Taguchi's objective is robust design, which means building a product, system, or process that works well even in the presence of degrading influences. That means products that deliver value without breaking and services that are enduring while being as simple as possible. Taguchi first determines the control factors that go into designing a product, their interdependencies, then generates an orthogonal array specifying the number of experiments required to find the optimal solution. [....]
That's when Kowalick turned the Taguchi Method to advertising, with the goal of significantly raising the response rate for ad campaigns. [....] The control factors included graphics, colors, and use of humor. The experiments themselves were 12 mailings to 625 addresses each -- two mailings per day over six days. [....]
The vanilla wafer recipe, however, will remain a secret.
------
So.... First of all, is it an advertisement behind an advertisement? Since many of us already has e-mail filter, how will this be not spotted by the filter? Hmm... I smell something really fishy.
We the corporations of the United States, in order to form a more perfect Profit, establish monopoly, insure domestic compliance, provide for the common interest, promote our welfare, and secure the blessings of liberty to ourselves and our posterity, do ordain and establish this Constitution for the United States of America.
Uh... When I heard the term "restricted music", I have the mental image of "Parental Advisory: Explicit Lyrics" or something like that. Maybe "copyrighted music" is a better term?
What is it and why should I care? The article doesnt explain it very well
Oh really? I suggest you read more carefully at the bottom of the page:
...Global Crossing, which operates a high-speed fiber optic network in 27 countries and went bankrupt in January 2002 after amassing $12.4 billion in debt amid sagging demand.
Now I hope you can see why Dept of Homeland Security have worries on securities thing.
It's true that some flaws in the OS are inherently design-based. However, even if we make certain design requirements to be incorporated in the OS, it still doesn't guarantee that the OS is secure. I would think that it even can't minimize the number of OS breaches. It would even hamper the OS development in order to comply with their standards.
About the quote regarding the "minimum expectations of consumers for security and general reliability by establishing a floor for these characteristics". I don't think it would be possible the goal of "the least restrictive requirement while not relenting the control" is vague. Unless it provides rigid post- or pre-conditions of each method (in first order logic if necessary) and provide each formal specifications unambiguously, I would still see some leaks here and there. And, guess what? They put the requirement like UML standards: Way to vague. Congratulations.
For those of you who are curious, click here for the draft.
There IS visuals... See the paper here (see page 5). It's a 9 MB PDF file. That paper also mentions that the fungus DID impact the error correcting part of the CD. And it mentions on page 7:
To begin with, each computation was started by placing one disc into a stand-alone CD or DVD player and pressing play. In all cases so far, placing treated discs into computer-based, CD, CD-RW and DVD players has caused the computer or laptop to crash. For this reason, this study focused only on CD and DVD players that are modular. In most cases the disc was recognized and begain playing. If the player failed to recognize the disc, it was necessary tor emove some cell material using soft cloth [...] Note that cells were completely removed from the innermost region of data at the beginning of the disc, since this is header information necessary for the player to mount the disc.
Then it goes on the snapshot of the "treated" DVD playback on figure 6 at page 9.
Not only that... If the fermented Linux CD is thrown into the dirt, the fungus would cause the dirt to form compost. Thus, SCO would sue compost maker too. Including us, who contribute it through septic tanks...:-)
Know that there are many ways to do one thing and there are pros and contras in each of them. In this case, it seems that NVidia's is not chosen and the way DX9 handles things undermines NVidia's method. It's not necessarily because NVidia sucks. Remember that there are politic struggles among Microsoft, NVidia, and ATI during the inception of DX9? I think NVidia now falls victim of it.
Re:64-benchmarks wont be good
on
AMD64 Preview
·
· Score: 2
If you read the article, the comparison is against Dual P4 Xeon. Some of the tests didn't enable any hyper-threading stuff (and thus it didn't take any advantage of the dualies. Opteron beat P4 by very high margin. Except for content creation & general usage stuff where the P4 wins. But take that with a grain of salt.
