No more free email?!
March 5, 2004 11:32 AM Subscribe
Bill Gates proposes an end to free email
If the U.S. Postal Service delivered mail for free, our mailboxes would surely runneth over with more credit-card offers, sweepstakes entries, and supermarket fliers. That's why we get so much junk e-mail: It's essentially free to send. So Microsoft Corp. chairman Bill Gates, among others, is now suggesting that we start buying "stamps" for e-mail.
If the U.S. Postal Service delivered mail for free, our mailboxes would surely runneth over with more credit-card offers, sweepstakes entries, and supermarket fliers. That's why we get so much junk e-mail: It's essentially free to send. So Microsoft Corp. chairman Bill Gates, among others, is now suggesting that we start buying "stamps" for e-mail.
...and what of the developing world? They don't get to participate in the global information sharing that is email? Even worse, if their fees are waived, all spam will move to their nations to skirt these costs.
posted by mathowie at 11:38 AM on March 5, 2004
posted by mathowie at 11:38 AM on March 5, 2004
Um, my mailbox does runneth over with credit-card offers, pennysavers, "have you seen me?"s, and similar forms of junk mail. It would appear to me that the price of postage itself isn't a strong deterrent to sending unwanted mailings.
posted by herc at 11:39 AM on March 5, 2004
posted by herc at 11:39 AM on March 5, 2004
Before anybody goes off (or, god forbid, spells Microsoft with a dollar sign), please read the article. wsg, I would have specified this in your post, it's way too sensational as is:
instead of paying a penny, the sender would "buy" postage by devoting maybe 10 seconds of computing time to solving a math puzzle. The exercise would merely serve as proof of the sender's good faith.
Which is not a terrible idea, but sounds hard to implement, whereas the other ideas in the article (from goodmail), which are just stupid.
I think it's pretty clear the solution doesn't lie down this road, the solution involves authentication in SMTP.
On preview: too late, i guess.
posted by malphigian at 11:40 AM on March 5, 2004
instead of paying a penny, the sender would "buy" postage by devoting maybe 10 seconds of computing time to solving a math puzzle. The exercise would merely serve as proof of the sender's good faith.
Which is not a terrible idea, but sounds hard to implement, whereas the other ideas in the article (from goodmail), which are just stupid.
I think it's pretty clear the solution doesn't lie down this road, the solution involves authentication in SMTP.
On preview: too late, i guess.
posted by malphigian at 11:40 AM on March 5, 2004
One problem, as the article points out, is with people who belong to e-mail lists. If you want to send one email to multiple people, do you have to pay more? What about group discussions via e-mail? I just don't see this working. Spam sucks, but I don't think charging for e-mail is the way to go. If it deters non-spammers from sending e-mail, it's no better than an overinclusive e-mail filter.
posted by Tin Man at 11:45 AM on March 5, 2004
posted by Tin Man at 11:45 AM on March 5, 2004
The only problem with that is, in my opinion, that people like myself who are responsible mail providers (self-linkage to my ToS), but non profit, will be forced offline. I have been running a server for about 3 years and provide free mail for about 50 friends and family. I do it because I enjoy it, not because I want to make money off of it. I have no banners and no advertising (except, perhaps, the attribution for web design done by a friend who's trying to launch a small business). I also really don't have computing resources to meet the distributing computing pan.
So, big money corporate people like Gates, who can pay someone to make sure he always has enough e-postage or have plenty of spare machines lying around for the distributed computing solution can bump out those of us who do it for the love of the game, so to speak.
Another step in the growing commercial industrialization of the internet.
(Plus - one thing I've never really seen mentioned is where this money is supposed to go. The receiving host? Some nebulous government agency? ICANN? The USPS?)
posted by Samizdata at 11:45 AM on March 5, 2004
So, big money corporate people like Gates, who can pay someone to make sure he always has enough e-postage or have plenty of spare machines lying around for the distributed computing solution can bump out those of us who do it for the love of the game, so to speak.
Another step in the growing commercial industrialization of the internet.
(Plus - one thing I've never really seen mentioned is where this money is supposed to go. The receiving host? Some nebulous government agency? ICANN? The USPS?)
posted by Samizdata at 11:45 AM on March 5, 2004
Email isn't free. I have to pay my ISP charges every month.
And anyway, email is just data. Mind you, it's data sent over port 25 instead of 80, but how is this supposed to affect web-based email? Do you charge for data because it's "email" 1's and 0's as opposed to "surfing" or "downloading" 1's and 0's?
As for the argument of "In real world (application), this sort of thing costs money. So it should be with the online version of (application)." Mail costs money not because they want to prevent people from spamming mailboxes. It costs money because you have to pay people to process your mail, drive the little truck around and put it in the box. Communication systems are social commodities that we deem important for the prosperity of our civilization, which is why it's controlled by the government and prices are kept at reasonable levels.
Using cost as a deterrent to illicit use is absurdly stupid, particularly to something as important as free and open communication. You want to stop spam? Fine. Build better filters, write better laws that go after spammers, etc. Putting a tax (or price) on email is like putting a tax on paper. Because, you know, criminals use paper.
posted by Civil_Disobedient at 11:48 AM on March 5, 2004
And anyway, email is just data. Mind you, it's data sent over port 25 instead of 80, but how is this supposed to affect web-based email? Do you charge for data because it's "email" 1's and 0's as opposed to "surfing" or "downloading" 1's and 0's?
As for the argument of "In real world (application), this sort of thing costs money. So it should be with the online version of (application)." Mail costs money not because they want to prevent people from spamming mailboxes. It costs money because you have to pay people to process your mail, drive the little truck around and put it in the box. Communication systems are social commodities that we deem important for the prosperity of our civilization, which is why it's controlled by the government and prices are kept at reasonable levels.
