How do you define "death" of an undead, ie. zombie?
Well, the "Living Dead" standards is to a shotgun (or other destructive implement) to the brains of the organixm.
In the commercial/legal world, I guess that the equivalent would be some combination of Grand Jury Indictment, class action lawsuit (with protective seizure of assets) and/or Chapter 11.
Uninhibitted by royalties, the companies are able to produce drugs without the overhead of having to pay for all of that pesky R&D that no self-respecting drug company should have to pay for.
Rong. Canada does respect drug company patents. It used to force them to license their stuff at (I think) a 15% royalty, but then negotiated a change where drug companies maintain a level of exclusivity, but have to be responsible in terms of pricing.
Drug companies still make a tidy profit in Canada... It's just not hand-over-fist like it is in the US.
The real question is what is the cost of this process, and how does it compare with laser modeling techniques?
In terms of computing, I expect this to be pretty cheap (at least from their quick description). The hard part is that 4 flash are going to eat your batteries like no tomorrow, and if you're trying to do snapshots, you want to grab the 4 images as quickly as possible.
The edge value for any given pixel is essentially max(4_pixels)-min(4_pixels). If you add a second buffer to the camera with 5 layers (R,G,B,Min,Max) you can do the computation as you read
from the buffer, If you add 2 bits to the R,G,B layers then you can compute an average without overflow. with a 600megabyte/second memory bandwidth, you should be able to take and calculate a 4megapixel image in under 1/30 of a second, and a 1 megapixel image in about 1/125.
The limiting factor, at that point, becomes the speed of your imaging chip.
Back around 1990 I got my first 1GB drive. It cost us %3500CDN (almost exactly $2900US -- not including 15 years of inflation).
It was a full-height 5" drive (most CD players are half-height). so this would actually make the 1GB drive larger than this 1TB unit.
Back then, 1GB was a friggin HUGE ammount of data to have in one box. Today it's a novelty thumb drive.
Conversely, might it open up SCO to litigation for having essentially sneakwrapped this "linux license" into their product?
Getting something for free is rarely a cause for litigation per se. I would think (IANAL) that it would only be to the extent to which the bundling caused a liability on the part of the receiving company.
Where a company may be able to claim a cause of liability because of SCO giving them a license might be if SCO tells the world "Company X bought Linux Licenses from us!", and then the company has to spend big money convincing their customers and investors that they're not brain-dead and/or SCO supporters.
On a different issue: If the company in question didn't sign off on the agreement, then there should be nothing that SCO can hold the company to. On the other hand, if SCO tries to sue the compamy, the company should be able to wave the license and say "see! They promised they wouldn't!". Of course, before I did something like that, I'd ask my lawyers to make sure that such an implicit acceptance of the license didn't create some sort of extra liability to SCO.
Remember that, with the exception of DV-1, I think that most of the companies who 'bought' an SCO license didn't even realize it. The way that they have things set up, people who buy other SCO software sometimes end up with a Linux License. Part of the reason why they're not identifying the companies who have purchased SCO Linux licenses may be that they don't want all of those companies to know what they've bought.
Computerland, for example, ended up with a bunch of Linux licenses as a reault of a (supposedly unrelated) out-of-court settlement with the Canopy group. They acted very surprised when it was announced that they were among the list of companies owning Linux licenses.
If your company has any vestigal connection to either SCO or Canopy, then it may turn out to be one of the '20 or 30' companies with Linux licenses.
AFAICT the only valid business logic for buying a LINUX license is to comfort skitish investors and/or customers with the knowledge that there is no legal liability. The only other reason for getting a Linux license is that it's a side effect of getting something else that your company needs from SCO. Companies in such a position who care, at all, about such acquisitions may actually be ashamed of them.
My quick way to describe the 8986 was "an 8085 with formalized bank
switching". as an 8 bit processor, it was a fine step up. As a 16
bit processor it was essentially braindead.
Less than 4 years after the IBM-PC was released, people were already
adding bank switching on top of the bank switching (EMS/XMS), and Inte
was creating (incompatible) replacements for the original (brain-damaged)
memory model. The memory models for the 8086, 80185, 80285 and 80386 were
all different. and the 80386 was so different from the others
that real and protected mode might as well have been designed as a
2-core processor system.
