Do Linux distributions have the equivalent of Solaris' ufsdump tool (aside from things like dd or cpio)
Yep. It's called 'dump' (that's the original name for dump, but the original was for BSD's FFS (I believe)). The way that dump was written, it requires some knowledge of the filesystem internals, so -- in this case -- Linux dump is only likely to work well with ext{2,3} filesystems (i.e. not, for example, reiserfs).
I can just see it now, Your spouse looks up your GPS data and finds that just about every night you said you were 'working late', you have thes long traces at their best friend's house
This isn't a big shock..MIL web servers are about 75% Microsoft -- and they've had hackers walking all over them. I'm betting that microsoft also charges them full list price, too (It's not a question of them buying a Windows server -- it's a question of which, and how many). For.MIL to move over to Open source would be a big hit.
For Microsoft to try and claim that they're as secure as open source so soon after the 'format your hard drive' bug created such a big fluff is likely to backfire on them. I think that they're shotgunning flies here, and more likely than not to hit their foot. They're actuzlly making it clear that they really do consider the security issue to be a real problem with their software. If these people start to look at the security differences between MS and OS.MIL usage of MS may just drop precipitiously.
I think I'm adhd, as well. One thing that I *LOVE* my palm for is the fact that I can program in the next thing I'm going to do and when I'm going to do it.
When I teach, I program in the times that I'm supposed to have the breaks and end of the course so that I can finish everything on time without leaving loose ends for the students.
Then there's the phone lookup -- real nice, and the 'find' function and I'll often write notes to myself on the bus...
The bigest problem I have with my palm is that I keep on losing or breaking it.
I've (catastrophically) dropped two palms, had another two stolen and had one in my pocket when I waded into a fight (to help break it up).
The fact that I keep buying them might be an indicator as to how much I love what a Palm provides. Of course, I'm buying M100s (nn the presumption that it'll suffer the same fate as it's predecessors), but I'm still buying them.
From the article:Japan lags behind Germany, the United States, China and other nations looking into or using open-source software such as Linux, which can be used and modified for free.
Last time I looked at the stats for various domains.mil (US MIL) was the only one that I saw where MS had an overwhelming dominance for web servers. (big surprise that they've had hundreds of boxes broken into?)..Gov was almost 50/50 and.us was only slightly better.
If they were talking about the US government being behind Japan, I think that they never bothered to check the statistics.
I think that they're expecting Microsoft to nay-say Open Source as much as possible. This means that if - despite having MS on the board - they conclude that OS is still ag ood idea, it'll be hard for MS to say "now give us a kick at the can". MS will have already had their kick at the can, but with knowledgeable people at hand and able to respond to MS FUD as precisely that.
I'd say that this is much better than MS being given an opportunity to respond to the report directly to the politicians and in the absense of of those same OS-knowledgeable experts.
The nastiest part of prel is the regular expressions.
If you've got access to a unix/linux system, man perlre will give you a reasonable introduction to perl regular expressions.
You might also want to download the perl reference manual.. It's designed to be printed double sided, then folded and stapled like a pocketbook. I find it a very usable summary of perl overall. I use it on a regular basis. (or just grab the pdf file and print it
In terms of books, I'd suggest the O'Reilly books (no shock there)
Re: abuse.net Re:Some corrections and arguments.
on
As the Spam Turns
·
· Score: 2
While your efforts are admirable, abuse.net is less than universally used
Abuse.net doesn't rely on the admin of the domain. They just keep a list of abuse addresses explicitly known for various domains. Where a domain doesn't have a known address, it defaults to abuse@the.domain. Anybody can submit an update for a domain if they have information (obviously, authoritative sources are preferred).
For sites with non-english reading admins, there's not a
whole lot more you can do (unless you know their language).
If they're lucky, they may be able to have someone (babelfish?) translate your letter for them. I wouldn't mind
learning mandarin, but I'm not going to learn it just so
that I can do multilingual spam warnings.
Re:Some corrections and arguments.
on
As the Spam Turns
·
· Score: 2
If you don't mind installing perl, I've got one that does 1 and 2. It needs to be fed emails in standard unix mailbox format. it's called peelhead. I put it in the nimhunt directory
It also needs to be configured with the dns names/ip addresses of hosts which handle your email for you.
If you're willing to marry the two pieces (shouldn't be that hard), you could...
You can block emails at the server, once youget the precedence:Spam header. On the other hand, once you've gotten enough spam from a given server, you can send a request to your routers to block that IP address. -- and it's not like the sender can't say that they didn't do any spamming because it's right there in the headers.