Using cost as a deterrent to illicit use is absurdly stupid, particularly to something as important as free and open communication. You want to stop spam? Fine. Build better filters, write better laws that go after spammers, etc. Putting a tax (or price) on email is like putting a tax on paper. Because, you know, criminals use paper.
posted by Civil_Disobedient at 11:48 AM on March 5, 2004
Pay postage for email? Oh, come on, folks - this is the oldest hoax in the book.
It sure is strange that Bill Gates fell for it, though. He must be hard up for money after giving out all those $100 bills for those emails I forwarded.
posted by soyjoy at 12:00 PM on March 5, 2004
It sure is strange that Bill Gates fell for it, though. He must be hard up for money after giving out all those $100 bills for those emails I forwarded.
posted by soyjoy at 12:00 PM on March 5, 2004
Don't be mistaken.
There are tons of possible solutions for this problem - he's just proposing the one that would make Microsoft the most money.
posted by cinderful at 12:01 PM on March 5, 2004
There are tons of possible solutions for this problem - he's just proposing the one that would make Microsoft the most money.
posted by cinderful at 12:01 PM on March 5, 2004
Yeah, I don't even get the idea behind this - real world junk mail is hardly tamed by the price of postage. Maybe it's worse for PO Boxes than for street addresses, but mail = bills, magazines, and lots and lots of junk. Personal messages are extremely rare, since they've shifted to email for the most part. Junk mail has not shifted to email at nearly the same rate. (I mean, there's new junk online, but the old junk still do a lot of business by way of bulk mail).
posted by mdn at 12:03 PM on March 5, 2004
posted by mdn at 12:03 PM on March 5, 2004
malphigian, if Gates is talking about using calculation as a form of payment, he's just claiming to have invented Adam Back's HashCash.
The problem with systems like this is that Moore's Law gets in the way. A useful payment in 2004 will probably be a spammer's delight in 2006, so you have to keep upping the ante. Except you're also penalizing people with less than up-to-date computer hardware, who have to "pay more" for the privilege of sending an email. Of course, if you're Microsoft, you want everyone to have current hardware and software...
The initial idea sounds okay, but it doesn't scale into reality very well.
posted by lowlife at 12:05 PM on March 5, 2004
The problem with systems like this is that Moore's Law gets in the way. A useful payment in 2004 will probably be a spammer's delight in 2006, so you have to keep upping the ante. Except you're also penalizing people with less than up-to-date computer hardware, who have to "pay more" for the privilege of sending an email. Of course, if you're Microsoft, you want everyone to have current hardware and software...
The initial idea sounds okay, but it doesn't scale into reality very well.
posted by lowlife at 12:05 PM on March 5, 2004
Civil, look at this:
You pay 45 bucks a month, say, to your ISP.
Now, you pay 40 bucks a month, plus 1 cent for each email.
Okay, you're back up to 45 bucks, maybe, by the end of the month. If it's more, you should have taken the email-heavy package.
Now, a spammer, wishing to send out a million emails every day, will pay a million cents. Or, $10000 per day. Or, they'll stop sending spam.
And, if this is part of the protocol, then nobody can avoid it. It's there.
The problem with filters is it still congests the network.
I pray M$ doesn't implement it.
posted by jon_kill at 12:06 PM on March 5, 2004
You pay 45 bucks a month, say, to your ISP.
Now, you pay 40 bucks a month, plus 1 cent for each email.
Okay, you're back up to 45 bucks, maybe, by the end of the month. If it's more, you should have taken the email-heavy package.
Now, a spammer, wishing to send out a million emails every day, will pay a million cents. Or, $10000 per day. Or, they'll stop sending spam.
And, if this is part of the protocol, then nobody can avoid it. It's there.
The problem with filters is it still congests the network.
I pray M$ doesn't implement it.
posted by jon_kill at 12:06 PM on March 5, 2004
I would have specified this in your post, it's way too sensational as is
Good point. Gates proposes the math puzzle idea (which doesn't sound very practical to me), but there are others who are talking about real money.
posted by wsg at 12:06 PM on March 5, 2004
Good point. Gates proposes the math puzzle idea (which doesn't sound very practical to me), but there are others who are talking about real money.
posted by wsg at 12:06 PM on March 5, 2004
The Road Ahead 1st Edition -there is no internet-
Yup, I'd say his abilities as seer for technology are well documented.
posted by nofundy at 12:16 PM on March 5, 2004
Yup, I'd say his abilities as seer for technology are well documented.
posted by nofundy at 12:16 PM on March 5, 2004
I'm amazed that after my post pointed out the line, multiple people (even a gold star winner!) said something about microsoft just trying to make money. Someone even used the freaking dollar sign. Jeez.
Let me try again:
The MS Research Lab Team suggested "paying" by having the sender's computer solve a small math problem, no money to Microsoft whatsoever.
Sorry if I sound snarky, I'm no MS cheerleader, but reflexive uninformed bashing is really tiresome.
lowlife makes a good point about why the computation method isn't terribly feasible, although I think even a computation that takes 10 seconds now and 1 second in a couple years would slow the spammers down a bit.
The first item on the agenda really needs to be making it harder to spoof the sender.
posted by malphigian at 12:19 PM on March 5, 2004
Let me try again:
The MS Research Lab Team suggested "paying" by having the sender's computer solve a small math problem, no money to Microsoft whatsoever.
Sorry if I sound snarky, I'm no MS cheerleader, but reflexive uninformed bashing is really tiresome.
lowlife makes a good point about why the computation method isn't terribly feasible, although I think even a computation that takes 10 seconds now and 1 second in a couple years would slow the spammers down a bit.