64-bit tests won't be fair to either side. It's like comparing apples to oranges. For me, I'm looking forward to see vis-a-vis comparison on programs that is optimized on either platform. For example: A program that is optimized on Itanium and Opteron and see how they fare.
Eh? This is different than I would expect. In the "formal" computer science definition would be something like generating code (often times machine code or byte code) in compiler context.
Also, in computer software design, it has another definition. Namely: Generating real code (C, C++, Java, whatever) from design templates (like from UML).
Now, these people are overloading the same term?? These make me even more confused.
Assume an elephant weighs about six tons, she says, that would mean that water inside a typical cumulous cloud would weigh about one hundred elephants.
NES processor (it's 6502, IIRC) is about 1.79 MHz and computer nowadays are thousand times faster than that. I would appreciate it if you can run your VB emulator on 386 computers and still have a decent speed.... I know that some of the older NES emulators like Nesticle still manage to run smoothly in old computers like that. Back then, Nesticle was a combination of C and assembly, IIRC.
Please read the bill first! I think this bill is fair enough for some part:
"For purposes of section 2319(b) of title 18, the placing of a copyrighted work, without authorization of the copyright owner, on a computer network accessible to members of the public who are able to copy the the work thorugh such access shall be considered to be the distribution, during a 180-day period, of at least 10 copies of that work with a retail values of more than $2,500.
Emphasis mine
Ok, the wording might be ambiguous and maybe interpreted by some stretch. But 10 copies of the same work?? I think none of us would have that. 6 month period? Hmmm... $2,500 retail value?? Mkay...
Also, look at the positive side: It also prohibits spyware and domain name fraud:
(Section 1822) (a)Whoever knowingly offers enabling software for download over the internet and does not --
(1) clearly and conspicuously warn any person downloading that software, before it is downloaded, that it is enabling software and could create security and privacy risk for the user's computer; and
(2) obtain that person's prior consent to the download after the warning;
shall be fined under this title or imprisoned not more than 6 months, or both.
(b)As used in this section, the term 'enabling software' means software that, when installed on the user's computer, enables 3rd parties to store the data on that computer, or use that computer to search other computers' contents over the Internet."
Ok, here's another ambiguous part. Part b can be a double-edged sword: It can be used for good (aka. against spyware) or just simply P2P network. You can petition this part to be clearly worded so that it cannot be specifically targetted for P2P.
The more they bombard us, the more we hate them and only idiots who make their own consumer hate them. If they at least wear their brains, they should exercise self-control. For me, personally, I never ever buy any products on any companies that advertise through spam and I believe most of you do too.
Now, the problem is that what if they advertise some other / competitor product so that the spam makes customers hate these innocent companies? Well, I suppose the innocent companies can sue these spammers for slander (assuming they can trace the culprits back and have long enough law tentacles to reach them). Therefore, either way, the spammers lose.
Our PC GOD Torvalds, which art in Transmeta^H^H^H^H^H^H^H^H^H OSDN
Hallowed be thy skillz
Thy kernel comes, in the US and all the earth
Give us this day our daily updates.
And forgive us our holes, as we apply thine patch.
And lead us not into closed source, but deliver us from Microsoft.
For thine is the kernel, the skillz, and the leetness for ever and ever. Amen.
Only THEN, he can say: "Praise the PC god and Linux open-source apostles, I'm a believer."
Well, the type of Bayesian learning used in this spam filtering is called "Naive Bayesian" and the engine is trained using "supervised learning" technique. Naive Bayes has been proven very successful for text categorization. Spam filtering is even more successful because we essentially categorize e-mails to two labels: "spam" or "not spam".
Supervised learning basically works like this. Feed the engine with multiple examples (in this case, e-mails) with labels (in this case, "spam" or "not spam"). The training usually takes thousands of examples to get good enough accuracy. And take note that we need both "spam" and "not spam" examples to enable the learning engine to distinguish them.
