If somebody were to purchase the company, they'd acquire its liabilities, too. Pending lawsuits with Novell, IBM, RedHat, AutoZone, as well as many more potential ones. These claims and counterclaims don't just "go away" if the company changes hands.
600 spam mails per day, versus approximayely 30 real e-mails [...] 1 false positive every few weeks (less than 1 in 10.000)
30 emails per day times a few weeks (say 3 weeks) is 630 real emails. That's a false positive rate of 1/630 or 0.16%, if you have really noticed all the false positives. Not 1 in 10,000.
Bogofilter works great. Or SpamAssassin but only if you force-feed it its own judgements. In both cases you have to correct classification errors.
Fidelis Assis (who has now gone solo after having participated in the CRM114 project) shows great results for his recent solo effort: OSBF-lua
Bratko's PPM spam filter -- the one that did great at TREC -- is not yet packaged as a drop-in filter. Same for my DMC spam filter.
The actual TREC 2005 tests referred to in TFA are here.
Personally, I can't accept unpredictable delays in my email, so I have opted out of greylisting. Also greylisting has a non-zero and very hard-to-measure false positive rate.
I assume the paper that you are describing is the 2004 study. The paper described in the talk (which was given 6 months ago or so) described results of the TREC 2005 Spam Track which took place in November 2005. It included a test SpamAssassin 3.x, not 2.3.
While it is reasonable to conjecture that spam has changed so as to defeat spam filtering techniques, or will change so as to defeat the PPM technique that did well at TREC, the historical evidence does not support this conjecture. In particular:
The spam filters tested in 2004 give pretty well exactly the same performance on 2005 and 2006 data.
New versions of the filters are a little bit better, but not by leaps and bounds, and also get about the same results over the last 2.5 years of data.
There is no evidence that "Bayesian poisining" is a viable technique for defeating statistical spam filters in anything but a very artifical laboratory environment where the poisoner has access to the recipient's inbox
The subject of the paper -- and the talk -- is primarily about testing methodology and the need for controlled scientific investigation. So I hesitate to endorse the simplistic notion of a "winner" of the TREC evaluation. However the technique that did very well was indeed quite novel, so here's a characterization.
Andrej Bratko used PPM -- a well-known data compression technique to compress ham and spam separately. Well actually he didn't compress them but just build the statistical model necessary to compress them. Then he simply (tentatively) added the unknown message to each model and chose the one that compressed it best. The general technique of using compression has been mentioned here and elsewhere but Bratko used a much stronger compression scheme and was somewhat clever about it.
I later reproduced Bratko's results using DMC -- a compression schem that I invented 20 years ago -- and got some interesting results. We have a journal article in press describing it and also an evaluation paper at CEAS 2006.
60 mph is about 100 km/h, or 27m/s. G is 9.8 m/s. So 0-60 in 3 sec is about 1G. Street tires don't do that.
But the 0-60 in 3 is McLuhanesque's fantasy. What the article says is 0-60-0 in "about 9 seconds." So let's say that "about 9" is really 10. 0-60 in 6 secs is about half-a-G, which is strong acceleration, and 60-0 in the remaining 4 secs is still most-of-a-G.
Why inflate damned good performance into a preposterous claim?
Is there some fundamental law of nature which states that two people cannot communicate over a distance without sound or visual cues?
Of course not. Smell, sense of motion, heat, touch work just fine.
But you mean, "is there evidence that some phenomenon for which there is no evidence could not exist?" Of course not. But that's not the point. There are an infinite number of preposterous claims that "cannot be disproved." The grass in my yard might be pink tomorrow morning. Or gravity might repel. Your fallacy is in assuming that "cannot be disproved" is any sort of evidence of their truth.
So-called on-line universities are in general not accredited and have no standards at all. They are frauds. All you do is send them a few thousand dollars and they send you a piece of paper (or maybe a pdf that you have to print yourself) that says "Degree" on top of it.
