The minimum ethernet frame is 64bytes. The actual UDP packet contains 29 bytes of data. Those bytes then get a UDP + IP header attached to it -- that's usually about 40 bytes. The ethernet card (driver, whatever) adds the ethernet MAC header (14 bytes?) and puts it on the cable...
Every layer the packet passes through with add and remove any necessary padding for transmission. For example, if that 69 byte IP frame were to pass through an ATM (AAL5) network, it would need two 53byte ATM cells.
Alright, I've seen enough of this... "OS-9" is an operating system designed by Microware in the early 80's. It's an extremely good, small, fast RTOS. It's also Microware's registered trademark -- hell, it's the product's name! It ticks me off to see people automatically associate "OS-9" as MacOS 9.
I'm now taking bets on how long before Microware wakes up it's lawyers.
It doesn't matter how much magic C++ hides from you, it's just as prone to bugs. You're still going to read in an unknown amount of data so you don't know how much space to allocate. Data from an untrusted source should be untrusted - period. All the C++ string crap does is provide a Microsoft-like uniform bug. It makes it even easier for the programmer to not care about what is being done.
If one does not think about security when writing one's code, then I can assure it has none. Outlook is very good example -- nobody thought about what kind of evil lurks on the internet.
"If you cannot trust your users, who can you trust? Exactly."
The boot loader uses the BIOS interface to the floppy. As long as you only need the boot floppy for the initial kernel and ramdisk (assuming the ramdisk is loaded by the boot loader), everything should work ok. Two of my machines have LS-120's so I've got the same problem -- once the kernel takes over, the "floppy" is gone (win98's the only thing that hasn't cared so far.)
Not that I like Redhat, but have you tried this with RH6.1? (boot from floppy and install from a CD? I think that's the only one that won't ask for the other floppy.)
At the time, it was beyond most grad students as well. Factoring a 128bit number is not that hard in the present computer world. As you point out, there's nothing stopping anyone from creating a netrek client client. (Altho' I'd invest the time in factoring the keys -- all of them:-))
For the record, the verification can (and usually did) happen at any point during game play -- paranoid servers would usually repeat the verification at random intervals.
Ain't that the damned truth... the Linux solution to just about everything is "stick it in/proc." procfs has become nothing more than a run-down gheto full of pimps and drug dealers.
Seeing the ass-backwards shit being done via procfs makes me want to wipe the hard drive and install Win2000. (RC2 even)
That's not really true. People like the openness of Linux... any idiot can mess about with the kernel, discuss messing about with the kernel, and submit patches and/or new stuff for the main kernel. FreeBSD is a rather closed environment. Of course, these are both good and bad things.
Compared to Linux, BSD is alot harder to install and always has been. Can you install FreeBSD with only a floppy and a net connection?
Amen brother! They might as well add a checkbox/option for "install every known security flaw and a few no body knows about" along side the "[dist] backdoor" button.
Every time I install Redhat, it takes about 5 minutes to install (read: waste 1.3G of drive space) and then an hour to remove the brain damage and other worthless crap it installs.
Oh spare me your bullshit. Buffer overflows happen everywhere (even in assembly.) It's fairly trivial to audit code for overflow problems.
The true problem is not the compiler or the language. It's the idiot with the keyboard writing the program in the first place. Everyone wants to flog Microsoft for their unbelievably stupid programming, but no one every has a bad word about anyone else's (open source, freeware, GPL, et. al.) bad code. If you think about what you are doing, then you don't have these problems.
Don't hail Matrox too highly... there's still a large amount of documentation they are sitting on. They just released the G100 specs a few weeks ago -- the reason for not posting it with the others? none. (I don't guess there was much developer demand.)
The entire community is greatful for the microcode Matrox provided, but we'd be much happier with the WARP instruction set (so would their competitors:-)) Matrox is making this the biggest PR move of the century.
Matrox isn't exactly brimming over with published specs. They have yet to tell anyone (aside from XiG) about the MGA-TVO (aka. maven) chip used for TV output and the secondary RAMDAC on the dual head G400's. They also have not released the programming specs for the WARP enigne(s). They DID provide WARP microcode for the GLX project, but the GLX team doesn't know what they are pushing into the WARP engine.
Note: I don't hate Matrox for holding back the WARP docs (that's Gx00 internal microcode,) but sitting on the TVO docs is just stupid and counter productive -- we will figure it out.
(I'm fairly certain they are sitting on them just like they did the G100 specs. Read the last developer newsletter for the word on the G100 specs.)
