Worse still is google's apparent reason for googlemail: so they can serve textads in the email...
It seems like Google has shot themselves in the foot recently. Their "adsense" program lets web site operators serve google adwords and receive a portion of the revenue. Some of these site operators have figured out how to spam the latest google algorithm. Now the first page of results for many search terms is little more than links to pages that have a google textad at the top and a bunch of index spamming text below it.
Actually there could be some credibility to this, as much of the spam is porn-related and organized crime is involved in the net porn industry. Apparently its a good way to launder money.
As a postgresql user, one thing I did notice is that postgresql uses the old one-process-per-connection style of daemon whereas mysql uses threads. Postgresql tends to fall apart under high connection concurrency whereas mysql might not due to its use of threads.
Let's say a SMTP replacement is in place and you now know for certain that the spam you just received did in fact originate from throwawayaccount@isp.net . Now what good is that information, since by the time you act on it, the spammer is done with the account?
Your message is kind of bizarre. First, you never discuss what the average death rates are in skydiving. Obviously, when a particular skydiving facility exceeds this rate, an investigation is in order. You only anecdotally discuss what the acceptable failure rate of the shuttle is. Is 1 shuttle down in 18 years within the design specifications? And most importantly, you seem to neglect the fact that whatever caused the shuttle issue must be located and corrected before another one is launched. The issue that caused the Challenger accident was not restricted to that one shuttle. It could've happened to the next flight if NASA just "got back on the horse" as you suggest doing.
maru
Re:XML is NOT just text!
on
XML and Perl
·
· Score: 1
What is the meaning of your sig? I am curious, because I feel that the lack of two-phase commit is one of the major issues preventing the open-source rdmbs servers from competing fully with Oracle. I was curious if your sig was related to this issue.
The issue is that postgresql doesn't yet scale as well as Oracle does. Postgresql doesn't currently support multimaster replication. The core development team is working on an implementation that will be groundbreaking but it probably won't be done for a year or two. Postgresql also launches one process per connection, which really bites. It doesn't prefork either and there is a fair amount of per-process startup overhead.
I don't understand what advantage in fighting spam accurate or trusted reverse-name lookup information would offer. The sender's IP is there as plain as day, what advantage does pansy-fying it do? Since you can't forge IPs for a mail connection, IPs can be used as the authentication for which you speak, you can allow or disallow mail transport based on IP. Everything you suggest can be done already with the current transport mechanisms.
The RBMK reaktor design is a piece of crap. It has a positive void coefficient, which means that coolant failure could lead to a strong increase in power output from the fission process. Worse still, the design has no containment dome. If Three Mile Island didn't have a containment dome there would've been a serious radiation release there as well.
This is going to sound like a joke, but it isn't. One of the primary mutations found in animals was missing anus. The detectable mutations weren't of the kind that would benefit the evolutionary process, they were of the type that would commonly rapidly result in death: absence of one or more extremities, deformation of the skull or spine, absence of eyes, overgrowth of the eyelids, lack of hair, exposed internal organs, or absence of an anus.
The problem is spam coming in, not going out, the serious spammers run their own SMTP servers and the spam is addressed to your domain. It is a complicated issue because large ISPs can receive numerous simultaneous connections from other email servers at other large ISPs, so some of the ideas you suggest as far as limiting concurrency really wouldn't work.
One of the most effective anti-spam resources I have used is the MAPS DUL. It's a list of dialup IP blocks. When your mail server is configured to use the DUL, it rejects mail from any DUL IPs. This is based on the premise that a dialup user should be sending mail through their mail gateway, not directly to your server. This doesn't help the issue with the large spamhausen but it does give a significant reduction in spam without any of the negative aspects of blacklist services.
What are you talking about? I have never seen a piece of spam that contained headers from which it was impossible to determine the spam's origin. Spammers do put in fake headers, but only to fool morons, the real headers are always right in there too. The real problem is that, for the most part, knowing the IP origin of the spammer accomplishes nothing.
What I don't understand is why you can't just say "I didn't write the infringing software". If Ford infringed a patent while making my car, I don't think I would be liable. If a software company infringed on another company's patent while making the streaming system I use, why am I, as a site owner, liable? I have wondered this for quite some time.
How exactly do you protect against an attack whose "payload" is sheer data volume? Make sure your pipe is bigger than the aggregate bandwidth available to every previously compromised host on the internet? How feasible is that? Aside from that, the attack wasn't even against a root server, it was against a DNS provider.
Slashdot Mirror
braniacs=cereal lovers?
Yeah, they've got some braniacs working on it and they are also aggressively recruiting Google's engineers.
Yes, the ads do account for the great majority of google's revenue. The search syndication produces a negligible amount.
Uh, vaporware? MS has been spidering for over six months now and the search interface is at http://beta.search.msn.com . The system definitely exists.
If the GFs are brighter then you probably have a water quality problem.
Worse still is google's apparent reason for googlemail: so they can serve textads in the email...