Spam blocking makes email unreliable. The way it is implemented is generally broad-brush and affects a lot more than just blocking some spam.
It's spam which makes mail unreliable. In a couple of cases, while skipping what looks like spam, I've missed a real email. If I didn't have a dozen unwanted spams a day to skip over, I'd have gotten those emails.
Yes. spam blocking sometimes will take out an entire C or even B class block. Usually it's done when there are signs that an ISP is allowing a spammer to relocate an email server once it's first address gets blocked for spamming.If you're running into problems like that, then chances are that you're with a spam-friendly ISP. That bodes ill just for starters.
If you're being careful to make sure that you're not doing anything that looks like it might be spam, then I'd suggest that you move to a provider who's not so spam friendly.
If you're doing borderline spam activities, you're not going to get a whole lot of sympathy here...
A lot of people on slashdot have had to waste a lot of time dealing with spam problems. We treat spammers like problems.. not puppies.
I define commercial use far more broadly than just providing web surfing. It includes selling products. In fact, I'm working with a friend who provides does web hosting, and ecommerce (see the tiny add at the bottom of my posting).
We buy adds, run contests, register our site with anybody who's interested in listing us, and I even mention us in my signature...but my postings are (to the best of my ability) intelligent and on-topic). On really good days, I've gotten a good hadfull of posts rated to +5.
Spamming people to sell them garbage, is also a commrecial use. It just happens to also be an action that's not welcome by 99.5 percent of the people on the net (to varying degrees). We have a right to not allow our email boxes used for your commercial purposes.
If it happens to be a joker that gave us a "spamtrap" address, we're blocked.
Spamgrap addresses are, by definition, not given out with permission to send to. If you can prove that someone gave you a spamtrap address, told you that they wanted you to send them email at that address and then complained that you spammed them, then you can sue the person who did that.
Don't bother the person who responded to that information. Sue the person who 'framed' you.
On the other hand, if you're bulk-sending email to addresses that aren't confirmed for sending mail to, then you're just asking to get blocked. Under those circumstances, I'd say you were just asking to get blocked.
Re:Some corrections and arguments.
on
As the Spam Turns
·
· Score: 2
It looks like he is familliar with things like whois, bash perl, etc... Unfortunately, his boss seems unwilling to part with money for a linux/bsd box...
(although a reasonable linux box could be built with about $200 worth of hardware from (your state).forsale.
A decent Linux server for those kinds of light duty purposes doesn't need much... an 8 bit video card, 64M of ram and a 4Gig hard disk should do you just fine.
Re:Some corrections and arguments.
on
As the Spam Turns
·
· Score: 2
Look at the number of systems in those lists. There is simply no feasible way to track down and notify everyone that gets listed.
Not so true. I have a script that I use to automagically find someone to notice. It first looks for a reverse DNS for the site. If that fails, then it does a traceroute and uses the last reverse DNS in that chain.
Then it reduces the DNS address it got to something 'reasonable' and uses that for a message through abuse.net.
It seems to work reasonably well... Unless you're faking (or refusing to set) your reverse DNS and your ISP is doing the same thing, chances are that somebody will get the notification who can at least pass it on to the proper person.
With over 1500 auto-generated reports, the program seemed to have a reasonably high success rate. It was part of a program that I called nimhunt
(I used it to track and report nimda-type virus attacks).
I've also modified it for auto-reporting other attacks as well.
... COMDEX was a dead week for the casinos 'cause techies don't gamble as much as white trash.
Makes sense. Techies are relatively good at numbers, so they know (or can easily calculate) that casinoes are nothing more than a money sink (unless you're good enough at card counting to get kicked out).
We all know that we're more likely to 'score' chasing the waitresses than playing roulette... and for most of us, our chances with the waitresses aren't that good.
had the citizens of Maryland, Virginia etc. been armed the sniper shooting would never had happened.
Right: so they figure that people could have shot a sniper that nobody ever saw before, during or immediately after the shootings?? (not to mention differentiating between good snipers and bad snipers!) And they want more people with their level of intelligence running around with guns?? God, am I glad that I live in Canada.
If anything, those snipings are an argument for gun control (or at least gun registration)
Don't mind me. I've set up this sniper's nest just in case the illegal sniper tries to shoot someone. (I've managed to shoot 4 of them, so far, this week.)