How Naive Bayes works? Well, think of the full Bayesian Network. Bayes net is basically a causal-effect graph with annotated Conditional Probability Table (CPT) on each node denoting the probabilities of possible values. Full Bayes Net takes Directed Acyclic Graph (DAG), but Naive Bayes takes a form of tree instead due to some "naive" assumptions. (Okay, I handwaved a whole lot of details here) And in Learning Naive Bayes, we basically try to construct the tree out of the examples.
Let P(spam) be the percentage of training e-mails that is labelled as "spam" and P(not spam) be the percentage of "not spam" e-mails.
First, let the filter reads all e-mails and collect the words out of them. Weed out duplicates and stop words (common words like "I", "you", "the", etc). Let NumVocab be the number of words after weeding.
Second, process e-mail one by one. Do weeding phase like the above. Let "n" be the number of words on that particular e-mail after the weeding. Scan the word one by one. Let "w" be the current word scanned and "nw" be the number of times word "w" occur in that e-mail. Imagine you have a big two dimensional array to store the result (let's call the array "P"). If the e-mail is labeled "spam", then store (nw+1)/(n+NumVocab) to P[w][spam].
Repeat until all training e-mails are read.
And here comes the testing phase...
When you encounter an e-mail and want to classify whether it's spam or not, you'll need to look up the array P you created earlier. First, you do the weeding phase and scan the word one by one. The algo is like this:
pspam = P(spam); pnospam = P(not spam); foreach unique words w in e-mail do
pspam = pspam * P[w][spam];
pnospam = pnospam * P[w][nospam]; endfor
if (pspam > pnospam) then return IS_SPAM; else return IS_NO_SPAM;
Slashdot Mirror
For those of you interested on learning Taguchi method. Here's a good intro.
Taguchi's objective is robust design, which means building a product, system, or process that works well even in the presence of degrading influences. That means products that deliver value without breaking and services that are enduring while being as simple as possible. Taguchi first determines the control factors that go into designing a product, their interdependencies, then generates an orthogonal array specifying the number of experiments required to find the optimal solution. [....]
That's when Kowalick turned the Taguchi Method to advertising, with the goal of significantly raising the response rate for ad campaigns. [....] The control factors included graphics, colors, and use of humor. The experiments themselves were 12 mailings to 625 addresses each -- two mailings per day over six days. [....]
The vanilla wafer recipe, however, will remain a secret.
------
So.... First of all, is it an advertisement behind an advertisement? Since many of us already has e-mail filter, how will this be not spotted by the filter? Hmm... I smell something really fishy.
In that case, they may as well make an amandment:
We the corporations of the United States, in order to form a more perfect Profit, establish monopoly, insure domestic compliance, provide for the common interest, promote our welfare, and secure the blessings of liberty to ourselves and our posterity, do ordain and establish this Constitution for the United States of America.
[it's a joke people]
Uh... When I heard the term "restricted music", I have the mental image of "Parental Advisory: Explicit Lyrics" or something like that. Maybe "copyrighted music" is a better term?
So say I sit down with my backpack on may lap, it gets me wrong.
Clippy: So, you gain 20 pounds in just 2 days. Would you like to:
[ ] Enroll as sumo wrestler (you are qualified now)
[ ] Enroll in Taco Gym
[ ] Cowboyneal!
[ ] All of above
What is it and why should I care? The article doesnt explain it very well
Oh really? I suggest you read more carefully at the bottom of the page:
Now I hope you can see why Dept of Homeland Security have worries on securities thing.
It's true that some flaws in the OS are inherently design-based. However, even if we make certain design requirements to be incorporated in the OS, it still doesn't guarantee that the OS is secure. I would think that it even can't minimize the number of OS breaches. It would even hamper the OS development in order to comply with their standards.
About the quote regarding the "minimum expectations of consumers for security and general reliability by establishing a floor for these characteristics". I don't think it would be possible the goal of "the least restrictive requirement while not relenting the control" is vague. Unless it provides rigid post- or pre-conditions of each method (in first order logic if necessary) and provide each formal specifications unambiguously, I would still see some leaks here and there. And, guess what? They put the requirement like UML standards: Way to vague. Congratulations.
