That paper provides a weak argument against PoW tokens. PoW tokens change the numbers of emails sent out by orders of magnitude. Certainly, a spammer can use cracked boxes to generate hashes, but that also reduces the utility of cracked boxes by several orders of magnitude.
PoW tokens are most effective when used in combination with other means of spam protection/prevention. Consider grandma, from the paper you linked, with her 100 Mhz 486. Almost all of the people she will be emailing are people that she already knows or is trying to establish a business relationship with (shopping at Amazon.com) Once identity has been established, it's no longer neccessary to use the tokens. Instead, whitelisting via digital signatures could be employed. This technique would also allow solicited bulk mail, such as mailing lists, to continue to be computationally cheap.
Moore's law does cause one to pause when thinking about PoW tokens, untill you realize that the size of the tokens is variable. Since they're merely there to denote that the sender has spent some of their own resources to make (initial) contact with you, then the size of the token could be used to show you how much that person cared about getting in touch with you. The larger the token, the greater the confidence in the message. In other words, the tokens grow with computing power.
The bottom line is that PoW tokens, while not a silver bullet, could add some barbed wire to the battle-field of spam. Since I've had the same email address for 14 years now, I would be happy with a 5% reduction in spam.
p.s. I didn't mean to minimize the importance of security here. In addition to locking down the email protocols, we also need to lock down the average user's computer. The fact that spammers have 10,000 0wned boxes is significant to the spam problem, but something to be considered elsewhere.
I'm amazed that I haven't seen more about Proof of work tokens for spam-fighting.
Proof of work tokens are hashes (like md5's) that take a relatively long time to compute and are very quick to validate. For most purposes, adding a few seconds to the delivery of email is unnoticable. For spammers, however, it greatly decreases the number of emails that can be sent out within a period of time.
Even though this does not completely eliminate the problem, it can significantly reduce the amount of time spent sifting through spam. Used in combination with public-key cryptography, it could even allow for mass-mailings from known users. (For instance, the Red Hat mailing list.)
The current problem with spam is a result of the fact that it takes almost no money to send spam. Increasing the amount of time spammers need to use in order to send out email is the only way to make a dent.
Most more pedestrain difficulties--bipolar, depression, ADHD [...] are best treated with cognitive behavioral therapy. Change the behavior and you change the brain chemistry. Psychotropics are terribly harmful and do nothing to address the underlying behavior that's causing the difficulties.
One might be able to argue that these conditions are overdiagnosed, but to say that these conditions are not greatly benifited by the use of medications is a sad, sad joke.
I am bipolar. My parents come from midwestern families that believe anything can be cured through force of will. I spent five years wondering why my life consistently crumbled around me and why I went through periods where I absolutely could not manage money. Eventually, I had a mania so intense that I was diagnosed schizophrenic. That was the first time that I had thought "Maybe this isn't all my fault and maybe it's not just a character flaw." When I finally did find the right medications, the world completely changed for me.
In a similar vein, I was there when my best friend was diagnosed ADD at 22. The contrast between before and after he started medications was astounding. No-one had a doubt that he had made the right decision by going on medications.
This is not to say that cognitive therapy is not part of the process. It has been helping me manage my illness, one I will have all my life, with great effect. However, the fact remains that no amount of management would be effective if I were not on medications to decrease the depth, duration and frequency of my mood swings.
There are currently only three drugs approved for bipolar disorder, yet there are hundreds which might be prescribed for it. If the few that are generally accepted as "first line" treatments don't work for a person, that person is almost definitely in for a long and tortuous trial and error process for finding the right medications.
I was put on risperdal for the few months when my bipolar had been misdiagnosed as an unknown flavor of schizophrenia. In the parent's poster's case, the dopamine receptors were brought back to a normal state by blocking some of the over-activity. I can speak as a person whose dopamine receptors were blocked beyond what they should be.
I was definitely much more productive once my mania had subsided. The problem was, I didn't derive much pleasure from anything. Sex became as much a chore as washing the dishes. My favorite television shows could no longer hold my attention. I just started doing things that I thought I should be doing in order to occupy time. Life felt distant and flat. It was productive, but quite meaningless.
After a couple of months, I headed into a depression. It was the worst that I had ever had. I saw no point whatsoever in continuing my existence and no possible ground that could be gained from any course of action. Eventually, I managed to drive to my best friend's house and say "Unless you can see something worth living for in me, and get me help finding it, I will kill myself." The only thing that saved me was my willingness to trust my friends to do what was right.