But they owe a king's ransome to pay off the price paid to acquire Comdex at the height of the tech frenzy.
In other words, someone could pay a pittance for the Comdex name, hire off all of their employees and continue to run the show. Comdex isn't likely to die-- just the company that runs it.
Hard drives with a high rotational rate are not necessarily faster at providing data than those with a slow rate. The bottleneck is often the time it takes to move the heads, and the time it takes to present the data to the CPU, not the latency of waiting until the data is under the head.
a 10K RPM drive has about 166 rotations/second -- or 6ms/revolution, giving an average rotational latency just over 3ms
Similarly, a 7200RPM drive has a rotational latency just over 4 ms. This compares to high-end seek times around 3ms. In other words, it's quite possible to be in a situation where it's cheaper to jump tracks to get data than to wait for the disk to rotate to the wanted data on the same track.
A
Rotational speed increases both rotational latency and overall transfer speeds.. (a 15K disk is going to give you about twice the transfer rate of a 7200RPM disk with the same geometry, if all of the data is on one cylinder) The effect is far from trivial compared to seek times. It is, however, pretty much guaranteed to wear out your bearings much faster. I, for one, would be very worried about keeping a 15K hard disk in my desktop for 5 years.
So are you actually saying that we would be better off without driver's licenses as well?
Don't you hate it when people put words into your mouth??
A driver's license does the job that it does well enough. It has also proven to be problematic as a security tool. Widening it to a national ID would increase the exposure of average citizens to abuse of/by the system. It would not, however, have done anything to stop the sept/11 attacks. Nor is it likely to do much to prevent future terror attacks.
People have been quick to forget that the second most serious terror attack on US soil was comitted by a blond, (so called) christian, ex-marine.
If you want to initiate a national ID system, then feel free to explain to me me how it could have stopped the Oklahoma bombing (or, for that matter, the Sept 11 attacks).
Now there's an interesting pattern here: Japan, Germany and Brazil. Conspiracy theories apart, these are countries where three UnitedLinux members (Turbolinux, SuSE and Connectiva) are based. Coincidence?
Chances are that Columbia is further along the road of officially adopting Open Source than any of those three. Is there a distribution coming out of Columbia??
Yes it's good to support a distribution that's based in oyour country, but the nice thing about Linux is that -- because it's OS -- you can provide meaningful support for the OS without being the official source of it. If Canada decided to go RedHat or Debian, for example, I could start up a company to provide support. I could either depend on the original distributors for support, or I could do the support myself.
As an example: a friend of mine is having problems with ftp/xinetd on a redhat installaton. We don't have to wait for RedHat to get around to fixing this problem. I can download the source, insert (or uncomment) debug statements and solve the problem myself.Once it's tracked, I can forward the fix to RedHat to fold into the official release.
No matter where the distribution is sourced, my company doesn't have to wait on what is financially advantageous to RedHat to get the fix. If RedHat isn't willing to do the work, then the work and, (almost as importantly) the payment for the work, can occur locally. Not being entirely dependent on a foreign company for support and being able to keep support payments inside the country are both incentives to any government.
The company that produced it gets sued out of existence for incorporating a known hazardous protein into a food.
Perhaps that's why companies are fighting tooth and nail to prevent marking GM foods -- or even marking foods as GM free. It's really hard to sue someone for GMing an alergen into corn when you can't prove that the corn is GMed.
If someone gets a serious reaction to some food in a restaurant, their knee jerk reaction isn't going to be to sue monsanto for GMing an alergen into the rice. It's going to be to sue the restaurant for not being careful enough with their utensils. Monsanto's just going to sit back and laugh.
Once people realize that the cause of their alergies is a GMed food, they're going to have to fight to prove that the food is GMed to begin with. Laws against labelling food as GMed (or even as GM free) will ensure that the GM companies can play 'prove it' in court for years.
Tobacco companies were able to do it for generations. There's no reason why Monsanto won't be able to do the same thing if it turns out that GM foods have unintended (or simply undocumented) fatal consequences for some consumers.
Does anyone find the irony of this and Pixar producing "A Bug's Life"
If they get together with The guys from Mainframe, I think we might have a hit.
Bob! Bob! We've got
bugs in Mainfraime!
Slow down mike. How many of them are there?
Millions! And some of them all over the diner. -- I can just see the headlines now: "Bugs slash Dot's." It'll be the biggest scandal in miliseconds!
One is the physical dark side... The side that's facing away from the sun (at the moment in question). The other dark side is the classical 'dark side' -- the side that's always facing away from the earth.