For those of you who are curious, click here for the draft.
Before people condemn the paper as lame, read the paper at the parent post. Probably the most important results to us are that the author suggests:
I may be wrong, but read the paper yourself.
There IS visuals... See the paper here (see page 5). It's a 9 MB PDF file. That paper also mentions that the fungus DID impact the error correcting part of the CD. And it mentions on page 7:
To begin with, each computation was started by placing one disc into a stand-alone CD or DVD player and pressing play. In all cases so far, placing treated discs into computer-based, CD, CD-RW and DVD players has caused the computer or laptop to crash. For this reason, this study focused only on CD and DVD players that are modular. In most cases the disc was recognized and begain playing. If the player failed to recognize the disc, it was necessary tor emove some cell material using soft cloth [...] Note that cells were completely removed from the innermost region of data at the beginning of the disc, since this is header information necessary for the player to mount the disc.
Then it goes on the snapshot of the "treated" DVD playback on figure 6 at page 9.
Not only that... If the fermented Linux CD is thrown into the dirt, the fungus would cause the dirt to form compost. Thus, SCO would sue compost maker too. Including us, who contribute it through septic tanks... :-)
Know that there are many ways to do one thing and there are pros and contras in each of them. In this case, it seems that NVidia's is not chosen and the way DX9 handles things undermines NVidia's method. It's not necessarily because NVidia sucks. Remember that there are politic struggles among Microsoft, NVidia, and ATI during the inception of DX9? I think NVidia now falls victim of it.
If you read the article, the comparison is against Dual P4 Xeon. Some of the tests didn't enable any hyper-threading stuff (and thus it didn't take any advantage of the dualies. Opteron beat P4 by very high margin. Except for content creation & general usage stuff where the P4 wins. But take that with a grain of salt.
64-bit tests won't be fair to either side. It's like comparing apples to oranges. For me, I'm looking forward to see vis-a-vis comparison on programs that is optimized on either platform. For example: A program that is optimized on Itanium and Opteron and see how they fare.
Eh? This is different than I would expect. In the "formal" computer science definition would be something like generating code (often times machine code or byte code) in compiler context.
Also, in computer software design, it has another definition. Namely: Generating real code (C, C++, Java, whatever) from design templates (like from UML).
Now, these people are overloading the same term?? These make me even more confused.
Assume an elephant weighs about six tons, she says, that would mean that water inside a typical cumulous cloud would weigh about one hundred elephants.
Somehow it reminds me of RIAA's math equivalent.
Here
Wrong. That's a different paper.
If the water's polluted enough, anyone can walk on it.
Yeah, like after pouring a decent amount of concrete mix in it...
No offense, but...
NES processor (it's 6502, IIRC) is about 1.79 MHz and computer nowadays are thousand times faster than that. I would appreciate it if you can run your VB emulator on 386 computers and still have a decent speed.... I know that some of the older NES emulators like Nesticle still manage to run smoothly in old computers like that. Back then, Nesticle was a combination of C and assembly, IIRC.
Please read the bill first! I think this bill is fair enough for some part:
"For purposes of section 2319(b) of title 18, the placing of a copyrighted work, without authorization of the copyright owner, on a computer network accessible to members of the public who are able to copy the the work thorugh such access shall be considered to be the distribution, during a 180-day period, of at least 10 copies of that work with a retail values of more than $2,500.
Emphasis mine
Ok, the wording might be ambiguous and maybe interpreted by some stretch. But 10 copies of the same work?? I think none of us would have that. 6 month period? Hmmm... $2,500 retail value?? Mkay...
Also, look at the positive side: It also prohibits spyware and domain name fraud:
(Section 1822) (a)Whoever knowingly offers enabling software for download over the internet and does not --
(1) clearly and conspicuously warn any person downloading that software, before it is downloaded, that it is enabling software and could create security and privacy risk for the user's computer; and
(2) obtain that person's prior consent to the download after the warning;
shall be fined under this title or imprisoned not more than 6 months, or both.
