I would have thought that the same sort of thing is probably true for Internet access - especially since sending data down a wire is just as expensive as not sending data down the same wire, once the wire has been laid.
Setting aside the possibility that they can charge you more (maybe they can't, or it would be too difficult, as you point out), you still have the "tragedy of the commons". If there's no incremental charge for bandwidth, there's no incentive not to use as much as you possbily can, leaving Napster on, running a warez ftp site, etc.
A lot of people here tend to say "I paid for it; let me do what I want", but in the all-you-can-eat plan, you're not really paying for unlimited bandwidth. You're paying for your share of the bandwidth and happen to receive unlimited bandwidth. The two ways to prevent abuse are incremental charges or just going after abusers explicitly ("Stop that! No bandwidth for you!") Are there any other good solutions for preventating bandwidth-hoarding under an all-you-can-eat plan?
Right, but what I'm trying to say here is that it doesn't matter WHY someone makes posts which are informative, funny, interesting, or what have you. It doesn't matter why people stay on topic or avoid trolling and/or flaming. What matters is that they DO do the things which, on slashdot, tend to make someone's karma go up.
Right, I agree that if that happens, that's a good thing, but I'm not sure that's what will always happen. I don't think it necessarily encourages posts which are funny, interesting, or informative but instead posts which will get moderated up. In the best case these are the same thing, but I think most people would agree that they aren't always.
Now the problem is the moderators; Acting falsely clever will cause a moderator with a lack of clue to moderate you up, thinking you are clever. So again, the broken part of moderation is the moderators, not the moderatees, IMO.
Well, I agree for the most part. I think we can blame people who try to abuse the system, too.
Maybe if "God" let us know when we were doing something positive or negative by simple incrementing or decrementing a counter, and the counter had to be over a certain positive number to get into "Heaven", then more people would do more good things.
More likely, people would try to figure the easiest way to increment the counter. That's the problem with feedback mechanisms: you're not necessarily reinforcing the desired behavior (good posting) but rather the measurable surrogate for the desired behavior (getting people to moderate you up). The same thing often happens with tests with lab animals. Positive reinforcement can work if the thing you measure is really well connected to the thing you want to measure, but quite often they're not connected. This leads to "karma whoring", where the goal is to get moderated up, not to necessarily add to the discussion. Removing karma isn't going to make this go away completely, since there's still the kick of having a high per-post score, but it might help.
Of course, perhaps Quantum computing may change some or all of this, but I am not qualified to comment on that.
I'm not either, but that won't stop me:)
Factoring a 256-bit number using Shor's algorithm for a quantum computer should take up to 769 qubits (we have, what, 5 or 7 so far?) and runs in O((lg n)^2 * lg lg n), which is O(really fast). For a 256-bit n the inner part works out to 524288, which doesn't tell you much but at least you can see it doesn't grow that fast.
I don't know what episode you're mentioning, but here's a quote by Bill Gates from my freshman-year CS textbook that has something to do with dumpsters:
No, the best way to prepare is to write programs, and to study great programs that other people have written. In my case, I went to the garbage cans at the Computer Science Center and I fished out listings of their operating system. You've got to be willing to read other people's code, then write your own, then have other people review your code.
One of my professors said she has a colleague with a big picture of Gore on his office wall. The colleague said that Al Gore stopped by his office (he does distributed computer) in 1994 and spent an hour in there asking questions. I had never heard of the Internet and didn't know the slightest thing about distributed computing at that point, so Gore's at least got some genuine interest in computers and networks.
The real problem isn't that they couldn't package Python for inclusion. The real problem is that it's questionably legal to embed Python in a GPL'd package. I'm doing some of that, but I'm not particularly worried.
CNRI, the previous employers of GvR, own the code; they were guaranteed a Python 1.6 release by Guido. BeOpen will do the 2.0 release. That's why the two are so close together. CNRI is being a bit difficult about the licensing issues (claiming that the previous Python license doesn't count as a real license, etc.)