It's not that Intel was coming up with all of these memory models
just for the fun of it. They knew that the 8086 memory model
was insufficient and they were essentially flailing around trying to
find something eles that didn't suck so much. That didn't happen until
they dompletely redesigned the processor to included 32 bit registers
(like just about every other 16 bit processor had since 1980)
When you look at all of the useless garbage that Intel came up
with, it seems like they suffer frm the same fundamental problem that
Microsoft has -- Design by Marketing. Their solutions made perfect sense
from a marketing point of view, but any competent engineer looking at
the result would often go "ewwwwww!".
I used to say that in the '80s.
I considered the 8086 to be one of the most brainded of the 16 bit chips that you could have designed if you were trying.... The 8086 did not make the bigtime because it was a good chip design. Quite the opposite.
I'm very honestly convinced that the reason why IBM chose it for their "IBM PC(tm)" is that it was way too braindead to have any real hope of competing against their big-money IBM/370-family mainframe engines.
It took them 4 tries (8086, 80186, 80286, 80386) to get the '86 family morphed into something half-assed, and it wasn't the only CPU family that came out of Intel and got laughed at until it died of shame.
(someone else can provide the list of {sh,n}ames).
Have you ever heard of journaling??? Log the changes the user is making. When the change log gets too big, write a checkpoint of the current file, and then empty the checkpoint log.
VI has had the checkpointing facility since the very early '80s. The actual file is never written to, but if the system borks before a final save is done, you have the option of restarting with the original file, or doing a restore from the last saved event (usually 1 or 2 changes behind what the user was actually doing when VI aborted).
As long as you're aloways appending to the file, you should either have a clean operation or an incomplete one which doesn't get used
One quick note about ntpasswd is that I ran into a situation where it only mounted the drive Read-Only... Everything worked up until the point where it copied the temp file back to the disk (which, of course failed).
What I managed to do in that case was to copy the file from/tmp to a permanent location (i think I scped it to another Linux box) Then I used a Knoppix disk and captive-ntfs to copy the saved temp files to the NT drive (names are pretty obvious).
That's the difference between America and Iraq. In Iraq, if an american uses an RC truck to get rid of a, uhm, "pest", it's known as 'counter insurgency'. If the same person does that in Washington, it's 'terrorism'.
But what is more likely is biological printers that grow stuff out of cells.
My mother had one of those.
(Don't know if it still works, thought.)
Prove my vote counted!
It did. It counted for Bush.
Feed the Fishers Red Herrings
on
Gone Phishing?
·
· Score: 1
One thing that I can think of to cut down on these scams is to start feeding the phishers red herrings -- fake accounts and IDs that -- when they try to use them, they get trapped. Something to make them worry.
I suggested it to at least one bank, but they didn't seem to give a rat's ass about the idea. Perhaps they get too much money back from their insurers?
A lot of social commentary shows up in SF and Fantasy. In some cases it was the only way to get such commentary past the censors ("We're not talking about you -- unless you're claiming to be a Space Cadet...").
( There Ain't No Such Thing As A Free Piece Of Cheese )
We had pad-tethered mice almost 20 years ago (although it was the mice that were wired back then). We don't have them any more because people didn't seem to like the pad tether. Now these people are giving us a new mouse that provies all of the advantages of a wireless mouse -- excep that it's tethered to a wired mousepad.
In other words, it's really wireless in name only.
To get a subpoena you need to send an application to a judge specifying precisely why you want it and what you want, then convince the judge to say "yes". The long part of this is handing the paper to the Judge and convincing him/her to sign it.
In theory there should never be a full automating of this process, since that would also imply that the requests get rubber-stamped.
Besides, you're gonna be spending way more time in the initial investigation (to get enough evidence to convince the judge) and in the subsequent analysis of the resulting data (presuming that you get any) than you will typing the details of the subpoena into the boilerplate for the application.
You can't get both TCP and UDP into the same packet without encapsulation.
Almost all of the properties that UDP shares with IP are a result of it sitting on top of IP. Kinda like the way that a minimal subclass shares properties with it's parent class.
Both TCP and UDP sit on top of IP, as does ICMP and a number of other protocls -- in fact both TCP and UDP sit on top of either IP4 or IP6. You don't have to change the tcp/udp code when you switch to IP4 -- just the IP code (that's the whole purpose of the layering process).
Re:Reliable... (TCP/IP vs TCP/UDP)
on
P2P Through Firewalls
·
· Score: 2, Insightful
UDP isn't similar to IP. It sits on top of IP, just like TCP does. The main difference between UDP and TCP is that TCP has reliability built in. The protocol will automatically re-order out of sequence packets, figure out which ones are missing and have them retransmitted before it delivers the data to the recipient.