When I finally found the right doc, obtained the proper diagnosis and was put on the right drugs (Lamictal and Effexor) it was a feeling not unlike watching the most beautiful sunrise ever. Untill then, I did not realize how awful of an experience that had been. Simply coming off the risperdal while starting my first (totally ineffective) drug regimen opened a new world.
Rent the movie Equilibrium. Not only is it a bad-ass action flick, it describes how I felt during that time. Now, one might think that this was a result of my bipolar, but I've never felt that way ever before in my life. I get depressed, but never before in that way. If I ever feel like that again, killing myself would definitely be an option.
Now, who wants to help test this article's process in humans?
Because what you're talking about is a Jim Crow law, and it's illegal. Even if it's just a tax-break, it's a tax break at the expense of knowledge...kind of like a tax break if you know a secret code. There's too much potential for abuse.
*sigh* I knew I should have addressed the Jim Crow clause in my original response. I assume that you're refferring to the literacy test and poll tax laws that were used to disenfranchise many people of color.
This would be no more of a Jim Crow Law than obscure itemized deductions. You can currently save a bundle on taxes if you know the secret code known as "Tax Law." The well-educated have a distinct advantage in this regard and many are willing to pay money to have an expert help them. This would only be a Jim Crow law if it disenfranchised someone, or created a gross inequity that settled along racial or economic lines.
Abuse only becomes an issue if you are incredibly uncreative in the setup of the system. First off, don't make it a huge tax break. Fifty bucks will do. That's a fairly large motivator for those with lower incomes. Secondly, earmark the fifty bucks that would not be paid out to someone who didn't vote/pass the test for civic education classes. Since no money was previously going to any efforts like this, it's not treading on anyone's turf. The classes might be publicly run, or they might be privately run, or both. That's a detail to be I feel is a bit beyond the scope of a conversation here. (In other words, let's not get lost in the details before we've mapped out some of the big questions)
Then, make the questions 100% factual, but not based on rote memory. 100 questions, multiple choice. Have all possible questions and the correct answers freely available in every post office, police station, etc.
You still have the question of those who are illiterate. I think that's a toughie. I'll have to think on that one.
Besides, knowing who your congressmen is doesn't mean you get less out of the system.
First, that statement attempts to hold one aspect of what I said as being representative of the whole. So, throw out that requirement. In it's place, put an extra question about the electoral college. Second, knowing the name of your congresscritter allows you to recognize their name in the headlines and it allows you to know who to vote out even if you know nothing of the competition.
Here in the US, if a guy is pretty much guaranteed to win your state, or if you don't care about either candidate, there's not really much of a point to voting.
[...] Obviously, if you live in a swing state, your ass should be in those polls. But c'mon...Bush is as likely to win New York as he is to apologize for his nationalistic foreign policies post 9-11.
While living in Tallahassee during the last election, I met quite a few people who wished they hadn't believed the polls and had instead gone out and voted. As a segue into the next section, is making someone grumble about how assinine it is that they have to vote when the stupid system effectively disenfranchises them such a horrible thing?
You want to see record numbers in the polls? Adopt Australia's vote-for-who-you-want policy. You'll start to see the ignored third parties -- right to life, independence, green party, libertarians, marijuana reform, socialists -- get more power real quick. I'm saying Libertarian congressmen by 2010 if we changed the voting law today. And Socialist senators -- probably from Vermont -- as early as 2014.
Let me know if you ever want help handing out fliers on this issue. I whole-heartedly agree, and I think that education is part of the solution. Aside from fiasco's like the last election, people can't understand how messed up our election system is untill the understand how it works. I would be happier if people simply understood that changes must happen at a st
'you can't shop around for better rates on a tax or reduce your taxation by taking a ten hour "defensive citizenship class"'
My gut reactions says that a tax reduction for those who could pass a civics test is not a bad idea. It's amazing how many people don't know how many senators there are, or the even the names of their local congress-critter. I think that if the majority of people knew how a bill gets passed or what rights and responsibilities are accorded to each branch of government, that the political process would be a bit saner. Hell, if an immigrant seeking nationalization has to pass a test like this, why shouldn't joe-sixpack?
Heck, why not double it up and provide a break for those who vote? I know that Australia imposes a fine on those who don't vote. We've all heard the slogan that it's not only your right to vote, but your duty. There's nothing that says you have to vote for anything, just show up to the voting booth and turn in a blank ballot, collect your receipt and move on with life.