In the best case, all of this will be worked out and people can distribute binaries of this package with Python linked in. In the worst case, none of this gets sorted out until Python 2.0 (BeOpen wants GPL compatibility) and no one can distribute said binaries with Python2.0.
You're right about the first part, and that's a reason that brick and mortar stores might be more efficient than internet stores (which have their own advantages). They ship in bulk, from distributor to the store, and the internet stores ship from the distributor (optionally to a warehouse) and then to the consumer.
But the USPS is only sort of a government agency; it is not subsidized, except perhaps in the real estate it owns (this is a really, really small part of its operating expenses). So the second part of your argument doesn't really make sense. Using the USPS is nothing at all like paying taxes.
The shipping costs are far greater than sales tax, so no, this is not true.
This gives online stores a disadvantage in pricing--they're forced to have HIGHER prices than brick-and-mortar stores, because they have to pay both taxes and shipping expenses.
Shipping is a real cost. Taxes are artificial. This is a very important point that not a lot of people seem to take into account. The cost of a good should be based upon supply in demand. Supply is in turn determined by how much effort and how many materials go into the product (among other things). It costs money to ship you a product, so that product should cost more. Indeed, you pay for the cost of shipping the product to the individual stores (from the distributors) when you buy from a brick and mortar store.
Think of it the other way around. Assume there are no taxes on anything. Would it be fair to the independent bookstores for the government to start charging just them and leave the internet stores alone? Would you say "Internet businesses have to pay shipping; we should charge local businesses so it will all be fair."?
The patent office is severely underfunded and was just forced to lay off something like 20% of their already meager staff. Of course, most of us would say that they should err on the side of caution when they don't have enough time to review patents, but the patent lawyers and patent seekers would probably prefer them to err towards leniency and those are the people who talk to the patent office more.
(1) Many of the frequencies of interest are not used by commercial broadcasting equipment (if I remember correctly, things like the spin flip transition of ground state atomic hydrogen, and the prominent HNO lines...but I could be remembering incorrectly here)
They're not that close at all, but as luck would have it, the frequency of spark plugs is in some way related to the frequency of interstellar hydrogen (not too close to a harmonic, I guess, because otherwise radio astronomy wouldn't work at all). As a result, only deisel engines are allowed inside the NRAO. And bikes of course; it's a really pretty place and can be better appreciated on a bike than in a diesel van.
Well, I wouldn't expect Linux to incorporate it, if only because I doubt the IBM license is compatible with the GPL. What I'm wondering about is arla, the existing free AFS implementation (I can't find what the license is, though I think it's GPL). It's amazing how many different kernels they support with that thing (a reason it won't/doesn't need to be incorporated into Linux). The server support is only experimental, but they client has been a blessing for those of us who wanted to use AFS but didn't want to wait for Transarc to release a new module to upgrade our kernels.
Anybody else notice that this might be the single biggest advantage of Microsoft's.NET program? Most of their program sounds vague, but the platform independent binaries actually sound like a good idea. They'll distribute the binaries in a java-bytecod-esque form and they'll be compiled native to the machine on installation (oo, that's going to make installing big things fun).
Of course, you can do basically the same thing with open source (I can't see how you wouldn't have similar portability problems), but this would be the solution for closed source stuff. So your new database or game or whatever will compile to your itanium/sledgehammer/z80/whatever. Well, I guess they can hope.
Well, it'll work with Linux right now, but you can't do anything 64-bit at all with it without a port. Sledgehammer has 3 modes. The first is legacy mode. You use that to run a 32-bit operating system and 32-bit applications. Linux will run in this way out-of-the-box. You can't run 64-bit applications in this mode and have to reboot to change. Then they have "long mode " which is split into two other modes, whose names I don't remember. Long mode runs a 64-bit operating system. The first of the long modes (compatibility mode or something) runs 32-bit applications on the 64-bit operating system. The second mode runs 64-bit applications. The two long modes can be changed by a context switch, so you can be running 64-bit and 32-bit applications on the same 64-bit OS.
