I think that an advagato styled trust metric might work out. Your mail server would be trusted by your upstream ISP's mail server, who would create trust relationships with a selection of other ISPs, enough that the large whole of the 'net would trust them.
I think getting a signifigant percentage of the mail servers in a trust system would be much less troublesome than trying to get every mail address on the 'net in a trust system.
I think a better resolution to the problem is to enforce a certain amount of purity in the mail headers.
If you are spam, you should mark your message as being such. If you are a mailing list, you should mark your message as being such.
And then we need to have a network of trust between the mail servers. Something lightweight enough that it works 90% of the time. Servers who are trusted are trusted that they will send out mail with proper headers. Servers who aren't trusted will get their mail bounced most of the time.
Thus, spam can be dropped on the floor at the option of any mail server. And server admins who don't mark spam as spam are marked as untrusted servers. At the option of the country that the mail server exists in, this can be declared as fraud.
I wrote up some notes on it on my webpage but I'm not sure how well it would really work in practice.
Actually, the biggest drive for Phillips is that they do not own any media companies and are big. They made quite a bit of money on the consumer CD-R drives.
So they are behaving quite rationally by protecting their CD-R drive income..;)
If you are going to teach computer architecture, teach a real architecture with warts. The MIPS architecture has only one wart, and that was mostly there as a leftover from previous assumptions, as far as I can tell.
I think you will get much more mileage out of teaching students on the x86 architecture. You can get a good emulator and also run things on a regular PC, depending. Most of your problem sets should fit on a 1.44MB floppy, which means that you can just boot an existing computer off of the floppy drive. Write a few pieces of code to make the student's life easier -- i.e. loading an executable image into memory and stuff.
1. Wired phones are cheaper than cordless phones but people buy cordless phones anyways.
2. A power plug is still a plug. Even though there's a power plug riight next to my chair in the living room, I will only plug in when the batteries on my laptop are low.
So therefore, I feel that the big usage for this is digital convergence -- the computer on your fridge to keep track of what you need to get from the store, the computer in your thermostat to let you set the temperature from anywhere, etc.
Most of them are large items that rarely change, especially if you are in an apartment.
Plus, these things are not here yet and probably won't be here for another few years. In the near term, people are only networking computers.
Plus, the digital convergence future will probably require a 802.11x like system anyways, because your MP3 player/computer in your car will hook up to your home network, you will be carying handheld/tablet computers around your house, etc. All of which will probably make you want to get 802.11x anyways.
Nope.. For a while, there were a whole crapload of products you plugged into your phoneline and you could use it for voice, or for data communications inside of the house at a few megabits.
Didn't really catch on.
I like my DSL, tho. Covad emerged from bankruptcy and I can run servers on speakeasy.net.;)
I mean, really. Your average non-savy user won't try to share a connection because they don't know it's possible. The Wireless product makers are cleaning up the partially-savy and geek markets. Hardcore geeks are wiring their houses with cat 5, etc.
Remember the telephone-over-powerline products to give you extra extensions? People just purchased cordless phones instead. Nobody bothers having a ton of wired extensions. And the network-over-a-phoneline? Just about nobody uses that one, either.
If people aren't wiring with the cables the designer intended (power on the power plug, telephone on the phone jack, network on the cat 5, cable on the coax, etc) they'll use wireless.
The backup plan, of course, is for Speakeasy to just buy lines from the phone company directly.
At least, I hope. I hope that in 5 years, there is still an ISP that will let you have static IP addresses and servers for a reasonable premium over normal rates. I love my speakeasy connection. If Covad sticks around for a while longer and is profitable, I'll do what I intended to do a while back and start hosting my own services, instead of paying for an ISP.
Of course, Covad is becomming part of the phone company by partnering with SBC.
If you solve an NP complete problem in O(n^65535) time, you have just shown that P == NP. However, you still wouldn't be able to crack any of the NP complete problems that cryptography is based on in a reasonable amount of time.
Because trust me, if it was a low exponent for x, we'd have found it already.;)
Besides, they'd just move to problems that are not NP complete for the popular cryptography algorythims. Cryptographers are too smart for their own good, you know.