The first item on the agenda really needs to be making it harder to spoof the sender.
posted by malphigian at 12:19 PM on March 5, 2004
Well, if this gets implemented then I'll just have to persuade my circle of friends to switch to sending messages between ourselves on a different port.
To start with we can just send SMTP, just not on port 25. When there's enough computing power to check content then we'll have to come up with a new protocol, or maybe just encrypt the content.
Of course, it may be that Gates wants to create a cut down version of the Internet, where everyone gets to pay Microsoft to take part (kind of an AOL on steroids). Those who paid would get to live in the officially sanctioned, safe Internet, while the rest of us would have to run the risk of being attacked by all sorts of evil-doers.
posted by daveg at 12:24 PM on March 5, 2004
To start with we can just send SMTP, just not on port 25. When there's enough computing power to check content then we'll have to come up with a new protocol, or maybe just encrypt the content.
Of course, it may be that Gates wants to create a cut down version of the Internet, where everyone gets to pay Microsoft to take part (kind of an AOL on steroids). Those who paid would get to live in the officially sanctioned, safe Internet, while the rest of us would have to run the risk of being attacked by all sorts of evil-doers.
posted by daveg at 12:24 PM on March 5, 2004
real world junk mail is hardly tamed by the price of postage
wanna bet? Here's my mailbox:
Paper "junk" received today: 3
E-mail "junk" received today: 125
And we still have another nine hours left in today.
posted by ednopantz at 12:27 PM on March 5, 2004
wanna bet? Here's my mailbox:
Paper "junk" received today: 3
E-mail "junk" received today: 125
And we still have another nine hours left in today.
posted by ednopantz at 12:27 PM on March 5, 2004
Now, a spammer, wishing to send out a million emails every day, will pay a million cents. Or, $10000 per day. Or, they'll stop sending spam.This would not prevent spam in the long term. It'd just change how they did their business.
And, if this is part of the protocol, then nobody can avoid it. It's there.
The problem with filters is it still congests the network.
I pray M$ doesn't implement it.
Right now they just send it out and forge a few of the addresses so it can't be tracked back easily. With the penny/email charge (or whatever it is) this will be closed to them but they'll resort to other methods. The first step would probably be massive spamming to popular bulletin boards.
The eventual step would be the introduction of trojan horses that allow them to send email from your computer. Spam coming from your MAC address isn't distinguishable from regular email in at least the short term. If they were smart, and I'm sure that somebody eventually would be, they'd rely on distributed spam. Rather than sending a million spams from a single computer (which would be easily detectable) send 100 spams from 10000 computers.
Spam is going to be around until the community is smart enough to realize that the services offered in spam are bogus, illegal or both. In other words, spam is here to stay.
posted by substrate at 12:48 PM on March 5, 2004
So I guess the lesson of this thread is:
Not only does no one read the article before posting, they also don't read anyone else's comments. Good to know.
posted by malphigian at 12:59 PM on March 5, 2004
Not only does no one read the article before posting, they also don't read anyone else's comments. Good to know.
posted by malphigian at 12:59 PM on March 5, 2004
Why doesn't the post office just charge commercial e-mail? That would seem to make too much sense.
posted by Bag Man at 1:02 PM on March 5, 2004
posted by Bag Man at 1:02 PM on March 5, 2004
malphigian, I read the article and your post and you are, of course, correct.
As for the post office/email analogy, there's a slightly different analogy that IMO works better: it's the unauthorized use of mailboxes. Isn't it true that (in the US, anyway) your mailbox is technically the property Postal Service? If I remember it right they take a dim view of people going around and putting fliers in mailboxes: in fact, if they discover that somebody's done that all over a neighborhood, the mailman collects the fliers, the Postal Police (who do exist) track down the perpetrator and charge him the mailing price of every flier that was discovered. And the system works perfectly: the mailman won't bust you for leaving a note in your neighbor's box, but the threshhold at which he gets annoyed is so low that you practically never get unauthorized stuff in your actual mailbox. I think those are the characteristics that a really great spam protection scheme should have.
On the hand, I find fliers on/under/around my door and car every day of the week.
posted by coelecanth at 1:14 PM on March 5, 2004
As for the post office/email analogy, there's a slightly different analogy that IMO works better: it's the unauthorized use of mailboxes. Isn't it true that (in the US, anyway) your mailbox is technically the property Postal Service? If I remember it right they take a dim view of people going around and putting fliers in mailboxes: in fact, if they discover that somebody's done that all over a neighborhood, the mailman collects the fliers, the Postal Police (who do exist) track down the perpetrator and charge him the mailing price of every flier that was discovered. And the system works perfectly: the mailman won't bust you for leaving a note in your neighbor's box, but the threshhold at which he gets annoyed is so low that you practically never get unauthorized stuff in your actual mailbox. I think those are the characteristics that a really great spam protection scheme should have.
On the hand, I find fliers on/under/around my door and car every day of the week.
posted by coelecanth at 1:14 PM on March 5, 2004
To start with, this seems to be another example of Gates getting headlines for repeating an idea that has been floating around for a few years.
The "pay to send" model would require a lot of work to provide sender authentication and avoid spoofing. The big question is who gets the cash? Perhaps the most promising aspect about it is that it turns open mail relays into a financial liability.
A more promising model is to make it more difficult to send large quantities of email from a single computer. Again, this is hardly anything revolutionary from Microsoft. Some networks are already switching to SMTP-over-SSL to reduce spam. Dynamic greylisting is another option. When a computer at a previously unknown ip address starts sending bulk email into the system, the recipient mail server starts bogging down the connection.
posted by KirkJobSluder at 1:22 PM on March 5, 2004
The "pay to send" model would require a lot of work to provide sender authentication and avoid spoofing. The big question is who gets the cash? Perhaps the most promising aspect about it is that it turns open mail relays into a financial liability.