For years, some universities have offered correspondence courses by mail. A few let you do an entire degree by correspondence. Typically those universities have some mechanism for proctored examinations and they make you do a full course of assignments. Typically the degrees offereed are generalist arts degrees.
There is no reason why correspondence courses must use snail mail for delivery of materials. On-line is a fine replacement. There is no reason why a university could not offer such degrees exclusively; however, it would be difficult in my opinion to maintain a high level of scholarship without a base established from real-live human interaction.
While I think it is unlikely that there are any decent specialist degrees offered by correspondence (on-line or snail mail), such programs should not be discounted out of hand. But neither should the mode of delivery be completely ignored. As is mentioned elsewhere, interaction with peers and mentors is an important part of education.
Bottom line: ignore non-accredited; treat on-line the same as any other "pass" degree -- evidence of the candidate's ability to complete something, not any particular expertise.
Quirks and Quarks had an episode on human hibernation discussing the known mechanisms and methods within the realm of immediate possibility. It is well established that cold-water near-drowning victims have survived several hours without oxygen. From an ethical point of view the first human subjects would have to be "last hope" interventions, where death would be inevitable if hibernation were not induced.
What on earth is "weekly percent of market share" anyway? The Slashdot article misparaphrases this as percentage of hits, which is nonsense.
It appears to me that they have selected commercial sites according to category, and then tracked the number of people that visit these sites in a week. Perhaps -- and I'm only guessing -- it means that about 5% of those surveyed visited myspace.com in a given week. That's a far cry from saying that 5% of all accesses were to Myspace.
The bottom line is that the statistic is completely meaningless unless defined. Without a definition there is no point in quibbling about how precisely or accurately it is being measured.
The article seems to indicate that for some fleeting moments there were some inaccuracies in the Ken Lay biography. I read it a couple of days ago and I just read it now. I didn't see any such inaccuracies.
Contested articles are flagged as "under dispute." That, plus a tiny bit of savvy, is all that is necessary to avail one's-self of the vast wealth of information available at Wikipedia.
No I haven't. Unless you think I can't tell what's below from correspondence from somebody I know.
--
Hello .
I think we had correspondence a long time ago if it was not you I am sorry. If it was I could not answer you because my Mozilla mail manager was down for a long time and I could not fix it only with my friend's help I got the emails address out for me..:) I hope it was whom we were corresponded with you are still interested, as I am, though I realize much time has passed since then... I really don't know where to start.... Maybe you could tell me a little about yourself since I lost our early letters, your appearance,age , hobbies, and are you still in the search? If it was you I wrote to and you are interested to get to know me better, I have a profile at : http://www.im-waiting-4you.net/
Don't really know what else to say for now I hope this is the right address
Let me know if you are interested, And I hope you won't run when you see my picture:-)
More specifically, it correctly classifies 84% of spam and 98% of non-spam.
The authors used the SpamAssassin corpus. Holden shows that, on the Spamassasin corpus, Bogofilter correctly classifies 90.3% of spam and 99.88% of non-spam. See http://sam.holden.id.au/writings/spam2/
This approach is nowhere near state of the art.
Chronological order is simple and powerful
on
What's In Your Inbox?
·
· Score: 2, Insightful
I am unconvinced that automatic sorting/threading methods are superior to chronological order for the vast majority of email users. Sure, I may want to be able to bring up "views" of my email that group it by thread or topic or whatever. And it is great idea to be able to tie in related data like bookmarks and files (chronology here is also a powerful search key).
But I think that only a very small number of people would trust an automated tool to determinine the order in which they see messages when they first arrive.
I rarely receive more than 100 messages in a day and I think that's at the high end of typical. It is not very onerous to look at the subject lines of these 100 messages, to triage them and decide if and when to read and respond to them. Maybe some very simple interface (click to remind me to deal with this later) would help but I'm pretty sure that anything more sophisticated than that would be way too intrusive. It would interfere rather than aid in my seeing the overall picture of my incoming email. I definitely don't want my email disappearing into some deep structure that I have to navigate in order to find it.