With UDP, if you want reliable transmission, then you have to do all that work yourself. If you have a generally reliable link, then this can be very cheap. It's also very cheap if you only want to send a little bit of data (say, one line of text).
With TCP it takes at least three packets just to open the connection -- before you can even transmit data -- and then another three to close it. That gives you a minimum of 6 packets with no actual transmitted data.
With UDP, a very simple transaction can consist of
a -> B (( DATA ))
B -> a (( ACK: CRC was X) [[ optional ]]
As long as the checksum is good, A need do nothing. If the checksum was bad (or no ACK comes back) then A can retransmit). If you don't mind losing the data altogether, then even the ACK is unnecessary.
UDP can be fast and cheap, but to avoid beating the internet to death on bulk transfers it depends on the application being well-designed to prevent bad side effects.... That intelligence is built into TCP (but at a cost).
As long as you can handle losing the occasional packet and/or having packets arrive out of order and you don't threaten hog the entire bandwidth of your pipe, then UDP is fine.
If you don't want to worry about the above, and you're willing to put up with the overhead of TCP, then TCP is your best bet. This turns out to be the case for the majority of applications, which is why TCP seems to be way more common than UDP -- SO much so that many people don't even relize that you can have IP without TCP (( the 'TCP/IP' misnomer contributes to this)).
So wierd: I woke up this morning with the same idea in my head. I finally tried using gtk-gnutella this weekend, and noticed that it was having problems with my firewall. I started thinking about how to punch thru a firewall without having to manually reconfigure and came up with the same solution.
Essentially, UDP is a stateless system, so stateful firewalls don't have SYN packets to signal the start of a connection, so you can do the following
Machine A sends a UDP packet from port X to machine B Port Y, and
Machine A's firewall will create a connection for the outgoing address/port pair
(Machine B's firewall will drop the packet because there is no connection associated with it)
Machine B then sends a UDP packet from port Y to machine A Port X
Machine B's firewall will create a connection based on the address/port pair.
Machine A's firewall will accept (and route) the connection as belonging to a legitmate connection.
All further connections between A and B based on this address/port pair should work until further notice
Note that this will not work on firewalls that use port randomizing on outgoing connections. More work has to be done to make that work.
This also (obviously) requires a mediating connection where the two can agree on what addresses and ports to send from and to.
(If a machine is inside of a NAT it may not realize what it's post-translated address is).
As long as you don't have port randomizing, this should work on both NATed and non-NATed statefull firewalls.
I like Linux, the tinkering and customization and the terminal, but Windows doesn't suck,
You'd be surprised at the number of people that I run into who
would disagree with you -- people who are sick of having to send their computer to some strange to get 'fixed' a few times a year. They're very happy to dump windows -- they're just afraid that they'll lose their access to the tools (really the information) that they currently have.
My most recent convert is a Medical intern who just doesn't have the time to mess with WIndows repeatedly dieing on her. The stability and security of Linux is what calls to her. She doesn't have the time to deal with Windows (or "work on her computer", as you put it).
The roommate who turned into a Linux cursader is a photographer. He is, similarly, uninterested in 'working on his computer'. His other choice is a MAC.
In Canada, a very high percentage of people have broadband, so winmodems are less and less an issue. I did, however, have one person who couldn't switch from Windows because he had a very esotheric MP3 voice recorder that I could not find a Linuz driver for. Other than that, hew was very happy with Linux.
Slashdot Mirror
Well, the "Living Dead" standards is to a shotgun (or other destructive implement) to the brains of the organixm.
In the commercial/legal world, I guess that the equivalent would be some combination of Grand Jury Indictment, class action lawsuit (with protective seizure of assets) and/or Chapter 11.
They never had a Linux version.
Rong. Canada does respect drug company patents. It used to force them to license their stuff at (I think) a 15% royalty, but then negotiated a change where drug companies maintain a level of exclusivity, but have to be responsible in terms of pricing.
Drug companies still make a tidy profit in Canada... It's just not hand-over-fist like it is in the US.
In terms of computing, I expect this to be pretty cheap (at least from their quick description). The hard part is that 4 flash are going to eat your batteries like no tomorrow, and if you're trying to do snapshots, you want to grab the 4 images as quickly as possible.
The edge value for any given pixel is essentially max(4_pixels)-min(4_pixels). If you add a second buffer to the camera with 5 layers (R,G,B,Min,Max) you can do the computation as you read from the buffer, If you add 2 bits to the R,G,B layers then you can compute an average without overflow. with a 600megabyte/second memory bandwidth, you should be able to take and calculate a 4megapixel image in under 1/30 of a second, and a 1 megapixel image in about 1/125.