A more promising model is to make it more difficult to send large quantities of email from a single computer. Again, this is hardly anything revolutionary from Microsoft. Some networks are already switching to SMTP-over-SSL to reduce spam. Dynamic greylisting is another option. When a computer at a previously unknown ip address starts sending bulk email into the system, the recipient mail server starts bogging down the connection.
posted by KirkJobSluder at 1:22 PM on March 5, 2004
My junk mail filter works fine. If no one is reading the spam, eventually there will be no spam. My browser blocks popups, too, which MSFT's browser doesn't, yet. Popups bother me a lot more than spam because they interrupt what I'm doing.
posted by Slagman at 1:32 PM on March 5, 2004
posted by Slagman at 1:32 PM on March 5, 2004
The reasons this won't work:
- Spammers send email through hijacked computers. They won't pay (either $s or clock cycles). Innocent third parties will.
- Spam brings in $12 billion a year. More than enough to fund the server farms which would be employed to get around this.
- Introducing slightly difficult math problems as a barrier doesn't scale well in the computer world.
- People with slower computers won't be willing to have their computers grind to a halt for several minutes every time they send an email.
- Much legitimate email is computer generated. I rely on several mailing lists which would be crippled by this.
- Trusting Microsoft with something like this is asking for trouble. Do we really want email to be closed source, proprietary, over-engineered, and poorly implimented?
Don't let robber barons who think monopolies drive innovation build your free mass communication system. Right?
posted by y6y6y6 at 1:33 PM on March 5, 2004
- Spammers send email through hijacked computers. They won't pay (either $s or clock cycles). Innocent third parties will.
- Spam brings in $12 billion a year. More than enough to fund the server farms which would be employed to get around this.
- Introducing slightly difficult math problems as a barrier doesn't scale well in the computer world.
- People with slower computers won't be willing to have their computers grind to a halt for several minutes every time they send an email.
- Much legitimate email is computer generated. I rely on several mailing lists which would be crippled by this.
- Trusting Microsoft with something like this is asking for trouble. Do we really want email to be closed source, proprietary, over-engineered, and poorly implimented?
Don't let robber barons who think monopolies drive innovation build your free mass communication system. Right?
posted by y6y6y6 at 1:33 PM on March 5, 2004
Paper "junk" received today: 3
E-mail "junk" received today: 125
well, you have a different ratio than I do, to start with - on my verizon account, I get like four or five email junk messages a day (hotmail gets a lot more, but it goes into the junk folder and I only check that once a week or so - still, I don't think it's more than 15 or 20 in a day).
But that aside, what's your ratio of non-junk paper mail to non-junk email? Isn't part of the disparity simply that you get a lot more email than paper mail? One reason for that is that it's much easier. Another reason is that it's free.
I'm not saying this is a terrible idea, but I think it's naive to expect it to end spam. The question is really whether it would help enough to offset the downside of having to pay for our own email...
posted by mdn at 1:39 PM on March 5, 2004
E-mail "junk" received today: 125
well, you have a different ratio than I do, to start with - on my verizon account, I get like four or five email junk messages a day (hotmail gets a lot more, but it goes into the junk folder and I only check that once a week or so - still, I don't think it's more than 15 or 20 in a day).
But that aside, what's your ratio of non-junk paper mail to non-junk email? Isn't part of the disparity simply that you get a lot more email than paper mail? One reason for that is that it's much easier. Another reason is that it's free.
I'm not saying this is a terrible idea, but I think it's naive to expect it to end spam. The question is really whether it would help enough to offset the downside of having to pay for our own email...
posted by mdn at 1:39 PM on March 5, 2004
I'm becoming increasingly convinced Bayesian filtering is what will kill spam. It is not wholly dependent on a technological solution. People have to make choices to teach the filter what is spam and what is not. This decision forces spammers to use increasingly creative methods to get by filters.
v!@grA anyone? A few more years and spam will become so obfuscated, random text generators will accidentally transponse vagina for penis.
Somehow, I don't think vagina enlargements will make money for spammers.
posted by pedantic at 1:39 PM on March 5, 2004
v!@grA anyone? A few more years and spam will become so obfuscated, random text generators will accidentally transponse vagina for penis.
Somehow, I don't think vagina enlargements will make money for spammers.
posted by pedantic at 1:39 PM on March 5, 2004
Paper "junk" received today: 3
E-mail "junk" received today: 125
And we still have another nine hours left in today.
So you're saying there's still time for the paper junk to catch up?
posted by Fezboy! at 1:59 PM on March 5, 2004
E-mail "junk" received today: 125
And we still have another nine hours left in today.
So you're saying there's still time for the paper junk to catch up?
posted by Fezboy! at 1:59 PM on March 5, 2004
y6y6y6:
- Spammers send email through hijacked computers. They won't pay (either $s or clock cycles). Innocent third parties will.
But wouldn't a system that slows down the ammount of messages that can be sent from a hijacked computer reduce the impact of trojans and viruses? And hrm, a slight pardon if I'm slightly less than sympathetic on the money problem. Most of the recent virus, trojans and worms have occured primarily due to the negligence of multiple parties. Treating the loss of control of your personal compter as equivalent to loss of control of your credit card or calling card number would be the critical leap needed to get most people to take security seriously.
Of course there is a nice mechanism for dealing with accidental trojans. If you loose your credit card once, many credit card companies will swallow the cost of fraud while minimizing the impact. Repeated claims of loss or compromise should send up red flags.
