You got it all wrong. Nobody wants to know what your window title is. However somebody wants to decide what your window title is going to become, and later they are going to have your shell execute that as if it were a command.
If as many people tried as hard to find security holes in OSX or Linux, there'd be reports for those daily as well.
You missed the part about claimed security hole. There was no hole, just a lot of people claiming something was a hole. The very same hole has often been claimed to exist in Linux as well. In reallity we are talking about a feature, that exists in any decent OS.
Google, when it's 'reading' a page, is having a bot spider it. If google is spidering a page and comes across a link to a page it has not 'read', then it follows the link, spiders the page, and includes it in the index.
Yes, but there can be a long time from the link is seen till it is downloaded. In the meantime it is able to provide the link in searches for words in the link text. As you crawl the web, you will find that the number of links to pages you have not yet downloaded is always larger than the number of pages you have downloaded. And it stays that way, because each time you download another page you will find on average more than one link you have never seen before.
Not as long as some people refuses to implement an idle priority. Even when running as nice as possible, a process will still get some amount of CPU time. And I only want to use the CPU time when the soundcard is actually in use, otherwise I want to halt the CPU to spare some resources.
When your CPU comes on and off of the HLT (halt) instruction, it creates a tiny electromagnetic field that is perceptible by your sound card.
Indeed I have had the very same problem with my computer. For a long time I simply added no-hlt to my kernel command line to remove the noise. To demonstrate that this was really the cause, I wrote a DOS program to provoke lots of noise on any computer suffering from this problem. The Turbo Pascal source is also available. (Don't try this program under Windows, It will not work).
Having lived with the no-hlt option for a long time, it came to my mind, that always wasting power in the CPU just because I occasionally want to use the soundcard without noise was stupid. Instead I wrote a patch that allowed me to switch the HLT instruction on and off as the soundcard driver was loaded and unloaded.
Another brand of software is in no means a security benefit. You have now doubled the requirements for a secure environment: both vendors now must be equally as secure.
Now you are assuming that for the system as a whole to be secure, each part has to be secure. However that is not how secure systems are designed. Though DNS is really not designed with much security in mind, so it may all fail due to weaknesses in design. Had the design been secure multiple implementations would be an advantage. The best you can do right now to protect yourself from false results from a cracked root DNS server is to ask more than one, but that of course does not help against DoS attacks. And of course always use secure protocols like ssh and https on top of IP+UDP+TCP+DNS.
Is Linux a legitimate solution to school districts facing a financial crunch?
Yes of course it is. Some people says students should be tought to use the software being used in the "real life". Why? If the students learn to acomplish the same task with cheaper software, how could that be bad?
But much rather than sticking with one choice of software, I'd see students trying a few different systems, so they can learn what are the differences and similarities between them, and they can learn how to learn using a new system, and they can make up their own minds about what they like and dislike. Because you cannot teach them how to use the software they are going to find themselves with in a few years, but you can theach them how to learn.
So let them try Linux, Unix, Windows, BSD, OSX, and let them find the best for each task.
Is the binary for Linux large enough that if the bits were mapped to an image, it would work as a background image?
I typically use kernels of size 2-3MB in uncompressed form and 0.7-1MB in compressed form. I typically use a screen resolution between 1024x768 and 1600x1200, if you use a black/white picture that is 96-234KB. So the kernel is certainly large enough to make a black/white picture for your background. If you want 24-bit truecolor image it might be difficult to fill your entire screen.
the hypothetical 12-year-old listening to all the advice about girls everyone else is giving.
When I was 12, I could really have used that advice. And since that me at the age of 12 did not know what to do, I think he might have listened to me. And here is what I would have told him: If you really like that girl, say it to her. And say it to everybody else, heck even say it to your parents. (If you are afraid to talk about feelings with your parents write it in a letter, they will understand. They are not really out to get you.) If you give up on the girl you like, you are going to regret forever about the chances you might have missed. It might not work out, and she might be breaking your heart. But you should rather take the chance and let her break your heart than go and break your own heart. Stand up and fight for what you want.
The error is generated by the receiving mailserver
The server will respond with a 550 SMTP error code. It will even refuse to accept the message if the recipient does not exist. Once the server has refused to receive the message it does no more about it.
Besides, if bounce messages had to be created during the initial contact, relaying wouldn't work.
Again it is the sending server that will generate the error message. However in this case the same message is being transfered with SMTP multiple times. It is in the last SMTP session the error is detected, and it is the sending server in this session that needs to generate the error mail.
which they don't because they use fake email addresses.
Again they can abuse the SMTP protocol to know if the address worked or not. Nobody forces the spammer to use a mailserver for sending spam, they can use any software they desire. In particular the spammer can use software, that rather than generating the bounce simply generate lists with good and bad addresses. This does not work if the spammer uses an open relay, because the relay accepts the message, and it is then the relay that generates the bounce message. But the spammer can still hide by using an open proxy rather than an open relay.