I've used Accel-X and Metro-X... I went back to using XFree. The only good thing about a comercial X server is that you have someone who's obligated to fix your bugs. They may be able to get chip specs faster as they can eat NDAs with no problems, but that doesn't amount to much -- if the OEM doesn't want to release the specs to programmers, then don't buy their "crap".
All of the comercial servers appeared to be bloated and overly complicated. That's not to say XFree isn't either, but "everyone" knows how to deal with XFree; and XFree offers alot more control over the display setup (manufacturer specs are often lies.)
If you look closely, most of the cards listed are "professional" OpenGL cards. None of those are the cheap cards most gamers overclock to get an extra thousandth of a frame per second. How many "gamers" have an Oxygen?
If you were paying someone to host your web site, wouldn't you have at least ONE phone number to call and find out what the deal is when your site disappears? While it may be true that they sent no e-mail or other notification to the entire world, no one has said their customers were totally in the dark.
When your telephone service goes out, do you watch the evening news for three months looking for a story about it? No, you find a working phone and start bitching at the phone company.
Having previously worked for an ISP (for four years), I can tell you with certainty that this sort of shit happens every day. The problem is -- and I'm not raggin' on the accounting workers -- the highly motivated, commited (and commitABLE) people in charge of the servers, routers, and other hardware are not the people in charge of the accounting.
Most accounting types couldn't tell the difference between HTML and C++. They speak a different language and live in a different world than us uber-geeks. They do only what they are paid to do and generally nothing more. There's certainly a great deal of detachment -- sysadmins tend to care more for their machine than their own children (assuming they ever have any:-))
And to make things worse, there's usually very little linkage between accounting and engineering. One speaks chinese and the other dutch so there's no big debate why this is. Thus, accounting puts a flag beside some customer's name in a database somewhere to indicate "terminate." The engineer has scripts that pick those entries out and turn them off -- poof, account terminated.
Where I used to work, the accounting drones required EVERYONE to report every second spent doing company work -- I guess they had some notion of "billable hours." This made no sense to those of us salaried employees, but we recorded it anyway. When they saw how much time we spent, voluntarily mind you, there was a collect jaw-drop that shook the building. The whole notion of "working from home" and "the need to access the network from home" were worse than prison to them. (Yet they still wanted every damned thing in the network to be 100% every second of the day and shit kittens when it wasn't.)
[FWIW, telco's have similar problems, altho' much less frequently.]
Don't take that as an excuss to put your computer on the roof during a lightening storm.
I'm also assuming the case doesn't melt:-) The average 24 guage zinc computer case would melt from a direct hit -- possablly even catch on fire.
I've seen telecom equipment survive lightening strikes (of course, it reset itself.) I've seen planes hit by lightening repeatedly. I own a Radio Shack PoS phone that's been hit by lightening three times (that I'm certain of) and it still works _perfectly_.
OTOH, I've dug up telco trunk cables fused by a lightening strike. And I've cut down trees "killed" by lightening strikes -- that was impressive to see (from a good distance luckily)
Like charges repell. As such, the electrons will move as far away from each other as possible. That means they end up on the outer most surface of the cage.
"The Faraday Cage also inhibits all electrical..." That's not entirely true. It will block all in-bound RF. Unless the cage is grounded, RF can and will leak out. (And even if it's grounded, some RF will still leak out.) Computer cases 1) keep other RF sources from getting inside the case and thus onto unshielded SCSI cables, etc. and 2) cuts down the amount of RF the computer spits into the atmosphere. Trust me, your computer is still emiting RF -- the FCC limits how much and with what power.
"... First, let us assume the cow is a perfect sphere and the milk is evenly distributed over the interior surface..."
That's because you screwed up the grounding for the PC. Most modern PCs are very unforgiving of slight voltage changes. As for EMF from a static discharge... that's laughable. There are bigger magnetic fields inside the case from the hard drives and power supply leads.
If the computer wasn't grounded to the case, and the case were grounded, then a direct lightening strike to fully encased computer system would hardly be noticed. The problem is powering a computer inside the cage -- we all know what happens when lightening runs in on a power cord:-)
(Note: when grounding self to computer, touch the metal power supply housing then there won't any ground potential differences across the system.)
It's a lot of work to clean up the spooge and other shit it throws around. Specifically, it installs AIM no matter what you tell it to do. It takes a few minutes to remove it and then edit the registry to remove the damned AIM icon from the player.