The limiting factor, at that point, becomes the speed of your imaging chip.
Back then, 1GB was a friggin HUGE ammount of data to have in one box. Today it's a novelty thumb drive.
Getting something for free is rarely a cause for litigation per se. I would think (IANAL) that it would only be to the extent to which the bundling caused a liability on the part of the receiving company.
Where a company may be able to claim a cause of liability because of SCO giving them a license might be if SCO tells the world "Company X bought Linux Licenses from us!", and then the company has to spend big money convincing their customers and investors that they're not brain-dead and/or SCO supporters.
On a different issue: If the company in question didn't sign off on the agreement, then there should be nothing that SCO can hold the company to. On the other hand, if SCO tries to sue the compamy, the company should be able to wave the license and say "see! They promised they wouldn't!". Of course, before I did something like that, I'd ask my lawyers to make sure that such an implicit acceptance of the license didn't create some sort of extra liability to SCO.
Computerland, for example, ended up with a bunch of Linux licenses as a reault of a (supposedly unrelated) out-of-court settlement with the Canopy group. They acted very surprised when it was announced that they were among the list of companies owning Linux licenses.
If your company has any vestigal connection to either SCO or Canopy, then it may turn out to be one of the '20 or 30' companies with Linux licenses.
AFAICT the only valid business logic for buying a LINUX license is to comfort skitish investors and/or customers with the knowledge that there is no legal liability. The only other reason for getting a Linux license is that it's a side effect of getting something else that your company needs from SCO. Companies in such a position who care, at all, about such acquisitions may actually be ashamed of them.
Less than 4 years after the IBM-PC was released, people were already adding bank switching on top of the bank switching (EMS/XMS), and Inte was creating (incompatible) replacements for the original (brain-damaged) memory model. The memory models for the 8086, 80185, 80285 and 80386 were all different. and the 80386 was so different from the others that real and protected mode might as well have been designed as a 2-core processor system.
It's not that Intel was coming up with all of these memory models just for the fun of it. They knew that the 8086 memory model was insufficient and they were essentially flailing around trying to find something eles that didn't suck so much. That didn't happen until they dompletely redesigned the processor to included 32 bit registers (like just about every other 16 bit processor had since 1980)
When you look at all of the useless garbage that Intel came up with, it seems like they suffer frm the same fundamental problem that Microsoft has -- Design by Marketing. Their solutions made perfect sense from a marketing point of view, but any competent engineer looking at the result would often go "ewwwwww!".
I'm very honestly convinced that the reason why IBM chose it for their "IBM PC(tm)" is that it was way too braindead to have any real hope of competing against their big-money IBM/370-family mainframe engines.
It took them 4 tries (8086, 80186, 80286, 80386) to get the '86 family morphed into something half-assed, and it wasn't the only CPU family that came out of Intel and got laughed at until it died of shame. (someone else can provide the list of {sh,n}ames).
VI has had the checkpointing facility since the very early '80s. The actual file is never written to, but if the system borks before a final save is done, you have the option of restarting with the original file, or doing a restore from the last saved event (usually 1 or 2 changes behind what the user was actually doing when VI aborted).
As long as you're aloways appending to the file, you should either have a clean operation or an incomplete one which doesn't get used
What I managed to do in that case was to copy the file from /tmp to a permanent location (i think I scped it to another Linux box) Then I used a Knoppix disk and captive-ntfs to copy the saved temp files to the NT drive (names are pretty obvious).
That's the difference between America and Iraq. In Iraq, if an american uses an RC truck to get rid of a, uhm, "pest", it's known as 'counter insurgency'. If the same person does that in Washington, it's 'terrorism'.
You don't need to be able to see the user's eyelashes to do accurate gesture recognition. A $20 webcam sounds like more than enough resolution to me.
My mother had one of those.
(Don't know if it still works, thought.)
Prove my vote counted!
It did. It counted for Bush.
I suggested it to at least one bank, but they didn't seem to give a rat's ass about the idea. Perhaps they get too much money back from their insurers?
A lot of social commentary shows up in SF and Fantasy. In some cases it was the only way to get such commentary past the censors ("We're not talking about you -- unless you're claiming to be a Space Cadet...").
Could somebody who actually made it to these web pages post their content so that the rest of us can read them.
I wonder what Microsoft is gonna have Jon charged with?
