feasible from infeasible algorithms. Sorting slower than O(n^2) really makes the TM look slow.
Does anyone know how to enter an arbitrary URI though, I couldn't do
it.
A handfull of slashdotters trying to type different URIs in the same field would explain that.
Oh well, this will be the first article to ever slashdot a VNC server, I think.
It is not the first time that demo was mentioned on slashdot.
Yesterday I was trying to help my brother with some configuration problems on his Windows XP machine. Maybe it is just me that is too used to better things from four years of using Linux. But at least I found it hard to do just some very simple tasks. For example I would like to generate the list of all files in a directory including subdirectories. I had to give up, I couldn't find a function to do that. I'd also like to compare the contents of two directories. Again I had to give up. Finally I realized that perhaps it would be easier to just copy all the stuff to my Linux computer, because at least there I have all the tools I need. However the attempts to copy files bailed out with cryptic error messages. And I wasn't even given the option to continue with copying the rest of the files. And the progress indicator was useless, for more than half an hour it said there was 17 minutes left. While I had the computers connected I also found, that this XP installation by default had a directory shared with read+write permissions for everybody in the world. It is fortunate that this machine is not on the internet yet.
2 windows vunerabilities in the last month
9 Linux vunderablilites in the last month
And just how many of those nine can be exploited remote? And how many of them will actually allow the attacker to take over the system? You see with Windows people don't really care about local attacks, as most users are running with administrator priveleges most of the time anyway. Secondly I don't think Microsoft cares about the security problem unless it has been proven, that an attacker can take control of the system remotely. Just leaking a litle bit of information doesn't make the headlines. And while being able to crash Windows remotely might be bad, people might not notice the difference anyway, and it certainly doesn't cause a lot of worms to flow all over the net.
I would think it would depend on the resolve of the individual author vs. how much he/she needs the money offered.
Exactly my point. And since they need an agreement with each and every developer, it means it only takes one developer unwilling to cooperate, to spoil that plan.
When is louder better?
on
Is Louder Better?
·
· Score: 2, Interesting
My first though when I read the headline and the resume was: Of course louder is better. Because louder allows you to utilize more of the 16 bit quality of the CD. If the sound was too low, it would maybe only use 15 or 14 bits of the quality. I started reading the article, and at first it sounded to me like this guy didn't know anything about what was really going on. But finally about one third into the article he got to the real point. That the sound is simply scaled beyond the 16bit. So as loud as possible was simply not enough, it had to be louder than that causing irrecoverable damage to the sound. Those trolls saying you could just turn down the volume either didn't read the article, or didn't understand it. Turning down the volume will not bring back what was lost.
So what can we do about this? It would be nice with some analysis software to evaluate individual CDs. Not that software can tell you how good something sounds, only the ear can tell you that. But still it is good with some subjective meassures instead of only objective meassures. But that is not all. How about releasing two masterings only differing in the volume. One of those too loud, and another one that is simply scaled just enough to not cause clipping. So people could listen to whatever version they prefer, or even mix the two in a way that would actually reproduce the original with more than 16 bits of quality.
What would be even better was a new format and a standard somehow forbiding this practice. From the article it sounds like they are pushing the volume about 9dB too loud. How about a format the forbids an average volume higher than the -18dB of the range allowed by the given number of bits. The problem is that everybody wants to have the highest volume, so standardizing a volume below what will cause damages to the sound seems like a good plan.
Of course requiring a lower volume will loose some bits of quality. 18dB equals to 6 bits of your samples, so my suggestion would be to use 32bit samples which is a nice number and 8 bits more than I have heard about anybody using. Sure it is not going to happen with CDDA, but it is about time to get a replacement format anyway. Unfortunately I'm afraid those designing that stuff today are not focusing on quality, but a lot of other stuff like screwing their customers as much as possible.
Apparently, this guy doesn't understand the process he's writing
about.
I really wonder why anybody would moderate that a flamebait. Insightful would have been more apropriate. I believe the risk is nonexistent. Today there are branches, some more popular than others. Clearly Linus' official versions are still the most popular. But still some of the development happens in other branches and eventually returns back to the mainstream. If the mainstream starts evolving in a direction which a major number of person dislikes, some other branch will simply take over. There might be some fight, but the end result will be the same as today: One branch is the most popular, but it still accepts code developed in the other branches. The model is simply so robust I don't believe anybody can break it.
Actually trying to make a branch closed and proprietary would be a violation of the license (and if you don't accept the license it would be a violation of copyright instead). So whoever do this will be infriging the rights of thousands of developers. They can expect no help from the community, rather the opposite.
According to the security people at my work Win98 is affected, and since Microsoft no longer supports it they didn't bother to write a patch.