I did the experiment last time I heard about this spam to unused addresses. I don't remember which browser I used, neither what problem I got. But I think it must have been a version of either Netscape or Mozilla.
Like the one you get when there is no address with that name.
That is probably not possible.
The error is generated by the sending mailserver, not the receiving mailserver. The sender is told already when sending wether the destination address is valid, so once your mailserver has said OK, it is already too late.
The sender address could be invalid, so the generated error message could be send to somebody else.
You would live in a constant risk of blacklisting a valid sender address.
Spammers have been bruteforcing honeypot domains for a few years now, sending spam to any and all combinations of letters and numbers.
Surely they can bruteforce some, but they cannot try all combinations. If the test addresses are long and random enough, they cannot be found by bruteforce. I wanted to try that experiment myself, but I wasn't even able to create an account on hotmail. When I was a hotmail user many years ago, it worked nicely in many browsers. And hotmail was even improved during the time I used it. I stopped using hotmail the day Microsoft bought it. But I did check if new mail arrived for some time after that, but Microsoft introduced more and more problems, and eventually I was no longer able to use it.
And BTW, some of the spammers verifying addresses on hotmail have been using my honeypot, which they thought was an open proxy. My honeypot said OK to all of those addresses, so there will be invalid email addresses on their lists.
The article says so, but that is incorrect. The upper limit is actually 63 sectors. And it has been a long time since I saw a harddisk with less. Because of the 504MB limit harddisk generaly used the 63 sectors to the limit. Later BIOSes did translation between IDE and BIOS interface geometry. That way the limit was pushed close to 8GB, but new problems were introduced. The 63 sectors was not touched by the translation, it was most convenient to keep that and only move bits between cylinder and side fields of the geometry. So today all harddisks on PCs have 62 sectors between the MBR and the first partition.
Hell, any kludge that two kids throw together can be considered a 'Linux distro.'
Of course. Since Linux is only a kernel it doesn't take much. Creating a usable system is a whole different matter. Then again that depends on what you want to use the system for, you can create systems specialized for any purpose.
Slashdot Mirror
If you really want my window title
You got it all wrong. Nobody wants to know what your window title is. However somebody wants to decide what your window title is going to become, and later they are going to have your shell execute that as if it were a command.
For instance, you can recover a root password in just 10-15 minutes on ANY machine.
You shouldn't make such claims without any evidence.
Heh, 98% of the packets on the internet are useless.
Hardly surprising since 98% of the users on the internet are useless.
If as many people tried as hard to find security holes in OSX or Linux, there'd be reports for those daily as well.
You missed the part about claimed security hole. There was no hole, just a lot of people claiming something was a hole. The very same hole has often been claimed to exist in Linux as well. In reallity we are talking about a feature, that exists in any decent OS.
I am shocked and appalled that Slashdot would ever report something from a source biased towards *NIX!
What you should be shocked about is that this time they actually admit it.
Google, when it's 'reading' a page, is having a bot spider it. If google is spidering a page and comes across a link to a page it has not 'read', then it follows the link, spiders the page, and includes it in the index.
Yes, but there can be a long time from the link is seen till it is downloaded. In the meantime it is able to provide the link in searches for words in the link text. As you crawl the web, you will find that the number of links to pages you have not yet downloaded is always larger than the number of pages you have downloaded. And it stays that way, because each time you download another page you will find on average more than one link you have never seen before.
yet another reason to be running distributed.net
Not as long as some people refuses to implement an idle priority. Even when running as nice as possible, a process will still get some amount of CPU time. And I only want to use the CPU time when the soundcard is actually in use, otherwise I want to halt the CPU to spare some resources.
When your CPU comes on and off of the HLT (halt) instruction, it creates a tiny electromagnetic field that is perceptible by your sound card.
Indeed I have had the very same problem with my computer. For a long time I simply added no-hlt to my kernel command line to remove the noise. To demonstrate that this was really the cause, I wrote a DOS program to provoke lots of noise on any computer suffering from this problem. The Turbo Pascal source is also available. (Don't try this program under Windows, It will not work).
Having lived with the no-hlt option for a long time, it came to my mind, that always wasting power in the CPU just because I occasionally want to use the soundcard without noise was stupid. Instead I wrote a patch that allowed me to switch the HLT instruction on and off as the soundcard driver was loaded and unloaded.
Another brand of software is in no means a security benefit. You have now doubled the requirements for a secure environment: both vendors now must be equally as secure.
Now you are assuming that for the system as a whole to be secure, each part has to be secure. However that is not how secure systems are designed. Though DNS is really not designed with much security in mind, so it may all fail due to weaknesses in design. Had the design been secure multiple implementations would be an advantage. The best you can do right now to protect yourself from false results from a cracked root DNS server is to ask more than one, but that of course does not help against DoS attacks. And of course always use secure protocols like ssh and https on top of IP+UDP+TCP+DNS.
Asleep at the wheel
Did anybody BTW patent the wheel?