It would also be really easy for President Bush Jr to release the Nuclear Launch Codes to Al Qaeda.
The difference is: You never know what Bush will do tomorrow.
We had pad-tethered mice almost 20 years ago (although it was the mice that were wired back then). We don't have them any more because people didn't seem to like the pad tether. Now these people are giving us a new mouse that provies all of the advantages of a wireless mouse -- excep that it's tethered to a wired mousepad.
In other words, it's really wireless in name only.
To get a subpoena you need to send an application to a judge specifying precisely why you want it and what you want, then convince the judge to say "yes". The long part of this is handing the paper to the Judge and convincing him/her to sign it.
In theory there should never be a full automating of this process, since that would also imply that the requests get rubber-stamped.
Besides, you're gonna be spending way more time in the initial investigation (to get enough evidence to convince the judge) and in the subsequent analysis of the resulting data (presuming that you get any) than you will typing the details of the subpoena into the boilerplate for the application.
You can't get both TCP and UDP into the same packet without encapsulation.
Almost all of the properties that UDP shares with IP are a result of it sitting on top of IP. Kinda like the way that a minimal subclass shares properties with it's parent class.
Both TCP and UDP sit on top of IP, as does ICMP and a number of other protocls -- in fact both TCP and UDP sit on top of either IP4 or IP6. You don't have to change the tcp/udp code when you switch to IP4 -- just the IP code (that's the whole purpose of the layering process).
With UDP, if you want reliable transmission, then you have to do all that work yourself. If you have a generally reliable link, then this can be very cheap. It's also very cheap if you only want to send a little bit of data (say, one line of text).
With TCP it takes at least three packets just to open the connection -- before you can even transmit data -- and then another three to close it. That gives you a minimum of 6 packets with no actual transmitted data.
With UDP, a very simple transaction can consist of
a -> B (( DATA ))
B -> a (( ACK: CRC was X) [[ optional ]]
As long as the checksum is good, A need do nothing. If the checksum was bad (or no ACK comes back) then A can retransmit). If you don't mind losing the data altogether, then even the ACK is unnecessary.
UDP can be fast and cheap, but to avoid beating the internet to death on bulk transfers it depends on the application being well-designed to prevent bad side effects.... That intelligence is built into TCP (but at a cost).
As long as you can handle losing the occasional packet and/or having packets arrive out of order and you don't threaten hog the entire bandwidth of your pipe, then UDP is fine.
If you don't want to worry about the above, and you're willing to put up with the overhead of TCP, then TCP is your best bet. This turns out to be the case for the majority of applications, which is why TCP seems to be way more common than UDP -- SO much so that many people don't even relize that you can have IP without TCP (( the 'TCP/IP' misnomer contributes to this)).
Essentially, UDP is a stateless system, so stateful firewalls don't have SYN packets to signal the start of a connection, so you can do the following
- Machine A sends a UDP packet from port X to machine B Port Y, and
- Machine A's firewall will create a connection for the outgoing address/port pair
- (Machine B's firewall will drop the packet because there is no connection associated with it)
- Machine B then sends a UDP packet from port Y to machine A Port X
- Machine B's firewall will create a connection based on the address/port pair.
- Machine A's firewall will accept (and route) the connection as belonging to a legitmate connection.
- All further connections between A and B based on this address/port pair should work until further notice
Note that this will not work on firewalls that use port randomizing on outgoing connections. More work has to be done to make that work. This also (obviously) requires a mediating connection where the two can agree on what addresses and ports to send from and to. (If a machine is inside of a NAT it may not realize what it's post-translated address is).As long as you don't have port randomizing, this should work on both NATed and non-NATed statefull firewalls.
You'd be surprised at the number of people that I run into who would disagree with you -- people who are sick of having to send their computer to some strange to get 'fixed' a few times a year. They're very happy to dump windows -- they're just afraid that they'll lose their access to the tools (really the information) that they currently have.
My most recent convert is a Medical intern who just doesn't have the time to mess with WIndows repeatedly dieing on her. The stability and security of Linux is what calls to her. She doesn't have the time to deal with Windows (or "work on her computer", as you put it).
The roommate who turned into a Linux cursader is a photographer. He is, similarly, uninterested in 'working on his computer'. His other choice is a MAC.
In Canada, a very high percentage of people have broadband, so winmodems are less and less an issue. I did, however, have one person who couldn't switch from Windows because he had a very esotheric MP3 voice recorder that I could not find a Linuz driver for. Other than that, hew was very happy with Linux.