The Xing mp3 engine inside realjukebox is the only reason I tolerate it at all. I do wish it would stop scanning the hard drives every time I start it -- it takes too damned long.
If you're crazy enough to enter the true information on the download forms, etc., then you're getting exactly what you deserve. Would you give the same information to some nut on the phone? I never put the truth on those types of forms. Do you really think my e-mail address is "foo@bar.com"?
People scream about privacy but then apply zero common sense to protecting it.
First rule: When a company stops giving you a paycheck you stop working for them.
By definition, you're not an employee; you're a volunteer.
As for what streaming media format one should use... that all depends on the party to whom you're streaming. Realplayer may be spyware, but it works just about everywhere. If you want the entire planet to be able to view it, MPEG is the solution. (maybe even AVI.)
no, the conversion should be simple... at least this type of thing is under *NIX -- it's just a commandline.
I'll be surprised if someone doesn't stand up and sue broadcast.com for some sort of predatory business practices, etc. And I'm certain some one will add this to the evidence of the "Microsoft Monopoly."
As I recall, there is a windows media player for *NIX (solaris at least.) The one time I tried it, it really sucked. The reason it (and others) run so well under windows is generally because they are partially built into the OS and they are designed to use windows APIs (published and private.) Thoses APIs don't exist under any *NIX. And no admin is going to patch the kernel to run IE. (People already frown on the VMware kernel "tweaks".)
As I recall, there's a flag in the ext2 super block that specifies the byteorder. I know there was some discussion years ago in this respect for sparc/linux systems -- back then, they file systems were not compat.
As for "MFS"... it make perfect sense for them to bury it in userland. Thus they'd never even have to think about releasing it. Of course, a new file system isn't a modification of the kernel as per the GPL. (Transarc has never released the source code for the linux AFS modules)
Reading that makes me wonder just how badly the TiVo works internally... a userspace file system cloned off of the worst possible file system: NFS.
So, who's up for reverse engineering their filesystem?
Slashdot Mirror
The minimum ethernet frame is 64bytes. The actual UDP packet contains 29 bytes of data. Those bytes then get a UDP + IP header attached to it -- that's usually about 40 bytes. The ethernet card (driver, whatever) adds the ethernet MAC header (14 bytes?) and puts it on the cable...
Every layer the packet passes through with add and remove any necessary padding for transmission. For example, if that 69 byte IP frame were to pass through an ATM (AAL5) network, it would need two 53byte ATM cells.
Alright, I've seen enough of this... "OS-9" is an operating system designed by Microware in the early 80's. It's an extremely good, small, fast RTOS. It's also Microware's registered trademark -- hell, it's the product's name! It ticks me off to see people automatically associate "OS-9" as MacOS 9.
I'm now taking bets on how long before Microware wakes up it's lawyers.
It doesn't matter how much magic C++ hides from you, it's just as prone to bugs. You're still going to read in an unknown amount of data so you don't know how much space to allocate. Data from an untrusted source should be untrusted - period. All the C++ string crap does is provide a Microsoft-like uniform bug. It makes it even easier for the programmer to not care about what is being done.
If one does not think about security when writing one's code, then I can assure it has none. Outlook is very good example -- nobody thought about what kind of evil lurks on the internet.
"If you cannot trust your users, who can you trust? Exactly."
The boot loader uses the BIOS interface to the floppy. As long as you only need the boot floppy for the initial kernel and ramdisk (assuming the ramdisk is loaded by the boot loader), everything should work ok. Two of my machines have LS-120's so I've got the same problem -- once the kernel takes over, the "floppy" is gone (win98's the only thing that hasn't cared so far.)
Not that I like Redhat, but have you tried this with RH6.1? (boot from floppy and install from a CD? I think that's the only one that won't ask for the other floppy.)
At the time, it was beyond most grad students as well. Factoring a 128bit number is not that hard in the present computer world. As you point out, there's nothing stopping anyone from creating a netrek client client. (Altho' I'd invest the time in factoring the keys -- all of them :-))
For the record, the verification can (and usually did) happen at any point during game play -- paranoid servers would usually repeat the verification at random intervals.
Ain't that the damned truth... the Linux solution to just about everything is "stick it in /proc." procfs has become nothing more than a run-down gheto full of pimps and drug dealers.
Seeing the ass-backwards shit being done via procfs makes me want to wipe the hard drive and install Win2000. (RC2 even)
That's not really true. People like the openness of Linux... any idiot can mess about with the kernel, discuss messing about with the kernel, and submit patches and/or new stuff for the main kernel. FreeBSD is a rather closed environment. Of course, these are both good and bad things.