My understanding of it is that you start out with a quantum register (made up of quantum bits) in all sorts of different states, do operations to them until only one possible state is left. Then and only then are you allowed to look at the answer (you can tell what an electron's spin is).
Re:The impossibility of Online Voting
on
Online Voting?
·
· Score: 2
There are some pretty good protocols for secure online voting outlined in Applied Cryptography. Here are the goals that Schneier says are desirable for online voting.
Only authorized voters can vote.
No one can vote more than once.
No one can determine for whom anyone else voted.
No one can duplicate anyone else's vote.
No one can change anyone else's vote without being discovered.
Every voter can make sure that his vote has been taken into account in the final tabulation.
And optionally:
Everyone knows who voted and who didn't (this one's optional)
A voter can change his mind.
If a voter finds out that his vote is miscounted, he can identiy and correct the problem without jeopardizing the secrecy of his ballot.
He describes a protocol which fulfulls all but #7. It's pretty complicated, but most of that falls on the implentation. It also doesn't seem to be susceptible to a replay attack, if implemented properly. The burden of proof on the individual would probably be about as valid and difficult as it is now. Instead of a voter registration card and an SSN as an ID, you'd have a public key, or something like that.
You can write an application however you want for MacOS, but they still have a UI standard that most developers try to follow and the users are probably better off for it (though they hate non-comforming apps as a result). I don't think anyone's suggesting you FORCE people to adhere to these things, but as it stands now UNIX doesn't even have these kinds of standards. How would they hurt?
Re:Network (infrastructure) Appliance?
on
Linux In A Box
·
· Score: 1
I have a 486 acting that's serving DHCP and masquerading for my 3-computer home network. It does just fine, and I don't the load's ever been near what would kill it (it does a take a while for sshd to connect me though). I used to have it running squid, bind, and apache but I figured those weren't worth the load (and keeping updated). It's really easy.
The webpage the previous poster pointed to should give you all the information you need and more. Start with the newbie-oriented howtos and help files. You'd just need a NIC for the 486. It doesn't even need a hard drive (though don't try running Squid or Samba off a floppy!) You don't need to edit any config files directly using LRP (though you can), since everything is wrapped in a console-based configuration utility. I'm a big fan of the LRP distribution.
I think a calculator is defined more by the UI than the functions it provides. I don't care how feature-complete a graphical (or text-based?) calculator program may be, I'll still be faster and more productive using my old HP (assuming I don't need online help, which is a pretty big assumption). If they're making the UI much more generic to allow MP3 playing, then that's bad, but if it still looks like a calculator, it's still a calculator to me.
But more importantly than just setting the thing up is if it will play nicely with NAT (aka IPMasq and that Win98 Internet Connection Sharing thing). I suppose it'll use directplay which is IIRC UDP, which (also IIRC) doens't work through NAT without explicit port forwarding. Among other things this means that you'd have to change your firewall if you wanted to switch between playing a DirectPlay game on your computer and one on your X-Box (or you could switch the IPs, which would be just as much of a hassle).
I'd like to get a networked game console some day (I have an N64). I just hope that whenever they come out with these things they'll work right with NAT. I'd think that with proliferation of all the consumer-level out-of-box router-like things which accomplish what my 486 is accomplishing right now, there would be pressure for this to work. I guess we'll see; I don't have my hopes up.
Well, if your mailing list was archived publically, then your address might have been picked up by spam-spiders or whatever.
I've seen mailing list programs which archive the messages changing myname@example.com to "myname at example dot com", which should be a decent deterant. I've also seen (in HTMLGen for Python, though I'm sure it's used elsewhere) replacing random letters from the email address with the HTML escape sequence, so that it looks perfectly normal to a person using a web browser but as text it's not a valid email (which will work until spider programs replace the escape sequences, which perhaps they do already).