There may be some flaws in this idea, but my gut says that it would improve the state of affairs. None of this would bar someone from taking part in the process, but instead would make it slightly more lucrative to take part.
While we're at it... Why not give welfare recipients a couple months to study and pass such a similar test or have their benifits cut? (but not eliminated.)
The thing I love about Biafra is that whether you agree with his ideas or not, you have to admit that he asks some interesting and important questions.
Many of the people on slashdot consider themselves to be activists of some degree or another. I'm right along with the majority of them who loathe microsoft, its business practices, its software and its API's.
Still, I can't quite summon myself to the kind of venom that I've seen on/. today. See, I don't think I would have the right to continue complaining if I did not give honest and constructive criticism when asked for it. Yes, this information will probably be used to ill conceived ends, but I have to say that I tried.
In Buddhism, Karma is not seen as a physical or metaphysical force that forces you to pay back your evil deeds, but rather as a psychological principle.
I could go on for a while about this, but I'll try to keep it short.
Ideally, once an action has happened, it is gone. Our mind, however, likes to hold on to what was and will not let go. Since the past is already gone, it is not a good place for the mind to dwell. Your karma is the part of you that holds onto the past and measure the present in terms of the past. If you measure the present in terms of the past, then you see more of what was than what is. A buddhist strives to see things as they are in the present (this is what meditation teaches) No buddhist claims that they will ever be free of karma, of their attachment to the past, but if they could be, they would be "enlightened."
A quick example: Let's say that you spent the past two years building your own house by hand. On the night that you move into the house, it burns down. If you hold on to the past and your memories of what was, then you will suffer at the thought of loss and the "wasted" time you spent building the house. If you see things as they are, then you will be happy that you are alive and start planning for a new place to live and perhaps rebuilding the house.
A rational number is any number that can be expressed as an integer divided by an integer. You're right that Pi is not rational, which is to say that it cannot be represented as an integer divided by another integer.
You are *completely* wrong in your assumption that an irrational number cannot have a pattern to it. Consider, for example, the number 0.10011000111000011110000011111...... The pattern in this number is very obvious. It cannot, however be represented as an integer divided by another integer and is thus irrational.
You say that like they're being paid enough to deal with BS that was proposed.
If enough people on a regular enough basis were to do something like this, sure, it would work. On the other hand, going to the manager with one month's worth of grocery receipts(be sure to have them totaled,) and telling him politely that you, your friends and neighbors are not going to be shopping at his store anymore because of the policy is a much more productive way of dealing with the situation. Trust me, if you give the manager hard numbers, like a stack of receipts, it will make them sweat more than loading up a cart, throwing a tantrum in line, then making employees run all over the store putting up items will. If you do that, you're just a jackass, not an activist.
This won't piss off the store, it'll only piss off the powerless employees working for minimum wage. The manager won't be putting those items away, it'll be the bagboy. What good will that do?
It's sad to watch people complain to clerks like they can do something about a policy such as this. Having worked enough low end jobs in my life, I can tell you that the clerks can gripe and grumble, but they're not the ones who will get things changed. Find the Manager! Talk to his boss! Make *them* sweat. Don't piss on the guy who's working his ass of trying to get through college. Don't shit on the mother of three who's there just to make ends meet. Find someone who can actually change the policy, and then piss them off. Leave the little guys alone, they probably dislike this as much as you.
In order to do some things, PGP must come up with large prime numbers. RSA key generation is the only thing I'm sure of, but there may be others. Before the primality test, it took much, much longer to make sure a particular number was prime.
It shouldn't do anything to the security of PGP, since the encryption is still based on factoring large numbers, a process which isn't (significantly) sped up by the primality test.
I remember a documentary from the gulf war that showed a huge spool of explosive cable shot from the back of the truck. When the cable exploded, it took the mines with it. Other than having to carry the cable with you, I wonder why that isn't sufficient.