Every few years, buy a good high-capacity removable disk drive and move everything over to the new disk format.
So take the thousands of CDRs and copy them to DVD-Rs. Wait a few years, buy the next format that comes out, copy the DVD-Rs to that. Rinse, repeat, etc.
And make sure that you burn two copies.
Also note that for sitting around and doing nothing, the CDRs have a long life. Using CDRs shortens the life considerably.
I know that I've at least seen advertising materials for Matrox's low-profile PCI video cards. I couldn't find it on Matrox's website, so you might have to contact them or a reseller that works on low-profile PCI products.
1) Easier implementation of large filesystems. 2^64*512 bytes of disk space per filesystem should be good enough for quite a few increments of Moore's Law. Ditto for larger than 4 gig files.
2) 64 bit processors take better to certain implementations of NUMA. SGI's implementation of NUMA gives each processor a range of memory that is local to that processor. If you had a 64 processor NUMA cluster, you'd have 64 megs local to each processor with 32 bit processors. You could have a few gigs per processor with 64-bit addressing.
3) With 64-bit processors, it's easier to map a file to memory again, without needing to map individual chunks. Over the near term, you could map your entire disk drive to memory space.
4) There are cases (i.e. bit packing) that don't take too well to vectorized MMX/SSE/etc. processing but do take well to 64 bit registers.
5) The ability to segment your memory space without creating annoying limitations. As in, you can have the lower 8,388,608 terabytes of RAM reserved for the user and the upper 8,388,608 terabytes of RAM reserved for the kernel. As opposed to Windows 2k, which leaves 2 gigs for the user and 2 gigs for the kernel. With the possibility of 3 gigs for the user, if you are running a higher-end version.
6) The ability to cache a data structure in the RAM attached to a given machine instead of buying solid state disk drives or other such things.
Most modern architectures don't have multi-word instructions. If I remember correctly, even the 68000 processor series used single-word instructions.
The P4 doesn't always use 20 bit addressing. It can, but it doesn't need to. There are pure 32 bit memory accessor operations that let you skip the translation.
Most non-x86 modern 32 bit processors (32 bit PowerPC processors, 32 bit MIPS processors, 32 bit SPARC processors etc.) have more registers than the PC. It does net you great performance increases without the register renaming features the latest x86 CPUs have.
64 bit processors, in their own right, don't net your average user much extra power right now. But we are reaching the point where mid and high end computing needs 64 bits. It's only a matter of time before consumer PCs start to need 64-bits.
Most programmers I know who actually have a CS degree or equivelent (That didn't come from Bob's Orange Smoothie and School of Computer Science) can pick up a new programming language fast enough to be useful.
But then, my definition of a mid-level programmer must be different than yours is.
I've had decent luck sending things in the origional packaging (So they can't gripe about the packaging) and sending it the first level above UPS ground.
And you have to watch your insurance carefully. Often times, they insure everything BUT electronics.
Yeah, but what does the leg up give it? SOAP is suffering from the too-many-cooks sort of problem that is taking it away from being a simple protocal.
And, since the standard isn't complete but people are implementing it, it probably won't be any more standard than old-fashioned RPC, where everybody extended it and such to build your own standard. (i.e. DCOM is technically an implementation of RPC, but not really)
What's most likely to happen is that nobody will win. MS people will use MS's purposefully broken SOAP implementation, some others will use real SOAP, and others will just use XML-RPC.
Which will eliminate many of the supposed advantages of any of the protocals.
Slashdot Mirror
I think that an advagato styled trust metric might work out. Your mail server would be trusted by your upstream ISP's mail server, who would create trust relationships with a selection of other ISPs, enough that the large whole of the 'net would trust them.
I think getting a signifigant percentage of the mail servers in a trust system would be much less troublesome than trying to get every mail address on the 'net in a trust system.
I think a better resolution to the problem is to enforce a certain amount of purity in the mail headers.
If you are spam, you should mark your message as being such. If you are a mailing list, you should mark your message as being such.
And then we need to have a network of trust between the mail servers. Something lightweight enough that it works 90% of the time. Servers who are trusted are trusted that they will send out mail with proper headers. Servers who aren't trusted will get their mail bounced most of the time.