I wrote a newsgroup-email-sucker-thing one time (for educational purposes only!); it was I think 16 lines of Python, not coded very well, a single thread of execution, and it still nabbed around 6000 addresses per hour (extrapolated; I didn't run it that long).
Slashdot Mirror
I would have thought that the same sort of thing is probably true for Internet access - especially since sending data down a wire is just as expensive as not sending data down the same wire, once the wire has been laid.
Setting aside the possibility that they can charge you more (maybe they can't, or it would be too difficult, as you point out), you still have the "tragedy of the commons". If there's no incremental charge for bandwidth, there's no incentive not to use as much as you possbily can, leaving Napster on, running a warez ftp site, etc.
A lot of people here tend to say "I paid for it; let me do what I want", but in the all-you-can-eat plan, you're not really paying for unlimited bandwidth. You're paying for your share of the bandwidth and happen to receive unlimited bandwidth. The two ways to prevent abuse are incremental charges or just going after abusers explicitly ("Stop that! No bandwidth for you!") Are there any other good solutions for preventating bandwidth-hoarding under an all-you-can-eat plan?
Right, but what I'm trying to say here is that it doesn't matter WHY someone makes posts which are informative, funny, interesting, or what have you. It doesn't matter why people stay on topic or avoid trolling and/or flaming. What matters is that they DO do the things which, on slashdot, tend to make someone's karma go up.
Right, I agree that if that happens, that's a good thing, but I'm not sure that's what will always happen. I don't think it necessarily encourages posts which are funny, interesting, or informative but instead posts which will get moderated up. In the best case these are the same thing, but I think most people would agree that they aren't always.
Now the problem is the moderators; Acting falsely clever will cause a moderator with a lack of clue to moderate you up, thinking you are clever. So again, the broken part of moderation is the moderators, not the moderatees, IMO.
Well, I agree for the most part. I think we can blame people who try to abuse the system, too.
Maybe if "God" let us know when we were doing something positive or negative by simple incrementing or decrementing a counter, and the counter had to be over a certain positive number to get into "Heaven", then more people would do more good things.
More likely, people would try to figure the easiest way to increment the counter. That's the problem with feedback mechanisms: you're not necessarily reinforcing the desired behavior (good posting) but rather the measurable surrogate for the desired behavior (getting people to moderate you up). The same thing often happens with tests with lab animals. Positive reinforcement can work if the thing you measure is really well connected to the thing you want to measure, but quite often they're not connected. This leads to "karma whoring", where the goal is to get moderated up, not to necessarily add to the discussion. Removing karma isn't going to make this go away completely, since there's still the kick of having a high per-post score, but it might help.
Oops, my bad. A danger of reading a +3 and only getting part of the conversation :(
Of course, perhaps Quantum computing may change some or all of this, but I am not qualified to comment on that.
:)
I'm not either, but that won't stop me
Factoring a 256-bit number using Shor's algorithm for a quantum computer should take up to 769 qubits (we have, what, 5 or 7 so far?) and runs in O((lg n)^2 * lg lg n), which is O(really fast). For a 256-bit n the inner part works out to 524288, which doesn't tell you much but at least you can see it doesn't grow that fast.
I don't know what episode you're mentioning, but here's a quote by Bill Gates from my freshman-year CS textbook that has something to do with dumpsters:
No, the best way to prepare is to write programs, and to study great programs that other people have written. In my case, I went to the garbage cans at the Computer Science Center and I fished out listings of their operating system. You've got to be willing to read other people's code, then write your own, then have other people review your code.
One of my professors said she has a colleague with a big picture of Gore on his office wall. The colleague said that Al Gore stopped by his office (he does distributed computer) in 1994 and spent an hour in there asking questions. I had never heard of the Internet and didn't know the slightest thing about distributed computing at that point, so Gore's at least got some genuine interest in computers and networks.