Personally, I think that signing individual users would be better than signing individual files. If you sign individual files, then someone can flood the system with bogus signings from a large number of random usernames, but if I instead assign trust only to people, it's much harder for a bogus client to get deeply into the web. There's also the issue of keeping track of trust. If I sign individual files, then I have to keep track of all those signatures. Since there are a great many fewer people than files, the amount of persistant information would be much smaller when only signing people. If you rank trust as a proportion, say 0% to 100%, then leaps through the trust web can be easy to calculate. I trust Toni 80%. Toni trusts Jenny 50%, thus I trust Jenny 40%. Let's say that Jenny and Toni are on the network when I connect, and I have don't even know of Jenny's existance yet. When I connect, I ask Toni to pass me his trust information. He passes me a file containing his signature for Jenny along with her current IP address. I calculate trust, mark it in my trust file, and then send Jenny my signed 40% trust. She can choose to discard this signature if she thinks it's too low (trust levels below 10%, for instance) and thus keep the really pointless entries out of the system. If, later, I decide that Jenny is actually a pretty good person, and bump her trust up to 75%, I would send her that signature, and she would keep the higher rating. This would keep the web large and open, but make it hard to do anything without a reputation or an introduction. The people who give out trust willy-nilly or who pass bad files will quickly be marginalized, but the people who build a good reputation will gain a larger base of files to choose from and contribute more to the system. The idea needs some work, but that's the basic algorithm as I see it. Oh, and there should be some way to tell people that they have a bum file and allow them to correct the situation before you automatically distrust them.
The biggest problem would be finding the IP address of other people you already trust. Any system that is specifically designed to keep track of users would be open to legal and DOS attacks. So, why not use something such as IRC. I tell all my friends to meet me in #SecureIndustrialMusic or some other such channel, and we could get each other's ip adress and become part of the network. Better yet, use some system that noone would dare to try and close down, like AIM or ICQ or YIM...or any combination of methods, but only for finding the people you already trust. Heck, if you did this in python, it would be portable and most of the routines you need are already contained in only two libraries: twisted has IRC, AIM and generic net libraries and PyCrypto has public key libraries. (that is, if my sleep deprived memory serves me correctly...I could be wrong about those libraries) Then slap a nice little tcl/tk front end onto it and you'd have a good system Moreover, this could help pare down each user's database size. If I hadn't seen a person in, say, six months, then I could just stop trusting them altogether.
In short, I think it's doable and I like the fact that it lends itself well to small groups slowly building reputations and relationships with each other. It is a system that self-corrects and builds up value over time. It rewards reliability, and makes poor-sportsmanship difficult to maintain for any length of time.
Over the past couple of months, I've been thinking that the solution might be a web-of-trust system similar to pgp key signings. It doesn't seem like such a thing would be too hard to impliment with actual key signings, perhaps even with gpg and the gnutella codebase. This would certainly reduce the size of the network of p2p clients, but I think most people tend to listen to music that's owned by someone within three or four degrees of seperation from themselves. Personally, I only use p2p for finding bands that have have been reccomended by friends, so it would almost certainly be within a couple of hops of trust from myself.
The only problem I can see with the moderation system that you're suggesting is that there would have to be a central authority for mod points. In the current political and legal climate, that's a direct weakness. You could, conceivably, combine the two systems. So, I could rate everyone that I've downloaded from based on Quality of Service and that would enter a special file, which could be picked up by each client that has trust in me. The client would then weight the entries based on how much they trust me. For instance, if they only had 50% trust in me, then my ratings could be cut in half. They could then decide on a threshold, below which they won't do business with a client. Someone could be allowed to enter into the network.
This system has a lot of possibilites. It would keep out unwanted parties, but also allow people to come in at a low level of trust and build from that. If you made it a generic fileswapper with searchable metadata (such as gif comments and id3 tags) then also allowed ssl transfers, it would be almost impossible to track.
Sorry if this is all a bit muddled and choppy. I've been up for more than 36 hours. Let me know if this sounds at all reasonable.
"This again proves, how much better software can be, if you remove management, lawyers, sales department etc. and make good programmers work together without short-term profit in mind."(emphasis mine)
When I read that scentence, I got the image of a darkly lit slave-pit filled with coders where the corporate entities you listed had been replaced by two or three dark-overlords whipping their slaves into submission. Why don't replace the word "make" with "allow." It is a much more pleasant term.
Other than that, I wholeheartedly agree with sentiment you expressed.
I get real sick of this line of reasoning. It's put out there everytime someone asks for information on a subject for a school project, and it's nothing but a troll. (yeah, I'm biting)
If this person relies solely on slashdot to do his homework, then yes, he is dumb. That's because slashdot is a gamble. However, slashdot is a very good place to get in touch with people who keep up with computer related. What if someone points him to the one resource he couldn't find on his own that can turn a mediocre paper into a stellar one? He would be dumb *not* to suppliment his normal research with that resource.