During the full moon, both dark sides are the same side. During a new moon, the dark side is actually brightly lit (by the sun) but since none of that light ever really makes it back to earth, it might as well be dark for most earth observers' purposes.
Of course when the near side of the moon is dark, it's lighter than the dark side is when it's dark because it's lit by the reflection of the earth. This means that the dark side of the moon is actually darker (when it's dark) than the 'light' side when it is dark.
This means that the dark side of the moon really is dark when it's dark -- as opposed to the bright side of the moon which is faintly lit when it's dark.
(go ahead.. just try and pass that through an AI parser!)
No, in fact, the corporations collected the same information on their own - the federal agency just duplicated their efforts, at taxpayer expense.
Not necessarily true. Some of that information is not available elsewhere. Much of it was produced as a result of government-funded research. In other words, you've paid taxes to have the resarch done, and the paper printed, but you can only get the results by going to a (often foreign company) and paying them money for the results that you've paid to have produced.
It's especially egregious when it's other government departments or other levels of government that have to pay a foreign company for the results of taxpayer-funded research. The government money that's going to go to foreign companies alone for the results of research is probably going to far exceed any savings from shutting down these sites. That we the general public also have access is simply a pleasant side-effect.
Profits to private companies aren't the only way to measure public good. The oklahoma bombing probably created massive profits for a number of companies -- but there are only a small handfull of people who are happy that it happened.
Some time ago I put together some spam tracking tools:
One is whoois it takes a URl peels off the http: and/path/ then does a whois lookup. If it recognizes a redirection it will follow (i.e. it will do lookups apnic and kornic, etc.)
The other is peelhead. Peelhead goes through a mail spool file and finds the IP of the machine which transmitted email to your MX host. You need to prime it with the hostnames/IPs of your box and macnines which accept email for you (e.g. your ISP and their secondary MX hosts) I found it useful for doing bulk statistics on the sources of spam.
One common use would be:
Slashdot Mirror
Yep. It's called 'dump' (that's the original name for dump, but the original was for BSD's FFS (I believe)). The way that dump was written, it requires some knowledge of the filesystem internals, so -- in this case -- Linux dump is only likely to work well with ext{2,3} filesystems (i.e. not, for example, reiserfs).
(Sorry: Former best friend).
For Microsoft to try and claim that they're as secure as open source so soon after the 'format your hard drive' bug created such a big fluff is likely to backfire on them. I think that they're shotgunning flies here, and more likely than not to hit their foot. They're actuzlly making it clear that they really do consider the security issue to be a real problem with their software. If these people start to look at the security differences between MS and OS.MIL usage of MS may just drop precipitiously.
When I teach, I program in the times that I'm supposed to have the breaks and end of the course so that I can finish everything on time without leaving loose ends for the students.
Then there's the phone lookup -- real nice, and the 'find' function and I'll often write notes to myself on the bus...
The bigest problem I have with my palm is that I keep on losing or breaking it.
I've (catastrophically) dropped two palms, had another two stolen and had one in my pocket when I waded into a fight (to help break it up).
The fact that I keep buying them might be an indicator as to how much I love what a Palm provides. Of course, I'm buying M100s (nn the presumption that it'll suffer the same fate as it's predecessors), but I'm still buying them.
Last time I looked at the stats for various domains .mil (US MIL) was the only one that I saw where MS had an overwhelming dominance for web servers. (big surprise that they've had hundreds of boxes broken into?). .Gov was almost 50/50 and .us was only slightly better.
If they were talking about the US government being behind Japan, I think that they never bothered to check the statistics.
I'd say that this is much better than MS being given an opportunity to respond to the report directly to the politicians and in the absense of of those same OS-knowledgeable experts.
If you've got access to a unix/linux system, man perlre will give you a reasonable introduction to perl regular expressions.
You might also want to download the perl reference manual.. It's designed to be printed double sided, then folded and stapled like a pocketbook. I find it a very usable summary of perl overall. I use it on a regular basis. (or just grab the pdf file and print it
In terms of books, I'd suggest the O'Reilly books (no shock there)
Abuse.net doesn't rely on the admin of the domain. They just keep a list of abuse addresses explicitly known for various domains. Where a domain doesn't have a known address, it defaults to abuse@the.domain. Anybody can submit an update for a domain if they have information (obviously, authoritative sources are preferred).