Darn, Slashdot ate my less-than sign between "Python" and "2.0"
The real problem isn't that they couldn't package Python for inclusion. The real problem is that it's questionably legal to embed Python in a GPL'd package. I'm doing some of that, but I'm not particularly worried.
CNRI, the previous employers of GvR, own the code; they were guaranteed a Python 1.6 release by Guido. BeOpen will do the 2.0 release. That's why the two are so close together. CNRI is being a bit difficult about the licensing issues (claiming that the previous Python license doesn't count as a real license, etc.)
In the best case, all of this will be worked out and people can distribute binaries of this package with Python linked in. In the worst case, none of this gets sorted out until Python 2.0 (BeOpen wants GPL compatibility) and no one can distribute said binaries with Python2.0.
You're right about the first part, and that's a reason that brick and mortar stores might be more efficient than internet stores (which have their own advantages). They ship in bulk, from distributor to the store, and the internet stores ship from the distributor (optionally to a warehouse) and then to the consumer.
But the USPS is only sort of a government agency; it is not subsidized, except perhaps in the real estate it owns (this is a really, really small part of its operating expenses). So the second part of your argument doesn't really make sense. Using the USPS is nothing at all like paying taxes.
The shipping costs are far greater than sales tax, so no, this is not true.
This gives online stores a disadvantage in pricing--they're forced to have HIGHER prices than brick-and-mortar stores, because they have to pay both taxes and shipping expenses.
Shipping is a real cost. Taxes are artificial. This is a very important point that not a lot of people seem to take into account. The cost of a good should be based upon supply in demand. Supply is in turn determined by how much effort and how many materials go into the product (among other things). It costs money to ship you a product, so that product should cost more. Indeed, you pay for the cost of shipping the product to the individual stores (from the distributors) when you buy from a brick and mortar store.
Think of it the other way around. Assume there are no taxes on anything. Would it be fair to the independent bookstores for the government to start charging just them and leave the internet stores alone? Would you say "Internet businesses have to pay shipping; we should charge local businesses so it will all be fair."?
The patent office is severely underfunded and was just forced to lay off something like 20% of their already meager staff. Of course, most of us would say that they should err on the side of caution when they don't have enough time to review patents, but the patent lawyers and patent seekers would probably prefer them to err towards leniency and those are the people who talk to the patent office more.
Just FYI, you've been able to do the same thing in GNOME with libglade for quite a while.
(1) Many of the frequencies of interest are not used by commercial broadcasting equipment (if I remember correctly, things like the spin flip transition of ground state atomic hydrogen, and the prominent HNO lines...but I could be remembering incorrectly here)
They're not that close at all, but as luck would have it, the frequency of spark plugs is in some way related to the frequency of interstellar hydrogen (not too close to a harmonic, I guess, because otherwise radio astronomy wouldn't work at all). As a result, only deisel engines are allowed inside the NRAO. And bikes of course; it's a really pretty place and can be better appreciated on a bike than in a diesel van.
Well, I wouldn't expect Linux to incorporate it, if only because I doubt the IBM license is compatible with the GPL. What I'm wondering about is arla, the existing free AFS implementation (I can't find what the license is, though I think it's GPL). It's amazing how many different kernels they support with that thing (a reason it won't/doesn't need to be incorporated into Linux). The server support is only experimental, but they client has been a blessing for those of us who wanted to use AFS but didn't want to wait for Transarc to release a new module to upgrade our kernels.
Anybody else notice that this might be the single biggest advantage of Microsoft's .NET program? Most of their program sounds vague, but the platform independent binaries actually sound like a good idea. They'll distribute the binaries in a java-bytecod-esque form and they'll be compiled native to the machine on installation (oo, that's going to make installing big things fun).
Of course, you can do basically the same thing with open source (I can't see how you wouldn't have similar portability problems), but this would be the solution for closed source stuff. So your new database or game or whatever will compile to your itanium/sledgehammer/z80/whatever. Well, I guess they can hope.