If you don't want to help him, then don't. However you waste your time if you trot out this dead horse every time and you waste your mod points if you mod it up. --
I only post to slashdot when I'm sleep deprived.
Slashdot Mirror
(btw... the link is broken: http://www.taugh.com/epostage.pdf)
That paper provides a weak argument against PoW tokens. PoW tokens change the numbers of emails sent out by orders of magnitude. Certainly, a spammer can use cracked boxes to generate hashes, but that also reduces the utility of cracked boxes by several orders of magnitude.
PoW tokens are most effective when used in combination with other means of spam protection/prevention. Consider grandma, from the paper you linked, with her 100 Mhz 486. Almost all of the people she will be emailing are people that she already knows or is trying to establish a business relationship with (shopping at Amazon.com) Once identity has been established, it's no longer neccessary to use the tokens. Instead, whitelisting via digital signatures could be employed. This technique would also allow solicited bulk mail, such as mailing lists, to continue to be computationally cheap.
Moore's law does cause one to pause when thinking about PoW tokens, untill you realize that the size of the tokens is variable. Since they're merely there to denote that the sender has spent some of their own resources to make (initial) contact with you, then the size of the token could be used to show you how much that person cared about getting in touch with you. The larger the token, the greater the confidence in the message. In other words, the tokens grow with computing power.
The bottom line is that PoW tokens, while not a silver bullet, could add some barbed wire to the battle-field of spam. Since I've had the same email address for 14 years now, I would be happy with a 5% reduction in spam.
p.s. I didn't mean to minimize the importance of security here. In addition to locking down the email protocols, we also need to lock down the average user's computer. The fact that spammers have 10,000 0wned boxes is significant to the spam problem, but something to be considered elsewhere.
I'm amazed that I haven't seen more about Proof of work tokens for spam-fighting.
Proof of work tokens are hashes (like md5's) that take a relatively long time to compute and are very quick to validate. For most purposes, adding a few seconds to the delivery of email is unnoticable. For spammers, however, it greatly decreases the number of emails that can be sent out within a period of time.
Even though this does not completely eliminate the problem, it can significantly reduce the amount of time spent sifting through spam. Used in combination with public-key cryptography, it could even allow for mass-mailings from known users. (For instance, the Red Hat mailing list.)
The current problem with spam is a result of the fact that it takes almost no money to send spam. Increasing the amount of time spammers need to use in order to send out email is the only way to make a dent.
Links:
HashCash.org
Reusable Proofs Of Work
Currently down, but look at the google cache
How would this effect valid email lists, like the gentoo newsletter, which people actually want to receive?
One might be able to argue that these conditions are overdiagnosed, but to say that these conditions are not greatly benifited by the use of medications is a sad, sad joke.
I am bipolar. My parents come from midwestern families that believe anything can be cured through force of will. I spent five years wondering why my life consistently crumbled around me and why I went through periods where I absolutely could not manage money. Eventually, I had a mania so intense that I was diagnosed schizophrenic. That was the first time that I had thought "Maybe this isn't all my fault and maybe it's not just a character flaw." When I finally did find the right medications, the world completely changed for me.
In a similar vein, I was there when my best friend was diagnosed ADD at 22. The contrast between before and after he started medications was astounding. No-one had a doubt that he had made the right decision by going on medications.
This is not to say that cognitive therapy is not part of the process. It has been helping me manage my illness, one I will have all my life, with great effect. However, the fact remains that no amount of management would be effective if I were not on medications to decrease the depth, duration and frequency of my mood swings.
Yep.
There are currently only three drugs approved for bipolar disorder, yet there are hundreds which might be prescribed for it. If the few that are generally accepted as "first line" treatments don't work for a person, that person is almost definitely in for a long and tortuous trial and error process for finding the right medications.
I was put on risperdal for the few months when my bipolar had been misdiagnosed as an unknown flavor of schizophrenia. In the parent's poster's case, the dopamine receptors were brought back to a normal state by blocking some of the over-activity. I can speak as a person whose dopamine receptors were blocked beyond what they should be.
I was definitely much more productive once my mania had subsided. The problem was, I didn't derive much pleasure from anything. Sex became as much a chore as washing the dishes. My favorite television shows could no longer hold my attention. I just started doing things that I thought I should be doing in order to occupy time. Life felt distant and flat. It was productive, but quite meaningless.