It seems like Google has shot themselves in the foot recently. Their "adsense" program lets web site operators serve google adwords and receive a portion of the revenue. Some of these site operators have figured out how to spam the latest google algorithm. Now the first page of results for many search terms is little more than links to pages that have a google textad at the top and a bunch of index spamming text below it.
Yeah, not only that... the $5 fish "glow" +significantly+ less than a 99c neon tetra.
Yeah, Walgreens is definitely hurting... sales up 18%, comps up 13%, no debt taken during expansion... looks like bankrupcy any day!!
I have read that most companies that file for chapter 11 do not successfully emerge.
Actually there could be some credibility to this, as much of the spam is porn-related and organized crime is involved in the net porn industry. Apparently its a good way to launder money.
As a postgresql user, one thing I did notice is that postgresql uses the old one-process-per-connection style of daemon whereas mysql uses threads. Postgresql tends to fall apart under high connection concurrency whereas mysql might not due to its use of threads.
maru
Quizilla uses mysql and serves 300,000 pageviews per day.
maru
Let's say a SMTP replacement is in place and you now know for certain that the spam you just received did in fact originate from throwawayaccount@isp.net . Now what good is that information, since by the time you act on it, the spammer is done with the account?
SMTP is clearly not the problem.
maru
Your message is kind of bizarre. First, you never discuss what the average death rates are in skydiving. Obviously, when a particular skydiving facility exceeds this rate, an investigation is in order. You only anecdotally discuss what the acceptable failure rate of the shuttle is. Is 1 shuttle down in 18 years within the design specifications? And most importantly, you seem to neglect the fact that whatever caused the shuttle issue must be located and corrected before another one is launched. The issue that caused the Challenger accident was not restricted to that one shuttle. It could've happened to the next flight if NASA just "got back on the horse" as you suggest doing.
maru
What is the meaning of your sig? I am curious, because I feel that the lack of two-phase commit is one of the major issues preventing the open-source rdmbs servers from competing fully with Oracle. I was curious if your sig was related to this issue.
maru
The issue is that postgresql doesn't yet scale as well as Oracle does. Postgresql doesn't currently support multimaster replication. The core development team is working on an implementation that will be groundbreaking but it probably won't be done for a year or two. Postgresql also launches one process per connection, which really bites. It doesn't prefork either and there is a fair amount of per-process startup overhead.
maru
I don't understand what advantage in fighting spam accurate or trusted reverse-name lookup information would offer. The sender's IP is there as plain as day, what advantage does pansy-fying it do? Since you can't forge IPs for a mail connection, IPs can be used as the authentication for which you speak, you can allow or disallow mail transport based on IP. Everything you suggest can be done already with the current transport mechanisms.
maru
The RBMK reaktor design is a piece of crap. It has a positive void coefficient, which means that coolant failure could lead to a strong increase in power output from the fission process. Worse still, the design has no containment dome. If Three Mile Island didn't have a containment dome there would've been a serious radiation release there as well.
maru
maru
This is going to sound like a joke, but it isn't. One of the primary mutations found in animals was missing anus. The detectable mutations weren't of the kind that would benefit the evolutionary process, they were of the type that would commonly rapidly result in death: absence of one or more extremities, deformation of the skull or spine, absence of eyes, overgrowth of the eyelids, lack of hair, exposed internal organs, or absence of an anus.
maru
The problem is spam coming in, not going out, the serious spammers run their own SMTP servers and the spam is addressed to your domain. It is a complicated issue because large ISPs can receive numerous simultaneous connections from other email servers at other large ISPs, so some of the ideas you suggest as far as limiting concurrency really wouldn't work.
One of the most effective anti-spam resources I have used is the MAPS DUL. It's a list of dialup IP blocks. When your mail server is configured to use the DUL, it rejects mail from any DUL IPs. This is based on the premise that a dialup user should be sending mail through their mail gateway, not directly to your server. This doesn't help the issue with the large spamhausen but it does give a significant reduction in spam without any of the negative aspects of blacklist services.
maru
What are you talking about? I have never seen a piece of spam that contained headers from which it was impossible to determine the spam's origin. Spammers do put in fake headers, but only to fool morons, the real headers are always right in there too. The real problem is that, for the most part, knowing the IP origin of the spammer accomplishes nothing.
maru
I suspect that most hallways are taupe because it's one of the cheapest colors of paint and it hides dirt and stains reasonably well.
maru
What I don't understand is why you can't just say "I didn't write the infringing software". If Ford infringed a patent while making my car, I don't think I would be liable. If a software company infringed on another company's patent while making the streaming system I use, why am I, as a site owner, liable? I have wondered this for quite some time.
maru
When they ask me "Don't you like to save money"? I typically answer "No", it really confuses them.
maru
How exactly do you protect against an attack whose "payload" is sheer data volume? Make sure your pipe is bigger than the aggregate bandwidth available to every previously compromised host on the internet? How feasible is that? Aside from that, the attack wasn't even against a root server, it was against a DNS provider.
maru