- Spam brings in $12 billion a year. More than enough to fund the server farms which would be employed to get around this.
Most spam is driven by the low costs of sending thousands of messages. Economic studies of spam suggest that will not take large increases in the costs of sending spam to result in dramatic reductions. (Blocking 50% of messages for 50% of the population almost doubles the costs.)
- People with slower computers won't be willing to have their computers grind to a halt for several minutes every time they send an email.
It actually would not take that bad of a problem, only doubling or trippling the ammount of time to send a message. (An even better strategy would be progressive delays for unknown hosts attempting to send large numbers of mail messages every few minutes.)
The mailing list problem is a bit more serious.
mdn: I'm not saying this is a terrible idea, but I think it's naive to expect it to end spam.
I don't think anyone is under the illusion that anything will end spam. Instead, what is being proposed are methods for shifting the economics of spam. Currently, the bandwidth costs for spam are trivial compared to the potential return. Any system or combination of systems that increase bandwidth costs can do a great deal for stopping spam. The trick is to find a way to increase the costs in such a way that legitimate users don't notice but spammers do. A legitimate user who sends 500 messages per day should not notice. A spammer who tries to send 50,000 per day from a host should.
posted by KirkJobSluder at 2:33 PM on March 5, 2004
- Spammers send email through hijacked computers. They won't pay (either $s or clock cycles). Innocent third parties will.
But wouldn't a system that slows down the ammount of messages that can be sent from a hijacked computer reduce the impact of trojans and viruses? And hrm, a slight pardon if I'm slightly less than sympathetic on the money problem. Most of the recent virus, trojans and worms have occured primarily due to the negligence of multiple parties. Treating the loss of control of your personal compter as equivalent to loss of control of your credit card or calling card number would be the critical leap needed to get most people to take security seriously.
Of course there is a nice mechanism for dealing with accidental trojans. If you loose your credit card once, many credit card companies will swallow the cost of fraud while minimizing the impact. Repeated claims of loss or compromise should send up red flags.
- Spam brings in $12 billion a year. More than enough to fund the server farms which would be employed to get around this.
Most spam is driven by the low costs of sending thousands of messages. Economic studies of spam suggest that will not take large increases in the costs of sending spam to result in dramatic reductions. (Blocking 50% of messages for 50% of the population almost doubles the costs.)
- People with slower computers won't be willing to have their computers grind to a halt for several minutes every time they send an email.
It actually would not take that bad of a problem, only doubling or trippling the ammount of time to send a message. (An even better strategy would be progressive delays for unknown hosts attempting to send large numbers of mail messages every few minutes.)
The mailing list problem is a bit more serious.
mdn: I'm not saying this is a terrible idea, but I think it's naive to expect it to end spam.
I don't think anyone is under the illusion that anything will end spam. Instead, what is being proposed are methods for shifting the economics of spam. Currently, the bandwidth costs for spam are trivial compared to the potential return. Any system or combination of systems that increase bandwidth costs can do a great deal for stopping spam. The trick is to find a way to increase the costs in such a way that legitimate users don't notice but spammers do. A legitimate user who sends 500 messages per day should not notice. A spammer who tries to send 50,000 per day from a host should.
posted by KirkJobSluder at 2:33 PM on March 5, 2004
I think this is a fantastic idea! Microsoft can start charging for their "email" protocol, and I'll sit over here with a mail transfer agent using some other protocol entirely! Maybe I'll exchange email over HTTP. Or lob spool files over FTP. Or the ever-popular UUCP.
Or even better! I've got it! I'll write up an RFC for some sort of... of... "Subverting Microsoft Tax Protocol!" All I'd have to do, just like Microsoft would, is convince the entire world to refuse to use any transport protocol but mine, and switch to it instantaneously so that interoperability doesn't become its weakness. After all, we don't want all that spam coming in from the "old" email infrastructure, so this new stuff had better not connect to it at all.
posted by majick at 3:50 PM on March 5, 2004
Or even better! I've got it! I'll write up an RFC for some sort of... of... "Subverting Microsoft Tax Protocol!" All I'd have to do, just like Microsoft would, is convince the entire world to refuse to use any transport protocol but mine, and switch to it instantaneously so that interoperability doesn't become its weakness. After all, we don't want all that spam coming in from the "old" email infrastructure, so this new stuff had better not connect to it at all.
posted by majick at 3:50 PM on March 5, 2004
"But wouldn't a system that slows down the ammount of messages that can be sent from a hijacked computer reduce the impact of trojans and viruses?"
No. To send the same amount of spam spammers would just have to hijack more computers. This doesn't seem to be a bottle neck they have trouble with.
"Blocking 50% of messages for 50% of the population almost doubles the costs."
Which would be overcome by spammers doubling the number of spams they send. Not good.
"It actually would not take that bad of a problem, only doubling or trippling the ammount of time to send a message."
Really? Then couldn't spammers overcome this by just doubling or tripling the number of computers they use?
"I don't think anyone is under the illusion that anything will end spam."
Moving to certification for authenticating SMTP servers, with dynamic grey lists, centralized opt-out reporting would end spam. But that would be hard. So we'll never do it.
posted by y6y6y6 at 4:02 PM on March 5, 2004
No. To send the same amount of spam spammers would just have to hijack more computers. This doesn't seem to be a bottle neck they have trouble with.
"Blocking 50% of messages for 50% of the population almost doubles the costs."
Which would be overcome by spammers doubling the number of spams they send. Not good.
"It actually would not take that bad of a problem, only doubling or trippling the ammount of time to send a message."
Really? Then couldn't spammers overcome this by just doubling or tripling the number of computers they use?
"I don't think anyone is under the illusion that anything will end spam."