For sites with non-english reading admins, there's not a whole lot more you can do (unless you know their language). If they're lucky, they may be able to have someone (babelfish?) translate your letter for them. I wouldn't mind learning mandarin, but I'm not going to learn it just so that I can do multilingual spam warnings.
If you're willing to marry the two pieces (shouldn't be that hard), you could...
You can block emails at the server, once youget the precedence:Spam header. On the other hand, once you've gotten enough spam from a given server, you can send a request to your routers to block that IP address. -- and it's not like the sender can't say that they didn't do any spamming because it's right there in the headers.
It's spam which makes mail unreliable. In a couple of cases, while skipping what looks like spam, I've missed a real email. If I didn't have a dozen unwanted spams a day to skip over, I'd have gotten those emails.
Yes. spam blocking sometimes will take out an entire C or even B class block. Usually it's done when there are signs that an ISP is allowing a spammer to relocate an email server once it's first address gets blocked for spamming.If you're running into problems like that, then chances are that you're with a spam-friendly ISP. That bodes ill just for starters.
If you're being careful to make sure that you're not doing anything that looks like it might be spam, then I'd suggest that you move to a provider who's not so spam friendly.
If you're doing borderline spam activities, you're not going to get a whole lot of sympathy here... A lot of people on slashdot have had to waste a lot of time dealing with spam problems. We treat spammers like problems.. not puppies.
We buy adds, run contests, register our site with anybody who's interested in listing us, and I even mention us in my signature...but my postings are (to the best of my ability) intelligent and on-topic). On really good days, I've gotten a good hadfull of posts rated to +5.
Spamming people to sell them garbage, is also a commrecial use. It just happens to also be an action that's not welcome by 99.5 percent of the people on the net (to varying degrees). We have a right to not allow our email boxes used for your commercial purposes.
If it happens to be a joker that gave us a "spamtrap" address, we're blocked.
Spamgrap addresses are, by definition, not given out with permission to send to. If you can prove that someone gave you a spamtrap address, told you that they wanted you to send them email at that address and then complained that you spammed them, then you can sue the person who did that.
Don't bother the person who responded to that information. Sue the person who 'framed' you.
On the other hand, if you're bulk-sending email to addresses that aren't confirmed for sending mail to, then you're just asking to get blocked. Under those circumstances, I'd say you were just asking to get blocked.
A decent Linux server for those kinds of light duty purposes doesn't need much... an 8 bit video card, 64M of ram and a 4Gig hard disk should do you just fine.
Not so true. I have a script that I use to automagically find someone to notice. It first looks for a reverse DNS for the site. If that fails, then it does a traceroute and uses the last reverse DNS in that chain.
Then it reduces the DNS address it got to something 'reasonable' and uses that for a message through abuse.net.
It seems to work reasonably well... Unless you're faking (or refusing to set) your reverse DNS and your ISP is doing the same thing, chances are that somebody will get the notification who can at least pass it on to the proper person.
With over 1500 auto-generated reports, the program seemed to have a reasonably high success rate. It was part of a program that I called nimhunt (I used it to track and report nimda-type virus attacks). I've also modified it for auto-reporting other attacks as well.
Makes sense. Techies are relatively good at numbers, so they know (or can easily calculate) that casinoes are nothing more than a money sink (unless you're good enough at card counting to get kicked out).
We all know that we're more likely to 'score' chasing the waitresses than playing roulette ... and for most of us, our chances with the waitresses aren't that good.
Right: so they figure that people could have shot a sniper that nobody ever saw before, during or immediately after the shootings?? (not to mention differentiating between good snipers and bad snipers!) And they want more people with their level of intelligence running around with guns?? God, am I glad that I live in Canada.
If anything, those snipings are an argument for gun control (or at least gun registration)
In other words, someone could pay a pittance for the Comdex name, hire off all of their employees and continue to run the show. Comdex isn't likely to die-- just the company that runs it.
a 10K RPM drive has about 166 rotations/second -- or 6ms/revolution, giving an average rotational latency just over 3ms
Similarly, a 7200RPM drive has a rotational latency just over 4 ms. This compares to high-end seek times around 3ms. In other words, it's quite possible to be in a situation where it's cheaper to jump tracks to get data than to wait for the disk to rotate to the wanted data on the same track.
A Rotational speed increases both rotational latency and overall transfer speeds.. (a 15K disk is going to give you about twice the transfer rate of a 7200RPM disk with the same geometry, if all of the data is on one cylinder) The effect is far from trivial compared to seek times. It is, however, pretty much guaranteed to wear out your bearings much faster. I, for one, would be very worried about keeping a 15K hard disk in my desktop for 5 years.