No, I'm implying (actually saying flat-out) that spam filter vendors lie to the press, and the press (including Slashdot) eat it up like candy. This isn't the first and won't be the last "product placement" Slashdot article.
I like the unsubstantiated side-bar claiming that porn spam had a 5.6% click-through rate. I suppose that's 5.6% of delivered spam? How could this possibly be measured? I'm not saying it can't be approximated but a sidebar and a couple of quotes are hardly sufficient evidence.
FIA have been trying to handicap the teams -- in particular the winning teams -- for years. They have required barge boards on the cars' bellies, smaller tires, grooved tires, inferior brakes, smaller engines.
They have also had difficulty stating and enforcing meaningful restrictions on driver aids like automatic transmissions, traction control and anti-lock brakes.
There's nothing new about FIA mandating a lame standardized component, ostensibly to level the playing field.
The same article mentions Michelin being squeezed out of F1 (i.e. Bridgestone being the only supplier). Competition provides too much incentive for improvement, and, I daresay, too little cash flow from suppliers to FIA.
I don't believe in science to the point that it really reflects more of nature than what our puny imagination
and little minds are capable of working with.
Are you saying that our imagination is just as good as science at predicting the effect of CO2? Qualifies as dont' believe in science in my book.
I don't believe in models clearly built to serve socio-economic interests.
State strawman. Repeat until believed. You dismiss all peer review while providing no evidence that any is so influenced.
Cast as many aspersions as you like on the Kyoto crowd. Unless they are the most powerful conspiracy that has ever existed -- even more powerful than the Catholic Church in its heydey -- there must surely be scientific venues not under its control that would be prepared to review and publish valid contrary evidence. So where is this evidence?
Dismissing the entire world's scientific evidence as the result of a conspiracy is just another way of saying you don't believe in science.
Apparently you don't believe in logic much, either: I will gleefully ignore all of your peer reviewed papers when Deutsche and Dresdner Bank
are doing the funding. This is a strawman fallacy. Can you point to any of the scientific literature funded by these banks? And even if you can, is the conclusion any different if you discount these sources? Can you point to any scientific literature that supports your implication that man's production of CO2 is not responsible for climate change?
In contrast to debate as to man's impact on climate, there is much legitimate debate as to how to mitigate this impact, as it involves social, economic and politicial issues in addition to scientific ones. There is ample evidence that "doing nothing" is not a solution.
As I understand it, it wouldn't be technically all that hard to "turn down the volume" enough to avoid runaway climate change -- so long as we did it before the change cascades out of control. Just stopping the burning of forests and a minor change in the vehicles we drive and/or how much we drive would go a long way. I say "just" but of course effecting even a small socio-economic-political change is a large undertaking.
Maybe some of these so-called "scientists" involved with Global Warming will get a wake-up call!
The so-called scientists involved in the the oil-and-gas disinformation campaign are indeed guilty of fraud. Only now are the scientists involved in the Tobacco disinformation campaign starting to be held to account. Let us hope that in this situation the wheels of justice move quicker.
If you think there is any legitimate question as to the relationship between CO2 and climate change, or as to humankind's contribution to CO2, or as to the acuteness of the problem, please cite reviewed scientific articles.
If you don't believe in science, say so. But don't dismiss the scientific evidence as "voodoo science" without offering some of what you consider "real science." Testimonials are not science, regardless of who makes them.
For those who bat about the words "proof" and "theory" I suggest you determine what they mean in a technical context. Americans are perfectly happy to kill their own citizens with a standard of proof known as "beyond reasonable doubt." And to kill the citizens of other countries on mere suspicion. There is no reasonable doubt about the relationships stated above. In terms of the popular legal use of the word proof, the case is proven.
There is some uncertainty as to how fast the climate change may cascade out of control. Just like when I turn up the volume control on a PA system, I know that I will eventually get horrible screeching feedback but I can't point to the exact position on the dial where it will occur. This does not make feedback "a theory" or "uncertain" at all. In the case of global there is strong evidence that this feedback has started to occur; we just don't know when it will be outright out of control.