Thus, spam can be dropped on the floor at the option of any mail server. And server admins who don't mark spam as spam are marked as untrusted servers. At the option of the country that the mail server exists in, this can be declared as fraud.
I wrote up some notes on it on my webpage but I'm not sure how well it would really work in practice.
Actually, the biggest drive for Phillips is that they do not own any media companies and are big. They made quite a bit of money on the consumer CD-R drives.
;)
So they are behaving quite rationally by protecting their CD-R drive income..
Yes, you are missing something...
This ruling means that they don't need to get a warant to sniff, just a court order.
So it's not rampant abuse of the system, but the privacy people see it as the thin end of a wedge.
If you are going to teach computer architecture, teach a real architecture with warts. The MIPS architecture has only one wart, and that was mostly there as a leftover from previous assumptions, as far as I can tell.
I think you will get much more mileage out of teaching students on the x86 architecture. You can get a good emulator and also run things on a regular PC, depending. Most of your problem sets should fit on a 1.44MB floppy, which means that you can just boot an existing computer off of the floppy drive. Write a few pieces of code to make the student's life easier -- i.e. loading an executable image into memory and stuff.
True...
But
1. Wired phones are cheaper than cordless phones but people buy cordless phones anyways.
2. A power plug is still a plug. Even though there's a power plug riight next to my chair in the living room, I will only plug in when the batteries on my laptop are low.
So therefore, I feel that the big usage for this is digital convergence -- the computer on your fridge to keep track of what you need to get from the store, the computer in your thermostat to let you set the temperature from anywhere, etc.
Most of them are large items that rarely change, especially if you are in an apartment.
Plus, these things are not here yet and probably won't be here for another few years. In the near term, people are only networking computers.
Plus, the digital convergence future will probably require a 802.11x like system anyways, because your MP3 player/computer in your car will hook up to your home network, you will be carying handheld/tablet computers around your house, etc. All of which will probably make you want to get 802.11x anyways.
Supposedly the main thing that kills any leakage is the transformer.
But, as has been already pointed out here, it's not perfect. That's probably why there's 56 bit encryption -- just in case.
And there's probably a cable-length limitation, too.
Nope.. For a while, there were a whole crapload of products you plugged into your phoneline and you could use it for voice, or for data communications inside of the house at a few megabits.
;)
Didn't really catch on.
I like my DSL, tho. Covad emerged from bankruptcy and I can run servers on speakeasy.net.
Is it just me, or is this product useless?
I mean, really. Your average non-savy user won't try to share a connection because they don't know it's possible. The Wireless product makers are cleaning up the partially-savy and geek markets. Hardcore geeks are wiring their houses with cat 5, etc.
Remember the telephone-over-powerline products to give you extra extensions? People just purchased cordless phones instead. Nobody bothers having a ton of wired extensions. And the network-over-a-phoneline? Just about nobody uses that one, either.
If people aren't wiring with the cables the designer intended (power on the power plug, telephone on the phone jack, network on the cat 5, cable on the coax, etc) they'll use wireless.
Thanks for the clarification.
;)
I do know that in some markets where there are no good DSL providers, you can IP tunnel through another ISP just to get the IP address.
Of course, that's an awful hack.
So do I. ;)
But I hope that I'll be able to do it while being able to have FTP, VNC, HTTP, and STMP servers.
The backup plan, of course, is for Speakeasy to just buy lines from the phone company directly.
At least, I hope. I hope that in 5 years, there is still an ISP that will let you have static IP addresses and servers for a reasonable premium over normal rates. I love my speakeasy connection. If Covad sticks around for a while longer and is profitable, I'll do what I intended to do a while back and start hosting my own services, instead of paying for an ISP.
Of course, Covad is becomming part of the phone company by partnering with SBC.
If you solve an NP complete problem in O(n^65535) time, you have just shown that P == NP. However, you still wouldn't be able to crack any of the NP complete problems that cryptography is based on in a reasonable amount of time.
;)
Because trust me, if it was a low exponent for x, we'd have found it already.
Besides, they'd just move to problems that are not NP complete for the popular cryptography algorythims. Cryptographers are too smart for their own good, you know.