So no more Win98 machines on the Internet.... Oops looks like I'm daydreaming, make that no more secure Win98 machines on the Internet.... Yeah right, as if there ever were any.... How long before somebody write a worm to upgrade those machines?;-)
I'd rather spend time manually checking the code which is executed 100000 times a second rather than getting told of buffer overruns in something probably never gets executed.
The number of times it gets executed is not an issue. If it is vulnurabe executing it once is enough for the cracker to take control. Even if it is never executed under normal circumstances, the cracker might be able to do something to get it executed.
Not for interop purposes though, as far as I'm aware.
In Denmark the law still says reverse engineering is legal for interoperability purposes. It also says you are allowed to fix bugs in the code and make backups. And finally the EULA is not allowed to state otherwise.
Slashdot Mirror
I suspect 90% of modems sold today are winmodems.
I suspect 90% of modems sold today are onboard modems which the buyer doesn't need anwyay. (In my case it is 100%)
In the case of a disk eject, the OS needs to forcibly unmount the disk and allow the user to eject, and it should be the responsibility of any programs to gracefully fail, or even better, handle the error, if they really needed to access that disk.
I don't completely agree, but something similar to what you describe would be a nice feature. (As long as we don't force it upon anybody, choice is the answer). I don't like the Windows way of handling removable media. I don't like the Linux way of handling removable media. I don't like the Machintosh way of handling removable media. I don't like the IRIX way of handling removable media. And I don't like the SunOS way of handling removable media. AmigaOS got it almost right at first attempt. Now if somebody will please tell me how to detect the eject button in software, I will try to make an AmigaOS-like implementation for Linux. I also need to know how to detect that a disc was inserted.
I don't care if I get an I/O error. Just give me the damned disk.
/proc/sys/dev/cdrom/lock
echo 0 >
Not much good for me; that would delete my nice squeaky-clean karma-free Linux /boot partition
/boot partition first on the disk, then a 2GB FAT partition (just in case I eventually need to run DOS, it can sit empty for years, because I never found any reason to install DOS).
/boot. Lucky me, that I only wrote that command on slashdot and non in my root shell.
Well, who knows who have what on which partition? I usually put a small
However it turned out the computer I wrote this on actually have a 32MB FAT partition before
In this particular case Win just means junk. Winmodems are known to be cripled hardware, that to save a few cents on the hardware let the CPU do an awfull lot of work with the driver code. In addition that it also means, that driver bugs can cause serious violations of standards. Finally I have heard about cases of attempts to hide the information you need to know to write the driver correctly. (Like if that would cause fewer broken drivers violating standards.)
DVD support is the only reason I keep a Windows partition.
ogle, xine, and mplayer.
Too many graphics chips, too few drivers. Blame vendors.
Ever spent hours trying (and failing) to get a printer driver to work on Linux?
No, I never tried that. OTOH after my mother had spent months trying (and failing) to get a printer to work on Windows, I got the same printer working on Linux in five minutes.
And...some things won't work on a conventional turning machine, at least not in polynomial time.
You are wrong about that. Anything that can be done on a conventional computer (or a RAM if you want a theoretical model) in polynomial time can also be done on a TM in polynomial time. However it will be a different polynomial. AFAIK at most you will add one or two to the degree of the polynomial. But in the real world it is less than that which seperates feasible from infeasible algorithms. Sorting slower than O(n^2) really makes the TM look slow.
Does anyone know how to enter an arbitrary URI though, I couldn't do it.
A handfull of slashdotters trying to type different URIs in the same field would explain that.
Oh well, this will be the first article to ever slashdot a VNC server, I think.
It is not the first time that demo was mentioned on slashdot.
The real question is how is Microsoft going to execute the next transfer of cash into the SCO litigation fund.
Microsoft could buy Linux licenses for all their computers and pay SCO $699 each time.
Is anyone actually taking SCO seriously anymore?
Maybe Chris Sontag?
I need claim that MS stole my IP and put it in windows
That is my idea. I already patended it. So you have to pay me 50% of the money you make.
Six keystrokes: "dir /s"
Wrong, that doesn't list all files. Besides it lists a lot besides the filenames. I just want the filenames as if I had typed find in a normal shell.
Yesterday I was trying to help my brother with some configuration problems on his Windows XP machine. Maybe it is just me that is too used to better things from four years of using Linux. But at least I found it hard to do just some very simple tasks. For example I would like to generate the list of all files in a directory including subdirectories. I had to give up, I couldn't find a function to do that. I'd also like to compare the contents of two directories. Again I had to give up. Finally I realized that perhaps it would be easier to just copy all the stuff to my Linux computer, because at least there I have all the tools I need. However the attempts to copy files bailed out with cryptic error messages. And I wasn't even given the option to continue with copying the rest of the files. And the progress indicator was useless, for more than half an hour it said there was 17 minutes left. While I had the computers connected I also found, that this XP installation by default had a directory shared with read+write permissions for everybody in the world. It is fortunate that this machine is not on the internet yet.