Well, it'll work with Linux right now, but you can't do anything 64-bit at all with it without a port. Sledgehammer has 3 modes. The first is legacy mode. You use that to run a 32-bit operating system and 32-bit applications. Linux will run in this way out-of-the-box. You can't run 64-bit applications in this mode and have to reboot to change. Then they have "long mode " which is split into two other modes, whose names I don't remember. Long mode runs a 64-bit operating system. The first of the long modes (compatibility mode or something) runs 32-bit applications on the 64-bit operating system. The second mode runs 64-bit applications. The two long modes can be changed by a context switch, so you can be running 64-bit and 32-bit applications on the same 64-bit OS.
My understanding of it is that you start out with a quantum register (made up of quantum bits) in all sorts of different states, do operations to them until only one possible state is left. Then and only then are you allowed to look at the answer (you can tell what an electron's spin is).
And optionally:
He describes a protocol which fulfulls all but #7. It's pretty complicated, but most of that falls on the implentation. It also doesn't seem to be susceptible to a replay attack, if implemented properly. The burden of proof on the individual would probably be about as valid and difficult as it is now. Instead of a voter registration card and an SSN as an ID, you'd have a public key, or something like that.
You can write an application however you want for MacOS, but they still have a UI standard that most developers try to follow and the users are probably better off for it (though they hate non-comforming apps as a result). I don't think anyone's suggesting you FORCE people to adhere to these things, but as it stands now UNIX doesn't even have these kinds of standards. How would they hurt?
I have a 486 acting that's serving DHCP and masquerading for my 3-computer home network. It does just fine, and I don't the load's ever been near what would kill it (it does a take a while for sshd to connect me though). I used to have it running squid, bind, and apache but I figured those weren't worth the load (and keeping updated). It's really easy.
The webpage the previous poster pointed to should give you all the information you need and more. Start with the newbie-oriented howtos and help files. You'd just need a NIC for the 486. It doesn't even need a hard drive (though don't try running Squid or Samba off a floppy!) You don't need to edit any config files directly using LRP (though you can), since everything is wrapped in a console-based configuration utility. I'm a big fan of the LRP distribution.
I think a calculator is defined more by the UI than the functions it provides. I don't care how feature-complete a graphical (or text-based?) calculator program may be, I'll still be faster and more productive using my old HP (assuming I don't need online help, which is a pretty big assumption). If they're making the UI much more generic to allow MP3 playing, then that's bad, but if it still looks like a calculator, it's still a calculator to me.
But more importantly than just setting the thing up is if it will play nicely with NAT (aka IPMasq and that Win98 Internet Connection Sharing thing). I suppose it'll use directplay which is IIRC UDP, which (also IIRC) doens't work through NAT without explicit port forwarding. Among other things this means that you'd have to change your firewall if you wanted to switch between playing a DirectPlay game on your computer and one on your X-Box (or you could switch the IPs, which would be just as much of a hassle).
I'd like to get a networked game console some day (I have an N64). I just hope that whenever they come out with these things they'll work right with NAT. I'd think that with proliferation of all the consumer-level out-of-box router-like things which accomplish what my 486 is accomplishing right now, there would be pressure for this to work. I guess we'll see; I don't have my hopes up.
Well, if your mailing list was archived publically, then your address might have been picked up by spam-spiders or whatever.
I've seen mailing list programs which archive the messages changing myname@example.com to "myname at example dot com", which should be a decent deterant. I've also seen (in HTMLGen for Python, though I'm sure it's used elsewhere) replacing random letters from the email address with the HTML escape sequence, so that it looks perfectly normal to a person using a web browser but as text it's not a valid email (which will work until spider programs replace the escape sequences, which perhaps they do already).
I wrote a newsgroup-email-sucker-thing one time (for educational purposes only!); it was I think 16 lines of Python, not coded very well, a single thread of execution, and it still nabbed around 6000 addresses per hour (extrapolated; I didn't run it that long).
No, I said you can't expect an OS to divide a single thread among separate procesors.