He simple [sic] said that according to this study about 1 out of 14 drivers will be in an accident within the same time as the study.
I saw no such conclusion in article describing the study.
I played Super Mario Brothers the other day. I suppose that means that in the interval I spent playing the game I'm likely to be killed by a donkey with a barrel.
Now I don't mean to say that this study is so far from reality as a video game, but nobody -- certainly not its authors -- claim that it in any way predicts the accident rate of these drivers.
Slashdot Mirror
If somebody were to purchase the company, they'd acquire its liabilities, too. Pending lawsuits with Novell, IBM, RedHat, AutoZone, as well as many more potential ones. These claims and counterclaims don't just "go away" if the company changes hands.
Liquidation is the only solution.
Here are the slides from the 400MB video presentation.
Fidelis Assis (who has now gone solo after having participated in the CRM114 project) shows great results for his recent solo effort: OSBF-lua Bratko's PPM spam filter -- the one that did great at TREC -- is not yet packaged as a drop-in filter. Same for my DMC spam filter.
The actual TREC 2005 tests referred to in TFA are here.
Personally, I can't accept unpredictable delays in my email, so I have opted out of greylisting. Also greylisting has a non-zero and very hard-to-measure false positive rate.
TREC 2006 evaluations are now underway.
While it is reasonable to conjecture that spam has changed so as to defeat spam filtering techniques, or will change so as to defeat the PPM technique that did well at TREC, the historical evidence does not support this conjecture. In particular:
- The spam filters tested in 2004 give pretty well exactly the same performance on 2005 and 2006 data.
- New versions of the filters are a little bit better, but not by leaps and bounds, and also get about the same results over the last 2.5 years of data.
- There is no evidence that "Bayesian poisining" is a viable technique for defeating statistical spam filters in anything but a very artifical laboratory environment where the poisoner has access to the recipient's inbox
The subject of the paper -- and the talk -- is primarily about testing methodology and the need for controlled scientific investigation. So I hesitate to endorse the simplistic notion of a "winner" of the TREC evaluation. However the technique that did very well was indeed quite novel, so here's a characterization.60 mph is about 100 km/h, or 27m/s. G is 9.8 m/s. So 0-60 in 3 sec is about 1G. Street tires don't do that.
But the 0-60 in 3 is McLuhanesque's fantasy. What the article says is 0-60-0 in "about 9 seconds." So let's say that "about 9" is really 10. 0-60 in 6 secs is about half-a-G, which is strong acceleration, and 60-0 in the remaining 4 secs is still most-of-a-G.
Why inflate damned good performance into a preposterous claim?
Is there some fundamental law of nature which states that two people cannot communicate over a distance without sound or visual cues?
Of course not. Smell, sense of motion, heat, touch work just fine.
But you mean, "is there evidence that some phenomenon for which there is no evidence could not exist?" Of course not. But that's not the point. There are an infinite number of preposterous claims that "cannot be disproved." The grass in my yard might be pink tomorrow morning. Or gravity might repel. Your fallacy is in assuming that "cannot be disproved" is any sort of evidence of their truth.
The responses to this message are way off track.
So-called on-line universities are in general not accredited and have no standards at all. They are frauds. All you do is send them a few thousand dollars and they send you a piece of paper (or maybe a pdf that you have to print yourself) that says "Degree" on top of it.
For years, some universities have offered correspondence courses by mail. A few let you do an entire degree by correspondence. Typically those universities have some mechanism for proctored examinations and they make you do a full course of assignments. Typically the degrees offereed are generalist arts degrees.
There is no reason why correspondence courses must use snail mail for delivery of materials. On-line is a fine replacement. There is no reason why a university could not offer such degrees exclusively; however, it would be difficult in my opinion to maintain a high level of scholarship without a base established from real-live human interaction.