Moving to certification for authenticating SMTP servers, with dynamic grey lists, centralized opt-out reporting would end spam. But that would be hard. So we'll never do it.
posted by y6y6y6 at 4:02 PM on March 5, 2004
I wonder how much spam could be prevented by simply making Windows a secure enough system that "zombie-bots" aren't a problem.
Naaah that won't work.
posted by ilsa at 4:13 PM on March 5, 2004
Naaah that won't work.
posted by ilsa at 4:13 PM on March 5, 2004
yawnz0r
There are tons of solutions out there, but no one bothers. A great first start would be to start checking sender permited from DNS records (which would get rid of TONS of spam right away).
This idea is as retarded as it always has been.
posted by delmoi at 6:14 PM on March 5, 2004
There are tons of solutions out there, but no one bothers. A great first start would be to start checking sender permited from DNS records (which would get rid of TONS of spam right away).
This idea is as retarded as it always has been.
posted by delmoi at 6:14 PM on March 5, 2004
y6y6y6:
I think you are misunderstanding the economics involved. Finding new hosts is the most expensive aspect of sending spam. Spammers have only a limited window of opportunity to use a host before it is blacklisted, or changes ip#. As a result, spammers can't simply avoid an increase in costs by increasing the number of hosts they use. Getting new hosts requires work, eliminating the "free lunch" motivation.
All spam prevention comes down to economics. Greylisting is just another way of increasing spammer costs by forcing inbound email from unknown hosts to be submitted twice. Real-Time Blacklists limit the ammount of time that a host remains a viable source for spam. Tar pits punish errors by keeping the spammer on the line for an extended period of time.
posted by KirkJobSluder at 6:28 PM on March 5, 2004
I think you are misunderstanding the economics involved. Finding new hosts is the most expensive aspect of sending spam. Spammers have only a limited window of opportunity to use a host before it is blacklisted, or changes ip#. As a result, spammers can't simply avoid an increase in costs by increasing the number of hosts they use. Getting new hosts requires work, eliminating the "free lunch" motivation.
All spam prevention comes down to economics. Greylisting is just another way of increasing spammer costs by forcing inbound email from unknown hosts to be submitted twice. Real-Time Blacklists limit the ammount of time that a host remains a viable source for spam. Tar pits punish errors by keeping the spammer on the line for an extended period of time.
posted by KirkJobSluder at 6:28 PM on March 5, 2004
I suggest Bill Gates suck my balls.
I betting he could get you to suck his for what he makes in 20 minutes . Just saying.
posted by thirteen at 7:46 PM on March 5, 2004
OK guys, let’s get some facts straight: Charlie Rose had an interview with Bill Gates on Wednesday (hopefully the interview will show up in the archive). Very likely CNN got the news but did not mention the interview. Please correct me if you know of any other recent source where Gates mentioned the "pay for email" stuff.
Indeed, this was only one solution to the problem (he spoke of several), and it was not to benefit MS, at least not directly. The way that I interpret his answer is a way to link email with authentication and, thus, with DRM. Gates has vision and is aware of other ideas floating around. He just made a push for the one that he/MS prefers.
He also mentioned, in a somehow different context, that he expects better search tool and applications that be able to react and adapt to different situations. This is the case of Bayesian filters and MS-Office's Clippy. Or, if you want to fully express the idea, the context recognition chip, first mentioned by Carl Sagan in Contact. But we do not want tools that just empower the user and add no benefit to company, do we?
Meanwhile, Gates had us talking about his idea. So, shall we accuse CNN of viral marketing?
posted by MzB at 8:49 PM on March 5, 2004
Indeed, this was only one solution to the problem (he spoke of several), and it was not to benefit MS, at least not directly. The way that I interpret his answer is a way to link email with authentication and, thus, with DRM. Gates has vision and is aware of other ideas floating around. He just made a push for the one that he/MS prefers.
He also mentioned, in a somehow different context, that he expects better search tool and applications that be able to react and adapt to different situations. This is the case of Bayesian filters and MS-Office's Clippy. Or, if you want to fully express the idea, the context recognition chip, first mentioned by Carl Sagan in Contact. But we do not want tools that just empower the user and add no benefit to company, do we?
Meanwhile, Gates had us talking about his idea. So, shall we accuse CNN of viral marketing?
posted by MzB at 8:49 PM on March 5, 2004
As a host of several non-profit mailing lists (some with several hundred members, all willing subscribers, thank you very much) this strikes me as pretty darned close to the stupidest possible way to handle this. If every post to the list slows our mail server to a crawl that would be a serious problem. So no, no, a thousand times no, Bill.
posted by litlnemo at 8:53 PM on March 5, 2004
posted by litlnemo at 8:53 PM on March 5, 2004
640K ought to be enough for anybody
-- Bill Gates circa 1981
Ok Bill Gates has among others said a lot of things that don't make much sense. The article itself doesn't make much sense as well as there is an -heavy- lack of quotations and it only conveys a few details of what was said. Not a good article for a serious discussion I think, but we can use a few points :
1) spam is considered by some as an annoyance
2) spam is considered by some as a damage
3) both usually think the CAUSE of spam is Free Email.
IT IS NOT
// start interesting story for average Joe user
Who are spammers
The "big names" in spamming business were script kiddies who started by exploiting security weakness of some email systems (I will not go in technical details as my target is not techies, but Mr. and Ms. Doe). Good techies are constantly fixing the security weakness and will continue to do so.
Spammers are -NOT- hackers , as real hackers despise the abuse of any system for personal gain, when the abuse is such that the system may be severely damaged or destroyed.
They are -abusers- as spammers abuse the structure that was built to do many things, including delivering email globally (at a cost that is so low it's hard to measure in fractions of cents.)