Compared to Linux, BSD is alot harder to install and always has been. Can you install FreeBSD with only a floppy and a net connection?
Amen brother! They might as well add a checkbox/option for "install every known security flaw and a few no body knows about" along side the "[dist] backdoor" button.
Every time I install Redhat, it takes about 5 minutes to install (read: waste 1.3G of drive space) and then an hour to remove the brain damage and other worthless crap it installs.
I really miss the simplicity of SLS!
Oh spare me your bullshit. Buffer overflows happen everywhere (even in assembly.) It's fairly trivial to audit code for overflow problems.
The true problem is not the compiler or the language. It's the idiot with the keyboard writing the program in the first place. Everyone wants to flog Microsoft for their unbelievably stupid programming, but no one every has a bad word about anyone else's (open source, freeware, GPL, et. al.) bad code. If you think about what you are doing, then you don't have these problems.
Don't hail Matrox too highly... there's still a large amount of documentation they are sitting on. They just released the G100 specs a few weeks ago -- the reason for not posting it with the others? none. (I don't guess there was much developer demand.)
:-)) Matrox is making this the biggest PR move of the century.
The entire community is greatful for the microcode Matrox provided, but we'd be much happier with the WARP instruction set (so would their competitors
Matrox isn't exactly brimming over with published specs. They have yet to tell anyone (aside from XiG) about the MGA-TVO (aka. maven) chip used for TV output and the secondary RAMDAC on the dual head G400's. They also have not released the programming specs for the WARP enigne(s). They DID provide WARP microcode for the GLX project, but the GLX team doesn't know what they are pushing into the WARP engine.
Note: I don't hate Matrox for holding back the WARP docs (that's Gx00 internal microcode,) but sitting on the TVO docs is just stupid and counter productive -- we will figure it out.
(I'm fairly certain they are sitting on them just like they did the G100 specs. Read the last developer newsletter for the word on the G100 specs.)
I've used Accel-X and Metro-X... I went back to using XFree. The only good thing about a comercial X server is that you have someone who's obligated to fix your bugs. They may be able to get chip specs faster as they can eat NDAs with no problems, but that doesn't amount to much -- if the OEM doesn't want to release the specs to programmers, then don't buy their "crap".
All of the comercial servers appeared to be bloated and overly complicated. That's not to say XFree isn't either, but "everyone" knows how to deal with XFree; and XFree offers alot more control over the display setup (manufacturer specs are often lies.)
If you look closely, most of the cards listed are "professional" OpenGL cards. None of those are the cheap cards most gamers overclock to get an extra thousandth of a frame per second. How many "gamers" have an Oxygen?
If you were paying someone to host your web site, wouldn't you have at least ONE phone number to call and find out what the deal is when your site disappears? While it may be true that they sent no e-mail or other notification to the entire world, no one has said their customers were totally in the dark.
When your telephone service goes out, do you watch the evening news for three months looking for a story about it? No, you find a working phone and start bitching at the phone company.
Having previously worked for an ISP (for four years), I can tell you with certainty that this sort of shit happens every day. The problem is -- and I'm not raggin' on the accounting workers -- the highly motivated, commited (and commitABLE) people in charge of the servers, routers, and other hardware are not the people in charge of the accounting.
:-))
Most accounting types couldn't tell the difference between HTML and C++. They speak a different language and live in a different world than us uber-geeks. They do only what they are paid to do and generally nothing more. There's certainly a great deal of detachment -- sysadmins tend to care more for their machine than their own children (assuming they ever have any
And to make things worse, there's usually very little linkage between accounting and engineering. One speaks chinese and the other dutch so there's no big debate why this is. Thus, accounting puts a flag beside some customer's name in a database somewhere to indicate "terminate." The engineer has scripts that pick those entries out and turn them off -- poof, account terminated.
Where I used to work, the accounting drones required EVERYONE to report every second spent doing company work -- I guess they had some notion of "billable hours." This made no sense to those of us salaried employees, but we recorded it anyway. When they saw how much time we spent, voluntarily mind you, there was a collect jaw-drop that shook the building. The whole notion of "working from home" and "the need to access the network from home" were worse than prison to them. (Yet they still wanted every damned thing in the network to be 100% every second of the day and shit kittens when it wasn't.)
[FWIW, telco's have similar problems, altho' much less frequently.]