While I think it is unlikely that there are any decent specialist degrees offered by correspondence (on-line or snail mail), such programs should not be discounted out of hand. But neither should the mode of delivery be completely ignored. As is mentioned elsewhere, interaction with peers and mentors is an important part of education.
Bottom line: ignore non-accredited; treat on-line the same as any other "pass" degree -- evidence of the candidate's ability to complete something, not any particular expertise.
Quirks and Quarks had an episode on human hibernation discussing the known mechanisms and methods within the realm of immediate possibility. It is well established that cold-water near-drowning victims have survived several hours without oxygen. From an ethical point of view the first human subjects would have to be "last hope" interventions, where death would be inevitable if hibernation were not induced.
What on earth is "weekly percent of market share" anyway? The Slashdot article misparaphrases this as percentage of hits, which is nonsense.
It appears to me that they have selected commercial sites according to category, and then tracked the number of people that visit these sites in a week. Perhaps -- and I'm only guessing -- it means that about 5% of those surveyed visited myspace.com in a given week. That's a far cry from saying that 5% of all accesses were to Myspace.
The bottom line is that the statistic is completely meaningless unless defined. Without a definition there is no point in quibbling about how precisely or accurately it is being measured.
The article seems to indicate that for some fleeting moments there were some inaccuracies in the Ken Lay biography. I read it a couple of days ago and I just read it now. I didn't see any such inaccuracies.
Contested articles are flagged as "under dispute." That, plus a tiny bit of savvy, is all that is necessary to avail one's-self of the vast wealth of information available at Wikipedia.
No I haven't. Unless you think I can't tell what's below from correspondence from somebody I know.
..:) ....
:-)
--
Hello .
I think we had correspondence a long time ago if it was not you I am sorry.
If it was I could not answer you because my Mozilla mail manager was down for a
long time and I could not fix it only with my friend's help I got the emails
address out for me
I hope it was whom we were corresponded with you are still interested, as I am,
though I realize much time has passed since then...
I really don't know where to start
Maybe you could tell me a little about yourself since I lost our early letters,
your appearance,age , hobbies, and are you still in the search?
If it was you I wrote to and you are interested to get to know me better, I have
a profile at :
http://www.im-waiting-4you.net/
Don't really know what else to say for now I hope this is the right address
Let me know if you are interested, And I hope
you won't run when you see my picture
talk to you soon.....
Galinka
More specifically, it correctly classifies 84% of spam and 98% of non-spam.
The authors used the SpamAssassin corpus. Holden shows that, on the Spamassasin corpus, Bogofilter correctly classifies 90.3% of spam and 99.88% of non-spam. See http://sam.holden.id.au/writings/spam2/
This approach is nowhere near state of the art.
I am unconvinced that automatic sorting/threading methods are superior to chronological order for the vast majority of email users. Sure, I may want to be able to bring up "views" of my email that group it by thread or topic or whatever. And it is great idea to be able to tie in related data like bookmarks and files (chronology here is also a powerful search key).
But I think that only a very small number of people would trust an automated tool to determinine the order in which they see messages when they first arrive.
I rarely receive more than 100 messages in a day and I think that's at the high end of typical. It is not very onerous to look at the subject lines of these 100 messages, to triage them and decide if and when to read and respond to them. Maybe some very simple interface (click to remind me to deal with this later) would help but I'm pretty sure that anything more sophisticated than that would be way too intrusive. It would interfere rather than aid in my seeing the overall picture of my incoming email. I definitely don't want my email disappearing into some deep structure that I have to navigate in order to find it.
No, I'm implying (actually saying flat-out) that spam filter vendors lie to the press, and the press (including Slashdot) eat it up like candy. This isn't the first and won't be the last "product placement" Slashdot article.
I like the unsubstantiated side-bar claiming that porn spam had a 5.6% click-through rate. I suppose that's 5.6% of delivered spam? How could this possibly be measured? I'm not saying it can't be approximated but a sidebar and a couple of quotes are hardly sufficient evidence.