After a couple of months, I headed into a depression. It was the worst that I had ever had. I saw no point whatsoever in continuing my existence and no possible ground that could be gained from any course of action. Eventually, I managed to drive to my best friend's house and say "Unless you can see something worth living for in me, and get me help finding it, I will kill myself." The only thing that saved me was my willingness to trust my friends to do what was right.
When I finally found the right doc, obtained the proper diagnosis and was put on the right drugs (Lamictal and Effexor) it was a feeling not unlike watching the most beautiful sunrise ever. Untill then, I did not realize how awful of an experience that had been. Simply coming off the risperdal while starting my first (totally ineffective) drug regimen opened a new world.
Rent the movie Equilibrium. Not only is it a bad-ass action flick, it describes how I felt during that time. Now, one might think that this was a result of my bipolar, but I've never felt that way ever before in my life. I get depressed, but never before in that way. If I ever feel like that again, killing myself would definitely be an option.
Now, who wants to help test this article's process in humans?
*sigh* I knew I should have addressed the Jim Crow clause in my original response. I assume that you're refferring to the literacy test and poll tax laws that were used to disenfranchise many people of color.
This would be no more of a Jim Crow Law than obscure itemized deductions. You can currently save a bundle on taxes if you know the secret code known as "Tax Law." The well-educated have a distinct advantage in this regard and many are willing to pay money to have an expert help them. This would only be a Jim Crow law if it disenfranchised someone, or created a gross inequity that settled along racial or economic lines.
Abuse only becomes an issue if you are incredibly uncreative in the setup of the system. First off, don't make it a huge tax break. Fifty bucks will do. That's a fairly large motivator for those with lower incomes. Secondly, earmark the fifty bucks that would not be paid out to someone who didn't vote/pass the test for civic education classes. Since no money was previously going to any efforts like this, it's not treading on anyone's turf. The classes might be publicly run, or they might be privately run, or both. That's a detail to be I feel is a bit beyond the scope of a conversation here. (In other words, let's not get lost in the details before we've mapped out some of the big questions)
Then, make the questions 100% factual, but not based on rote memory. 100 questions, multiple choice. Have all possible questions and the correct answers freely available in every post office, police station, etc.
You still have the question of those who are illiterate. I think that's a toughie. I'll have to think on that one.
First, that statement attempts to hold one aspect of what I said as being representative of the whole. So, throw out that requirement. In it's place, put an extra question about the electoral college. Second, knowing the name of your congresscritter allows you to recognize their name in the headlines and it allows you to know who to vote out even if you know nothing of the competition.
While living in Tallahassee during the last election, I met quite a few people who wished they hadn't believed the polls and had instead gone out and voted. As a segue into the next section, is making someone grumble about how assinine it is that they have to vote when the stupid system effectively disenfranchises them such a horrible thing?
Let me know if you ever want help handing out fliers on this issue. I whole-heartedly agree, and I think that education is part of the solution. Aside from fiasco's like the last election, people can't understand how messed up our election system is untill the understand how it works. I would be happier if people simply understood that changes must happen at a st
Heck, why not double it up and provide a break for those who vote? I know that Australia imposes a fine on those who don't vote. We've all heard the slogan that it's not only your right to vote, but your duty. There's nothing that says you have to vote for anything, just show up to the voting booth and turn in a blank ballot, collect your receipt and move on with life.
There may be some flaws in this idea, but my gut says that it would improve the state of affairs. None of this would bar someone from taking part in the process, but instead would make it slightly more lucrative to take part.
While we're at it... Why not give welfare recipients a couple months to study and pass such a similar test or have their benifits cut? (but not eliminated.)
He did a great job with the keynote at H2K.
Listen here.
The thing I love about Biafra is that whether you agree with his ideas or not, you have to admit that he asks some interesting and important questions.
Many of the people on slashdot consider themselves to be activists of some degree or another. I'm right along with the majority of them who loathe microsoft, its business practices, its software and its API's.
/. today. See, I don't think I would have the right to continue complaining if I did not give honest and constructive criticism when asked for it. Yes, this information will probably be used to ill conceived ends, but I have to say that I tried.
Still, I can't quite summon myself to the kind of venom that I've seen on
What a coincidence! I just started on my own open source project to do this.
/dev/random > /dev/sound/dsp
Here's the source:
#!/bin/bash
cat
Any improvements and bugfixes welcome.