2 windows vunerabilities in the last month
9 Linux vunderablilites in the last month
And just how many of those nine can be exploited remote? And how many of them will actually allow the attacker to take over the system? You see with Windows people don't really care about local attacks, as most users are running with administrator priveleges most of the time anyway. Secondly I don't think Microsoft cares about the security problem unless it has been proven, that an attacker can take control of the system remotely. Just leaking a litle bit of information doesn't make the headlines. And while being able to crash Windows remotely might be bad, people might not notice the difference anyway, and it certainly doesn't cause a lot of worms to flow all over the net.
did not give away information about the suspect (eg. race or eye colour
Why? Wouldn't it be cool if they could just take a DNA sample and produce a picture of the person?
I would think it would depend on the resolve of the individual author vs. how much he/she needs the money offered.
Exactly my point. And since they need an agreement with each and every developer, it means it only takes one developer unwilling to cooperate, to spoil that plan.
My first though when I read the headline and the resume was: Of course louder is better. Because louder allows you to utilize more of the 16 bit quality of the CD. If the sound was too low, it would maybe only use 15 or 14 bits of the quality. I started reading the article, and at first it sounded to me like this guy didn't know anything about what was really going on. But finally about one third into the article he got to the real point. That the sound is simply scaled beyond the 16bit. So as loud as possible was simply not enough, it had to be louder than that causing irrecoverable damage to the sound. Those trolls saying you could just turn down the volume either didn't read the article, or didn't understand it. Turning down the volume will not bring back what was lost.
So what can we do about this? It would be nice with some analysis software to evaluate individual CDs. Not that software can tell you how good something sounds, only the ear can tell you that. But still it is good with some subjective meassures instead of only objective meassures. But that is not all. How about releasing two masterings only differing in the volume. One of those too loud, and another one that is simply scaled just enough to not cause clipping. So people could listen to whatever version they prefer, or even mix the two in a way that would actually reproduce the original with more than 16 bits of quality.
What would be even better was a new format and a standard somehow forbiding this practice. From the article it sounds like they are pushing the volume about 9dB too loud. How about a format the forbids an average volume higher than the -18dB of the range allowed by the given number of bits. The problem is that everybody wants to have the highest volume, so standardizing a volume below what will cause damages to the sound seems like a good plan.
Of course requiring a lower volume will loose some bits of quality. 18dB equals to 6 bits of your samples, so my suggestion would be to use 32bit samples which is a nice number and 8 bits more than I have heard about anybody using. Sure it is not going to happen with CDDA, but it is about time to get a replacement format anyway. Unfortunately I'm afraid those designing that stuff today are not focusing on quality, but a lot of other stuff like screwing their customers as much as possible.
Apparently, this guy doesn't understand the process he's writing about.
I really wonder why anybody would moderate that a flamebait. Insightful would have been more apropriate. I believe the risk is nonexistent. Today there are branches, some more popular than others. Clearly Linus' official versions are still the most popular. But still some of the development happens in other branches and eventually returns back to the mainstream. If the mainstream starts evolving in a direction which a major number of person dislikes, some other branch will simply take over. There might be some fight, but the end result will be the same as today: One branch is the most popular, but it still accepts code developed in the other branches. The model is simply so robust I don't believe anybody can break it.
Actually trying to make a branch closed and proprietary would be a violation of the license (and if you don't accept the license it would be a violation of copyright instead). So whoever do this will be infriging the rights of thousands of developers. They can expect no help from the community, rather the opposite.
According to the security people at my work Win98 is affected, and since Microsoft no longer supports it they didn't bother to write a patch.
;-)
So no more Win98 machines on the Internet.... Oops looks like I'm daydreaming, make that no more secure Win98 machines on the Internet.... Yeah right, as if there ever were any.... How long before somebody write a worm to upgrade those machines?
Windows, IRIX, OS X.... How often do you replace your hardware?
I'd rather spend time manually checking the code which is executed 100000 times a second rather than getting told of buffer overruns in something probably never gets executed.
The number of times it gets executed is not an issue. If it is vulnurabe executing it once is enough for the cracker to take control. Even if it is never executed under normal circumstances, the cracker might be able to do something to get it executed.
Not for interop purposes though, as far as I'm aware.
In Denmark the law still says reverse engineering is legal for interoperability purposes. It also says you are allowed to fix bugs in the code and make backups. And finally the EULA is not allowed to state otherwise.