FIA have been trying to handicap the teams -- in particular the winning teams -- for years. They have required barge boards on the cars' bellies, smaller tires, grooved tires, inferior brakes, smaller engines.
They have also had difficulty stating and enforcing meaningful restrictions on driver aids like automatic transmissions, traction control and anti-lock brakes.
There's nothing new about FIA mandating a lame standardized component, ostensibly to level the playing field.
The same article mentions Michelin being squeezed out of F1 (i.e. Bridgestone being the only supplier). Competition provides too much incentive for improvement, and, I daresay, too little cash flow from suppliers to FIA.
See Monty Python's The Argument Sketch .
Apparently one of us is in the wrong room.
Are you saying that our imagination is just as good as science at predicting the effect of CO2? Qualifies as dont' believe in science in my book.
I don't believe in models clearly built to serve socio-economic interests.
State strawman. Repeat until believed. You dismiss all peer review while providing no evidence that any is so influenced.
Cast as many aspersions as you like on the Kyoto crowd. Unless they are the most powerful conspiracy that has ever existed -- even more powerful than the Catholic Church in its heydey -- there must surely be scientific venues not under its control that would be prepared to review and publish valid contrary evidence. So where is this evidence?
Apparently you don't believe in logic much, either: I will gleefully ignore all of your peer reviewed papers when Deutsche and Dresdner Bank are doing the funding. This is a strawman fallacy. Can you point to any of the scientific literature funded by these banks? And even if you can, is the conclusion any different if you discount these sources? Can you point to any scientific literature that supports your implication that man's production of CO2 is not responsible for climate change?
In contrast to debate as to man's impact on climate, there is much legitimate debate as to how to mitigate this impact, as it involves social, economic and politicial issues in addition to scientific ones. There is ample evidence that "doing nothing" is not a solution.
As I understand it, it wouldn't be technically all that hard to "turn down the volume" enough to avoid runaway climate change -- so long as we did it before the change cascades out of control. Just stopping the burning of forests and a minor change in the vehicles we drive and/or how much we drive would go a long way. I say "just" but of course effecting even a small socio-economic-political change is a large undertaking.
Maybe some of these so-called "scientists" involved with Global Warming will get a wake-up call!
The so-called scientists involved in the the oil-and-gas disinformation campaign are indeed guilty of fraud. Only now are the scientists involved in the Tobacco disinformation campaign starting to be held to account. Let us hope that in this situation the wheels of justice move quicker.
If you think there is any legitimate question as to the relationship between CO2 and climate change, or as to humankind's contribution to CO2, or as to the acuteness of the problem, please cite reviewed scientific articles.
If you don't believe in science, say so. But don't dismiss the scientific evidence as "voodoo science" without offering some of what you consider "real science." Testimonials are not science, regardless of who makes them.
For those who bat about the words "proof" and "theory" I suggest you determine what they mean in a technical context. Americans are perfectly happy to kill their own citizens with a standard of proof known as "beyond reasonable doubt." And to kill the citizens of other countries on mere suspicion. There is no reasonable doubt about the relationships stated above. In terms of the popular legal use of the word proof, the case is proven.
There is some uncertainty as to how fast the climate change may cascade out of control. Just like when I turn up the volume control on a PA system, I know that I will eventually get horrible screeching feedback but I can't point to the exact position on the dial where it will occur. This does not make feedback "a theory" or "uncertain" at all. In the case of global there is strong evidence that this feedback has started to occur; we just don't know when it will be outright out of control.
Can you supply us some details? [Not that I think such details would've helped RIM, whose appeal had been denied.]
He simple [sic] said that according to this study about 1 out of 14 drivers will be in an accident within the same time as the study.
I saw no such conclusion in article describing the study.
I played Super Mario Brothers the other day. I suppose that means that in the interval I spent playing the game I'm likely to be killed by a donkey with a barrel.
Now I don't mean to say that this study is so far from reality as a video game, but nobody -- certainly not its authors -- claim that it in any way predicts the accident rate of these drivers.