Why "free" email ?
Much like telephones and faxes (or any other communication device) email becomes very demanded and very interesting ONLY when -a lot- of people use email and that's why email was given for "free" to millions of people : to promote the use of email, to make people appreciate the -fact- you can send messages instantly to Australia or anywhere else and not wait for slow unreliable paper mail to arrive.
Actually, you email isn't free when you pay a subscription to an Internet Service Provider (ISP), but ISP companies know very well that the average user doesn't send hundred of emails a day. Even if you DID send one hundred of emails a day, the cost of production and delivery of email is still very very low and probably well within the price of your moderaltely unexpensive monthly subscription, which includes more services then just email.
Why Spammer Are THE Problem, not unexpensive email
The problem of spammers starts when THEY send dozen of thousands of emails a day : that is still -technically sustainable- but when the numbers start to add to many many millions email an hour the techies can only respond by 1) shutting down the spammer offender email, when they can or 2) start adding more expensive hardware to sustain the email traffic, which starts adding costs.
Some ISP learned the lesson and become wise: they started charging serious money to spammers (spammer is the dude who sends many thousands email an hour if not more, remember) and that is a good measure to shift the costs to the person that -makes- the cost happen, without charging Joe user for the abuse of Spammer. This way you hit ONLY the spammer, making HIS costs higher, NOT yours.
This way ISPs can finance their hardware and wages at expense of the spammer.
Enter "Rogue" ISPs : ISPs specialized in selling email to spammers for a price. A few "rogue" isps are technically sustainable, but when the number of such ISPs increases the traffic of email they start hitting mailbox machines of any other ISPs with predictable result: more money is to be spent for faster and better mailbox machines.
Many technical attempts are being made to stop such ISPs or to make their life miserable and expensive.
From the user point of view, filtering techniques will reduce the impact of the problem, and very good isps already are trying to implement free filters for Joe users and to use filters to contain spammers at well.
// end story
Why the Email Tax is IDIOTIC and Dangerous
To put it simply, the COST of distribution and production of a single piece of email is very close to zero, so close to zero it's hard to measure it.
The problem is NOT email, the problem ARE spammers. One must hit spammers, not Joe the ordinary user, because spammers are overloading the system, not Joe user.
By placing an extra cost (or even worse a State Tax) on EVERY email Joe user will actually PAY the way to pay even more the now
"Rogue" ISPS will happily pay the price of such Tax or Extra Cost and attempts will be made by them to reduce other costs so that, at then end, Mr.Spammer will remain a welcome customer. Even if email becomes as expensive at normal paper mail, it is still much more faster and reliable then regular mail ; guess what, companies will still prefer email to mail and spam will be only marginally reduced.
What if Email Tax/Stamp is introduced ?
Let's imagine that the EMail Tax (or stamp, or whatever) manages to contain and reduce spammers. If the spammers are brought to their knees, then the Email Tax-Stamp should be removed , but guess what: ISPs will happily KEEP it forever as it increases their revenue and they'll keep on saying that they must keep the Tax to contain evil spammers. Some ISP is salivating over the idea of making you pay a lot more for email using the spammer boo boo scare.
Joe user will pay for spammers threat to be used as a scaring argument to keep the Email Tax/Stamp, and will CONTINUE to receive spam, even if less then now.
Why ? Because Email is still better then Mail, so rogue companies pushed by Marketing Idiots will think "hey I'll spam by email EVEN if it costs as normal mail, because it's faster !!! "
What do you prefer, extremely unexpensive email and spam (that can be reduced by free filters anyway) or real expensive email and less spam (not zero spam, forget zero spam there will never be zero spam in your mailbox) ???
It is only YOUR choice and it shouldn't be imposed on others with blanket coverage . If ISPs want to extra-charge for keeping _EVERY_ spam email out of your mailbox that's fine as they're providing an extra-service that you may choose to pay or not to pay for.
But if you can do the filtering for yourself, or if you're not bothered so much by spam, why should you pay for a service you have no interest into ?
posted by elpapacito at 7:16 AM on March 6, 2004
-- Bill Gates circa 1981
Ok Bill Gates has among others said a lot of things that don't make much sense. The article itself doesn't make much sense as well as there is an -heavy- lack of quotations and it only conveys a few details of what was said. Not a good article for a serious discussion I think, but we can use a few points :
1) spam is considered by some as an annoyance
2) spam is considered by some as a damage
3) both usually think the CAUSE of spam is Free Email.
IT IS NOT
// start interesting story for average Joe user
Who are spammers
The "big names" in spamming business were script kiddies who started by exploiting security weakness of some email systems (I will not go in technical details as my target is not techies, but Mr. and Ms. Doe). Good techies are constantly fixing the security weakness and will continue to do so.
Spammers are -NOT- hackers , as real hackers despise the abuse of any system for personal gain, when the abuse is such that the system may be severely damaged or destroyed.
They are -abusers- as spammers abuse the structure that was built to do many things, including delivering email globally (at a cost that is so low it's hard to measure in fractions of cents.)
Why "free" email ?
Much like telephones and faxes (or any other communication device) email becomes very demanded and very interesting ONLY when -a lot- of people use email and that's why email was given for "free" to millions of people : to promote the use of email, to make people appreciate the -fact- you can send messages instantly to Australia or anywhere else and not wait for slow unreliable paper mail to arrive.
Actually, you email isn't free when you pay a subscription to an Internet Service Provider (ISP), but ISP companies know very well that the average user doesn't send hundred of emails a day. Even if you DID send one hundred of emails a day, the cost of production and delivery of email is still very very low and probably well within the price of your moderaltely unexpensive monthly subscription, which includes more services then just email.