Has anyone mirrored this? Where'd they stick that stuff, on the other side of a 300 baud modem?
As I recall, Lucas was once quoted as saying he'd rather buy up every copy in existance and smash them with a hammer.
I vaguely remember this special (and one with some Ewoks?) from years and years ago. I didn't think it was that bad at the time.
Wasn't Lucas on one of "The Muppet Show" episodes?
Follow up/Disclaimer:
:-) The average 24 guage zinc computer case would melt from a direct hit -- possablly even catch on fire.
Don't take that as an excuss to put your computer on the roof during a lightening storm.
I'm also assuming the case doesn't melt
I've seen telecom equipment survive lightening strikes (of course, it reset itself.) I've seen planes hit by lightening repeatedly. I own a Radio Shack PoS phone that's been hit by lightening three times (that I'm certain of) and it still works _perfectly_.
OTOH, I've dug up telco trunk cables fused by a lightening strike. And I've cut down trees "killed" by lightening strikes -- that was impressive to see (from a good distance luckily)
Actually, the cage doesn't have to be grounded.
..."
Like charges repell. As such, the electrons will move as far away from each other as possible. That means they end up on the outer most surface of the cage.
"The Faraday Cage also inhibits all electrical
That's not entirely true. It will block all in-bound RF. Unless the cage is grounded, RF can and will leak out. (And even if it's grounded, some RF will still leak out.) Computer cases 1) keep other RF sources from getting inside the case and thus onto unshielded SCSI cables, etc. and 2) cuts down the amount of RF the computer spits into the atmosphere. Trust me, your computer is still emiting RF -- the FCC limits how much and with what power.
"... First, let us assume the cow is a perfect sphere and the milk is evenly distributed over the interior surface..."
That's because you screwed up the grounding for the PC. Most modern PCs are very unforgiving of slight voltage changes. As for EMF from a static discharge... that's laughable. There are bigger magnetic fields inside the case from the hard drives and power supply leads.
:-)
If the computer wasn't grounded to the case, and the case were grounded, then a direct lightening strike to fully encased computer system would hardly be noticed. The problem is powering a computer inside the cage -- we all know what happens when lightening runs in on a power cord
(Note: when grounding self to computer, touch the metal power supply housing then there won't any ground potential differences across the system.)
It's a lot of work to clean up the spooge and other shit it throws around. Specifically, it installs AIM no matter what you tell it to do. It takes a few minutes to remove it and then edit the registry to remove the damned AIM icon from the player.
The Xing mp3 engine inside realjukebox is the only reason I tolerate it at all. I do wish it would stop scanning the hard drives every time I start it -- it takes too damned long.
If you're crazy enough to enter the true information on the download forms, etc., then you're getting exactly what you deserve. Would you give the same information to some nut on the phone? I never put the truth on those types of forms. Do you really think my e-mail address is "foo@bar.com"?
People scream about privacy but then apply zero common sense to protecting it.
First rule: When a company stops giving you a paycheck you stop working for them.
By definition, you're not an employee; you're a volunteer.
As for what streaming media format one should use... that all depends on the party to whom you're streaming. Realplayer may be spyware, but it works just about everywhere. If you want the entire planet to be able to view it, MPEG is the solution. (maybe even AVI.)
no, the conversion should be simple... at least this type of thing is under *NIX -- it's just a commandline.
I'll be surprised if someone doesn't stand up and sue broadcast.com for some sort of predatory business practices, etc. And I'm certain some one will add this to the evidence of the "Microsoft Monopoly."
As I recall, there is a windows media player for *NIX (solaris at least.) The one time I tried it, it really sucked. The reason it (and others) run so well under windows is generally because they are partially built into the OS and they are designed to use windows APIs (published and private.) Thoses APIs don't exist under any *NIX. And no admin is going to patch the kernel to run IE. (People already frown on the VMware kernel "tweaks".)
As I recall, there's a flag in the ext2 super block that specifies the byteorder. I know there was some discussion years ago in this respect for sparc/linux systems -- back then, they file systems were not compat.
As for "MFS"... it make perfect sense for them to bury it in userland. Thus they'd never even have to think about releasing it. Of course, a new file system isn't a modification of the kernel as per the GPL. (Transarc has never released the source code for the linux AFS modules)
Reading that makes me wonder just how badly the TiVo works internally... a userspace file system cloned off of the worst possible file system: NFS.
So, who's up for reverse engineering their filesystem?