In Buddhism, Karma is not seen as a physical or metaphysical force that forces you to pay back your evil deeds, but rather as a psychological principle.
I could go on for a while about this, but I'll try to keep it short.
Ideally, once an action has happened, it is gone. Our mind, however, likes to hold on to what was and will not let go. Since the past is already gone, it is not a good place for the mind to dwell. Your karma is the part of you that holds onto the past and measure the present in terms of the past. If you measure the present in terms of the past, then you see more of what was than what is. A buddhist strives to see things as they are in the present (this is what meditation teaches) No buddhist claims that they will ever be free of karma, of their attachment to the past, but if they could be, they would be "enlightened."
A quick example: Let's say that you spent the past two years building your own house by hand. On the night that you move into the house, it burns down. If you hold on to the past and your memories of what was, then you will suffer at the thought of loss and the "wasted" time you spent building the house. If you see things as they are, then you will be happy that you are alive and start planning for a new place to live and perhaps rebuilding the house.
*sigh*
I'll bite.
A rational number is any number that can be expressed as an integer divided by an integer. You're right that Pi is not rational, which is to say that it cannot be represented as an integer divided by another integer.
You are *completely* wrong in your assumption that an irrational number cannot have a pattern to it. Consider, for example, the number 0.10011000111000011110000011111......
The pattern in this number is very obvious. It cannot, however be represented as an integer divided by another integer and is thus irrational.
You say that like they're being paid enough to deal with BS that was proposed.
If enough people on a regular enough basis were to do something like this, sure, it would work. On the other hand, going to the manager with one month's worth of grocery receipts(be sure to have them totaled,) and telling him politely that you, your friends and neighbors are not going to be shopping at his store anymore because of the policy is a much more productive way of dealing with the situation. Trust me, if you give the manager hard numbers, like a stack of receipts, it will make them sweat more than loading up a cart, throwing a tantrum in line, then making employees run all over the store putting up items will. If you do that, you're just a jackass, not an activist.
This won't piss off the store, it'll only piss off the powerless employees working for minimum wage. The manager won't be putting those items away, it'll be the bagboy. What good will that do?
It's sad to watch people complain to clerks like they can do something about a policy such as this. Having worked enough low end jobs in my life, I can tell you that the clerks can gripe and grumble, but they're not the ones who will get things changed. Find the Manager! Talk to his boss! Make *them* sweat. Don't piss on the guy who's working his ass of trying to get through college. Don't shit on the mother of three who's there just to make ends meet. Find someone who can actually change the policy, and then piss them off. Leave the little guys alone, they probably dislike this as much as you.
(post or moderate, that is the question)
Very simply, it could make some things faster.
In order to do some things, PGP must come up with large prime numbers. RSA key generation is the only thing I'm sure of, but there may be others. Before the primality test, it took much, much longer to make sure a particular number was prime.
It shouldn't do anything to the security of PGP, since the encryption is still based on factoring large numbers, a process which isn't (significantly) sped up by the primality test.
I remember a documentary from the gulf war that showed a huge spool of explosive cable shot from the back of the truck. When the cable exploded, it took the mines with it. Other than having to carry the cable with you, I wonder why that isn't sufficient.
Hmmm... Good thoughts.
Personally, I think that signing individual users would be better than signing individual files. If you sign individual files, then someone can flood the system with bogus signings from a large number of random usernames, but if I instead assign trust only to people, it's much harder for a bogus client to get deeply into the web. There's also the issue of keeping track of trust. If I sign individual files, then I have to keep track of all those signatures. Since there are a great many fewer people than files, the amount of persistant information would be much smaller when only signing people. If you rank trust as a proportion, say 0% to 100%, then leaps through the trust web can be easy to calculate. I trust Toni 80%. Toni trusts Jenny 50%, thus I trust Jenny 40%. Let's say that Jenny and Toni are on the network when I connect, and I have don't even know of Jenny's existance yet. When I connect, I ask Toni to pass me his trust information. He passes me a file containing his signature for Jenny along with her current IP address. I calculate trust, mark it in my trust file, and then send Jenny my signed 40% trust. She can choose to discard this signature if she thinks it's too low (trust levels below 10%, for instance) and thus keep the really pointless entries out of the system. If, later, I decide that Jenny is actually a pretty good person, and bump her trust up to 75%, I would send her that signature, and she would keep the higher rating. This would keep the web large and open, but make it hard to do anything without a reputation or an introduction. The people who give out trust willy-nilly or who pass bad files will quickly be marginalized, but the people who build a good reputation will gain a larger base of files to choose from and contribute more to the system. The idea needs some work, but that's the basic algorithm as I see it. Oh, and there should be some way to tell people that they have a bum file and allow them to correct the situation before you automatically distrust them.