How about this: Send an email
I believe Echelon already possesses quite a few emails copyrighted by me.
Is Linux a legitimate solution to school districts facing a financial crunch?
Yes of course it is. Some people says students should be tought to use the software being used in the "real life". Why? If the students learn to acomplish the same task with cheaper software, how could that be bad?
But much rather than sticking with one choice of software, I'd see students trying a few different systems, so they can learn what are the differences and similarities between them, and they can learn how to learn using a new system, and they can make up their own minds about what they like and dislike. Because you cannot teach them how to use the software they are going to find themselves with in a few years, but you can theach them how to learn.
So let them try Linux, Unix, Windows, BSD, OSX, and let them find the best for each task.
With classic arcade games running on everything but toasters.
Don't say that too loud, somebody might invent a new toaster.
Is the binary for Linux large enough that if the bits were mapped to an image, it would work as a background image?
I typically use kernels of size 2-3MB in uncompressed form and 0.7-1MB in compressed form. I typically use a screen resolution between 1024x768 and 1600x1200, if you use a black/white picture that is 96-234KB. So the kernel is certainly large enough to make a black/white picture for your background. If you want 24-bit truecolor image it might be difficult to fill your entire screen.
the hypothetical 12-year-old listening to all the advice about girls everyone else is giving.
When I was 12, I could really have used that advice. And since that me at the age of 12 did not know what to do, I think he might have listened to me. And here is what I would have told him: If you really like that girl, say it to her. And say it to everybody else, heck even say it to your parents. (If you are afraid to talk about feelings with your parents write it in a letter, they will understand. They are not really out to get you.) If you give up on the girl you like, you are going to regret forever about the chances you might have missed. It might not work out, and she might be breaking your heart. But you should rather take the chance and let her break your heart than go and break your own heart. Stand up and fight for what you want.
Not true.
You are wrong.
The error is generated by the receiving mailserver
The server will respond with a 550 SMTP error code. It will even refuse to accept the message if the recipient does not exist. Once the server has refused to receive the message it does no more about it.
Besides, if bounce messages had to be created during the initial contact, relaying wouldn't work.
Again it is the sending server that will generate the error message. However in this case the same message is being transfered with SMTP multiple times. It is in the last SMTP session the error is detected, and it is the sending server in this session that needs to generate the error mail.
which they don't because they use fake email addresses.
Again they can abuse the SMTP protocol to know if the address worked or not. Nobody forces the spammer to use a mailserver for sending spam, they can use any software they desire. In particular the spammer can use software, that rather than generating the bounce simply generate lists with good and bad addresses. This does not work if the spammer uses an open relay, because the relay accepts the message, and it is then the relay that generates the bounce message. But the spammer can still hide by using an open proxy rather than an open relay.
What browser are you using?
I did the experiment last time I heard about this spam to unused addresses. I don't remember which browser I used, neither what problem I got. But I think it must have been a version of either Netscape or Mozilla.
at long last i will be able to get my BSOD
Oh, you want a BSOD? I have implemented a BSOD for Linux.
That is probably not possible.
Spammers have been bruteforcing honeypot domains for a few years now, sending spam to any and all combinations of letters and numbers.
Surely they can bruteforce some, but they cannot try all combinations. If the test addresses are long and random enough, they cannot be found by bruteforce. I wanted to try that experiment myself, but I wasn't even able to create an account on hotmail. When I was a hotmail user many years ago, it worked nicely in many browsers. And hotmail was even improved during the time I used it. I stopped using hotmail the day Microsoft bought it. But I did check if new mail arrived for some time after that, but Microsoft introduced more and more problems, and eventually I was no longer able to use it.
And BTW, some of the spammers verifying addresses on hotmail have been using my honeypot, which they thought was an open proxy. My honeypot said OK to all of those addresses, so there will be invalid email addresses on their lists.
one JPEG to ruin an entire TLD forever.
.jpg or .jpeg?
That got me thinking, how about a new TLD named
It has 64 sectors, numbered 0 to 63.
Nope, it has 63 sectors, numbered 1 to 63.
Track 0 has 64 sectors (generally)
The article says so, but that is incorrect. The upper limit is actually 63 sectors. And it has been a long time since I saw a harddisk with less. Because of the 504MB limit harddisk generaly used the 63 sectors to the limit. Later BIOSes did translation between IDE and BIOS interface geometry. That way the limit was pushed close to 8GB, but new problems were introduced. The 63 sectors was not touched by the translation, it was most convenient to keep that and only move bits between cylinder and side fields of the geometry. So today all harddisks on PCs have 62 sectors between the MBR and the first partition.
the software departement will find a workaround
Unfortunately that is not always possible.
Hell, any kludge that two kids throw together can be considered a 'Linux distro.'
Of course. Since Linux is only a kernel it doesn't take much. Creating a usable system is a whole different matter. Then again that depends on what you want to use the system for, you can create systems specialized for any purpose.