Why Spammer Are THE Problem, not unexpensive email
The problem of spammers starts when THEY send dozen of thousands of emails a day : that is still -technically sustainable- but when the numbers start to add to many many millions email an hour the techies can only respond by 1) shutting down the spammer offender email, when they can or 2) start adding more expensive hardware to sustain the email traffic, which starts adding costs.
Some ISP learned the lesson and become wise: they started charging serious money to spammers (spammer is the dude who sends many thousands email an hour if not more, remember) and that is a good measure to shift the costs to the person that -makes- the cost happen, without charging Joe user for the abuse of Spammer. This way you hit ONLY the spammer, making HIS costs higher, NOT yours.
This way ISPs can finance their hardware and wages at expense of the spammer.
Enter "Rogue" ISPs : ISPs specialized in selling email to spammers for a price. A few "rogue" isps are technically sustainable, but when the number of such ISPs increases the traffic of email they start hitting mailbox machines of any other ISPs with predictable result: more money is to be spent for faster and better mailbox machines.
Many technical attempts are being made to stop such ISPs or to make their life miserable and expensive.
From the user point of view, filtering techniques will reduce the impact of the problem, and very good isps already are trying to implement free filters for Joe users and to use filters to contain spammers at well.
// end story
Why the Email Tax is IDIOTIC and Dangerous
To put it simply, the COST of distribution and production of a single piece of email is very close to zero, so close to zero it's hard to measure it.
The problem is NOT email, the problem ARE spammers. One must hit spammers, not Joe the ordinary user, because spammers are overloading the system, not Joe user.
By placing an extra cost (or even worse a State Tax) on EVERY email Joe user will actually PAY the way to pay even more the now
"Rogue" ISPS will happily pay the price of such Tax or Extra Cost and attempts will be made by them to reduce other costs so that, at then end, Mr.Spammer will remain a welcome customer. Even if email becomes as expensive at normal paper mail, it is still much more faster and reliable then regular mail ; guess what, companies will still prefer email to mail and spam will be only marginally reduced.
What if Email Tax/Stamp is introduced ?
Let's imagine that the EMail Tax (or stamp, or whatever) manages to contain and reduce spammers. If the spammers are brought to their knees, then the Email Tax-Stamp should be removed , but guess what: ISPs will happily KEEP it forever as it increases their revenue and they'll keep on saying that they must keep the Tax to contain evil spammers. Some ISP is salivating over the idea of making you pay a lot more for email using the spammer boo boo scare.
Joe user will pay for spammers threat to be used as a scaring argument to keep the Email Tax/Stamp, and will CONTINUE to receive spam, even if less then now.
Why ? Because Email is still better then Mail, so rogue companies pushed by Marketing Idiots will think "hey I'll spam by email EVEN if it costs as normal mail, because it's faster !!! "
What do you prefer, extremely unexpensive email and spam (that can be reduced by free filters anyway) or real expensive email and less spam (not zero spam, forget zero spam there will never be zero spam in your mailbox) ???
It is only YOUR choice and it shouldn't be imposed on others with blanket coverage . If ISPs want to extra-charge for keeping _EVERY_ spam email out of your mailbox that's fine as they're providing an extra-service that you may choose to pay or not to pay for.
But if you can do the filtering for yourself, or if you're not bothered so much by spam, why should you pay for a service you have no interest into ?
posted by elpapacito at 7:16 AM on March 6, 2004
WILL YOUR IDIOTS GET IT INTO YOUR HEADS THAT THEY DO NOT MEAN A $$ COST. THEY MEAN SLOWING DOWN THE SENDING PROCESS BY INTERACTIVE AUTHENTICATION.
Sorry to do that, but the uninformed reactions in this thread are truly nauseating.
posted by tiamat at 9:59 AM on March 6, 2004
Sorry to do that, but the uninformed reactions in this thread are truly nauseating.
posted by tiamat at 9:59 AM on March 6, 2004
-1 : Flamebait
posted by elpapacito at 4:24 PM on March 6, 2004
posted by elpapacito at 4:24 PM on March 6, 2004
Damn, just like Micro$oft to try and charge us for email. Those bastards.
posted by Civil_Disobedient at 5:04 PM on March 6, 2004
posted by Civil_Disobedient at 5:04 PM on March 6, 2004
Tiamat, I am aware of this, and it's still a problem for me as a mailing list host. Unless there is someway for each mailing list post to authenticate once instead of hundreds of times, this would kill a lot of the social/fan/subject mailing lists. I don't see that as being worth the trade-off.
Please note also that the article does refer to paying actual money for e-mail:
posted by litlnemo at 5:16 PM on March 6, 2004
Please note also that the article does refer to paying actual money for e-mail:
"Meanwhile, Goodmail Systems Inc. has been in touch with Yahoo! Inc. and other e-mail providers about using cash. Goodmail envisions charging bulk mailers a penny a message to bypass spam filters and avoid being incorrectly tossed as junk. (...) Goodmail chief executive Richard Gingras said individuals might get to send a limited number for free, while mailing lists and nonprofit organizations might get price breaks."So, uh, calm down. (About this thread, at least. Not about these stupid, stupid proposals.)
posted by litlnemo at 5:16 PM on March 6, 2004
Who is going to be the postal office? who's gonna sell the stamps? MS.. yeah.. big $$ on the horizon..
posted by 12345 at 9:36 AM on March 8, 2004
posted by 12345 at 9:36 AM on March 8, 2004
« Older Toho Tokusatsu Movie Poster Gallery | a savage drunken pinball high on black-tar heroin Newer »
This thread has been archived and is closed to new comments
posted by MrLint at 11:38 AM on March 5, 2004