I think your best bet is a pragmatic one...
Every few years, buy a good high-capacity removable disk drive and move everything over to the new disk format.
So take the thousands of CDRs and copy them to DVD-Rs. Wait a few years, buy the next format that comes out, copy the DVD-Rs to that. Rinse, repeat, etc.
And make sure that you burn two copies.
Also note that for sitting around and doing nothing, the CDRs have a long life. Using CDRs shortens the life considerably.
I know that I've at least seen advertising materials for Matrox's low-profile PCI video cards. I couldn't find it on Matrox's website, so you might have to contact them or a reseller that works on low-profile PCI products.
Now, this is hacking cars.. ;)
Some advantages nobody's touched on yet..
1) Easier implementation of large filesystems. 2^64*512 bytes of disk space per filesystem should be good enough for quite a few increments of Moore's Law. Ditto for larger than 4 gig files.
2) 64 bit processors take better to certain implementations of NUMA. SGI's implementation of NUMA gives each processor a range of memory that is local to that processor. If you had a 64 processor NUMA cluster, you'd have 64 megs local to each processor with 32 bit processors. You could have a few gigs per processor with 64-bit addressing.
3) With 64-bit processors, it's easier to map a file to memory again, without needing to map individual chunks. Over the near term, you could map your entire disk drive to memory space.
4) There are cases (i.e. bit packing) that don't take too well to vectorized MMX/SSE/etc. processing but do take well to 64 bit registers.
5) The ability to segment your memory space without creating annoying limitations. As in, you can have the lower 8,388,608 terabytes of RAM reserved for the user and the upper 8,388,608 terabytes of RAM reserved for the kernel. As opposed to Windows 2k, which leaves 2 gigs for the user and 2 gigs for the kernel. With the possibility of 3 gigs for the user, if you are running a higher-end version.
6) The ability to cache a data structure in the RAM attached to a given machine instead of buying solid state disk drives or other such things.
Most modern architectures don't have multi-word instructions. If I remember correctly, even the 68000 processor series used single-word instructions.
The P4 doesn't always use 20 bit addressing. It can, but it doesn't need to. There are pure 32 bit memory accessor operations that let you skip the translation.
Most non-x86 modern 32 bit processors (32 bit PowerPC processors, 32 bit MIPS processors, 32 bit SPARC processors etc.) have more registers than the PC. It does net you great performance increases without the register renaming features the latest x86 CPUs have.
64 bit processors, in their own right, don't net your average user much extra power right now. But we are reaching the point where mid and high end computing needs 64 bits. It's only a matter of time before consumer PCs start to need 64-bits.
I suspect that the Comcast and Cox deal made sense because the @Home people figure they can liquidate their assets for more than $80M.
It was likely a good bet. AT&T could have blinked and paid up for more than $400M.
Ooops.
If you check Mozilla Releases, you will find releases for Free and Net BSD builds, but no OpenBSD builds.
Most things that compile for Linux will work under BSD.
So vim and emacs work, mozilla works, and whatever MP3 player you want will work.
Trust me....
SourceSafe will have problems before CVS will have problems.
So that shouldn't be too much of a concern.
Most programmers I know who actually have a CS degree or equivelent (That didn't come from Bob's Orange Smoothie and School of Computer Science) can pick up a new programming language fast enough to be useful.
But then, my definition of a mid-level programmer must be different than yours is.
I've had decent luck sending things in the origional packaging (So they can't gripe about the packaging) and sending it the first level above UPS ground.
And you have to watch your insurance carefully. Often times, they insure everything BUT electronics.
Yeah, but what does the leg up give it? SOAP is suffering from the too-many-cooks sort of problem that is taking it away from being a simple protocal.
And, since the standard isn't complete but people are implementing it, it probably won't be any more standard than old-fashioned RPC, where everybody extended it and such to build your own standard. (i.e. DCOM is technically an implementation of RPC, but not really)
What's most likely to happen is that nobody will win. MS people will use MS's purposefully broken SOAP implementation, some others will use real SOAP, and others will just use XML-RPC.
Which will eliminate many of the supposed advantages of any of the protocals.