The biggest problem would be finding the IP address of other people you already trust. Any system that is specifically designed to keep track of users would be open to legal and DOS attacks. So, why not use something such as IRC. I tell all my friends to meet me in #SecureIndustrialMusic or some other such channel, and we could get each other's ip adress and become part of the network. Better yet, use some system that noone would dare to try and close down, like AIM or ICQ or YIM...or any combination of methods, but only for finding the people you already trust. Heck, if you did this in python, it would be portable and most of the routines you need are already contained in only two libraries: twisted has IRC, AIM and generic net libraries and PyCrypto has public key libraries. (that is, if my sleep deprived memory serves me correctly...I could be wrong about those libraries) Then slap a nice little tcl/tk front end onto it and you'd have a good system Moreover, this could help pare down each user's database size. If I hadn't seen a person in, say, six months, then I could just stop trusting them altogether.
In short, I think it's doable and I like the fact that it lends itself well to small groups slowly building reputations and relationships with each other. It is a system that self-corrects and builds up value over time. It rewards reliability, and makes poor-sportsmanship difficult to maintain for any length of time.
Email me if you want to talk about it further.
Over the past couple of months, I've been thinking that the solution might be a web-of-trust system similar to pgp key signings. It doesn't seem like such a thing would be too hard to impliment with actual key signings, perhaps even with gpg and the gnutella codebase. This would certainly reduce the size of the network of p2p clients, but I think most people tend to listen to music that's owned by someone within three or four degrees of seperation from themselves. Personally, I only use p2p for finding bands that have have been reccomended by friends, so it would almost certainly be within a couple of hops of trust from myself.
The only problem I can see with the moderation system that you're suggesting is that there would have to be a central authority for mod points. In the current political and legal climate, that's a direct weakness. You could, conceivably, combine the two systems. So, I could rate everyone that I've downloaded from based on Quality of Service and that would enter a special file, which could be picked up by each client that has trust in me. The client would then weight the entries based on how much they trust me. For instance, if they only had 50% trust in me, then my ratings could be cut in half. They could then decide on a threshold, below which they won't do business with a client. Someone could be allowed to enter into the network.
This system has a lot of possibilites. It would keep out unwanted parties, but also allow people to come in at a low level of trust and build from that. If you made it a generic fileswapper with searchable metadata (such as gif comments and id3 tags) then also allowed ssl transfers, it would be almost impossible to track.
Sorry if this is all a bit muddled and choppy. I've been up for more than 36 hours. Let me know if this sounds at all reasonable.
Oh, there was plenty of settlement in the Americas well before even Columbus got there. History did not begin in Europe.
Dude, don't go around poking Sigfried and Roy... You don't know where they've been.
"This again proves, how much better software can be, if you remove management, lawyers, sales department etc. and make good programmers work together without short-term profit in mind."(emphasis mine)
When I read that scentence, I got the image of a darkly lit slave-pit filled with coders where the corporate entities you listed had been replaced by two or three dark-overlords whipping their slaves into submission. Why don't replace the word "make" with "allow." It is a much more pleasant term. Other than that, I wholeheartedly agree with sentiment you expressed.
So, why not post those somewhere?
A suite of common office documents can help devlopers know what type of features a non-technical job requires.
Who knows, it might become the kernel of a set of standards used to compare Word import filters across the board.
I get real sick of this line of reasoning. It's put out there everytime someone asks for information on a subject for a school project, and it's nothing but a troll. (yeah, I'm biting)
If this person relies solely on slashdot to do his homework, then yes, he is dumb. That's because slashdot is a gamble. However, slashdot is a very good place to get in touch with people who keep up with computer related. What if someone points him to the one resource he couldn't find on his own that can turn a mediocre paper into a stellar one? He would be dumb *not* to suppliment his normal research with that resource.
If you don't want to help him, then don't. However you waste your time if you trot out this dead horse every time and you waste your mod points if you mod it up.
--
I only post to slashdot when I'm sleep deprived.
Watching for pupil dilation considered harmfull.
Movie at 11.
--
I only post to slashdot when I'm sleep deprived.