(b)As used in this section, the term 'enabling software' means software that, when installed on the user's computer, enables 3rd parties to store the data on that computer, or use that computer to search other computers' contents over the Internet."
Ok, here's another ambiguous part. Part b can be a double-edged sword: It can be used for good (aka. against spyware) or just simply P2P network. You can petition this part to be clearly worded so that it cannot be specifically targetted for P2P.
The more they bombard us, the more we hate them and only idiots who make their own consumer hate them. If they at least wear their brains, they should exercise self-control. For me, personally, I never ever buy any products on any companies that advertise through spam and I believe most of you do too.
Now, the problem is that what if they advertise some other / competitor product so that the spam makes customers hate these innocent companies? Well, I suppose the innocent companies can sue these spammers for slander (assuming they can trace the culprits back and have long enough law tentacles to reach them). Therefore, either way, the spammers lose.
The author should close with the Linux Prayer:
Our PC GOD Torvalds, which art in Transmeta^H^H^H^H^H^H^H^H^H OSDN
Hallowed be thy skillz
Thy kernel comes, in the US and all the earth
Give us this day our daily updates.
And forgive us our holes, as we apply thine patch.
And lead us not into closed source, but deliver us from Microsoft.
For thine is the kernel, the skillz, and the leetness for ever and ever. Amen.
Only THEN, he can say: "Praise the PC god and Linux open-source apostles, I'm a believer."
Oh, please... It's a joke, isn't it?
I'll be glad to see if there's any managers USE Debian. Managers INSTALLING Debian... ?? Wow! It's so... "news that matters".
Screening is not a good solution just like censoring is a bad idea.
The better way is to explain or display negative examples in a positive way. This way we can educate people on what is good and what is bad.
Well, the type of Bayesian learning used in this spam filtering is called "Naive Bayesian" and the engine is trained using "supervised learning" technique. Naive Bayes has been proven very successful for text categorization. Spam filtering is even more successful because we essentially categorize e-mails to two labels: "spam" or "not spam".
Supervised learning basically works like this. Feed the engine with multiple examples (in this case, e-mails) with labels (in this case, "spam" or "not spam"). The training usually takes thousands of examples to get good enough accuracy. And take note that we need both "spam" and "not spam" examples to enable the learning engine to distinguish them.
How Naive Bayes works? Well, think of the full Bayesian Network. Bayes net is basically a causal-effect graph with annotated Conditional Probability Table (CPT) on each node denoting the probabilities of possible values. Full Bayes Net takes Directed Acyclic Graph (DAG), but Naive Bayes takes a form of tree instead due to some "naive" assumptions. (Okay, I handwaved a whole lot of details here) And in Learning Naive Bayes, we basically try to construct the tree out of the examples.
Let P(spam) be the percentage of training e-mails that is labelled as "spam" and P(not spam) be the percentage of "not spam" e-mails.
First, let the filter reads all e-mails and collect the words out of them. Weed out duplicates and stop words (common words like "I", "you", "the", etc). Let NumVocab be the number of words after weeding.
Second, process e-mail one by one. Do weeding phase like the above. Let "n" be the number of words on that particular e-mail after the weeding. Scan the word one by one. Let "w" be the current word scanned and "nw" be the number of times word "w" occur in that e-mail. Imagine you have a big two dimensional array to store the result (let's call the array "P"). If the e-mail is labeled "spam", then store (nw+1)/(n+NumVocab) to P[w][spam].
Repeat until all training e-mails are read.
And here comes the testing phase...
When you encounter an e-mail and want to classify whether it's spam or not, you'll need to look up the array P you created earlier. First, you do the weeding phase and scan the word one by one. The algo is like this:
Hope this helps.
...please come by and take a look.
Translation:
Please slashdot me and don't even peek... :-)