Don't you hate it when people put words into your mouth??
A driver's license does the job that it does well enough. It has also proven to be problematic as a security tool. Widening it to a national ID would increase the exposure of average citizens to abuse of/by the system. It would not, however, have done anything to stop the sept/11 attacks. Nor is it likely to do much to prevent future terror attacks.
People have been quick to forget that the second most serious terror attack on US soil was comitted by a blond, (so called) christian, ex-marine.
If you want to initiate a national ID system, then feel free to explain to me me how it could have stopped the Oklahoma bombing (or, for that matter, the Sept 11 attacks).
Chances are that Columbia is further along the road of officially adopting Open Source than any of those three. Is there a distribution coming out of Columbia??
Yes it's good to support a distribution that's based in oyour country, but the nice thing about Linux is that -- because it's OS -- you can provide meaningful support for the OS without being the official source of it. If Canada decided to go RedHat or Debian, for example, I could start up a company to provide support. I could either depend on the original distributors for support, or I could do the support myself.
As an example: a friend of mine is having problems with ftp/xinetd on a redhat installaton. We don't have to wait for RedHat to get around to fixing this problem. I can download the source, insert (or uncomment) debug statements and solve the problem myself.Once it's tracked, I can forward the fix to RedHat to fold into the official release.
No matter where the distribution is sourced, my company doesn't have to wait on what is financially advantageous to RedHat to get the fix. If RedHat isn't willing to do the work, then the work and, (almost as importantly) the payment for the work, can occur locally.
Not being entirely dependent on a foreign company for support and being able to keep support payments inside the country are both incentives to any government.
Perhaps that's why companies are fighting tooth and nail to prevent marking GM foods -- or even marking foods as GM free. It's really hard to sue someone for GMing an alergen into corn when you can't prove that the corn is GMed.
If someone gets a serious reaction to some food in a restaurant, their knee jerk reaction isn't going to be to sue monsanto for GMing an alergen into the rice. It's going to be to sue the restaurant for not being careful enough with their utensils. Monsanto's just going to sit back and laugh.
Once people realize that the cause of their alergies is a GMed food, they're going to have to fight to prove that the food is GMed to begin with. Laws against labelling food as GMed (or even as GM free) will ensure that the GM companies can play 'prove it' in court for years.
Tobacco companies were able to do it for generations. There's no reason why Monsanto won't be able to do the same thing if it turns out that GM foods have unintended (or simply undocumented) fatal consequences for some consumers.
If they get together with The guys from Mainframe, I think we might have a hit.
One is the physical dark side... The side that's facing away from the sun (at the moment in question). The other dark side is the classical 'dark side' -- the side that's always facing away from the earth.
During the full moon, both dark sides are the same side. During a new moon, the dark side is actually brightly lit (by the sun) but since none of that light ever really makes it back to earth, it might as well be dark for most earth observers' purposes.
Of course when the near side of the moon is dark, it's lighter than the dark side is when it's dark because it's lit by the reflection of the earth. This means that the dark side of the moon is actually darker (when it's dark) than the 'light' side when it is dark.
This means that the dark side of the moon really is dark when it's dark -- as opposed to the bright side of the moon which is faintly lit when it's dark.
(go ahead.. just try and pass that through an AI parser!)
Not necessarily true. Some of that information is not available elsewhere. Much of it was produced as a result of government-funded research. In other words, you've paid taxes to have the resarch done, and the paper printed, but you can only get the results by going to a (often foreign company) and paying them money for the results that you've paid to have produced.
It's especially egregious when it's other government departments or other levels of government that have to pay a foreign company for the results of taxpayer-funded research. The government money that's going to go to foreign companies alone for the results of research is probably going to far exceed any savings from shutting down these sites. That we the general public also have access is simply a pleasant side-effect.
Profits to private companies aren't the only way to measure public good. The oklahoma bombing probably created massive profits for a number of companies -- but there are only a small handfull of people who are happy that it happened.
The other is peelhead. Peelhead goes through a mail spool file and finds the IP of the machine which transmitted email to your MX host. You need to prime it with the hostnames/IPs of your box and macnines which accept email for you (e.g. your ISP and their secondary MX hosts) I found it useful for doing bulk statistics on the sources of spam. One common use would be:
or:would give you a list of your top-20 spam sources