Imperative languages are based on the execution of individual commands. Fundamentally they are based on the concept of assignment -- moving data from one place to another. Functional languages are based on the evaluation of expressions and the absence of side-effects. Pseudo-functional languages have variables, loops, and side-effects but are mainly based on functional concepts. Declarative languages are based on the concept of goals, and the recursive description of how those goals should be achieved, or the definition of what constitutes achievement of the goals.
I'm not sure why you consider Forth a declarative language. To me it seems more like an imperative language with an unusual syntax.
But an engineer doesn't take a whole physics workload. He/she takes the basic classes, and then some classes more specific to their engineering field. A mechanical engineer would probably take a class on the physics of vibration, for example.
Similarly, software engineering would be based in computer science but would focus on engineering itself -- designing large systems, fault tolerance, working as a team, and things of that sort. I can attest that many incredibly intelligent computer scientists don't have any idea how to write a coherent large program! Just checking for error return codes seems too much of a bother for some:-)
I say all this as a person who considers himself foremost a "computer scientist" but who is stuck doing software engineering for the moment.
The CS program at Portland State University starts with assembly in their introductory classes. At the time I was there, it was x86 assembly, but I've heard that some professors are using Sparc assembly as well -- not a good idea in my opinion, simply because of 1) the delay slot and 2) the sethi instruction, both of which are a little confusing for someone who's never coded before, let alone never coded in assembly language.
I think it's a little weird to call this "Learning Computer Science via Assembly Language." It's programming, not computer science. Computer science is really only marginally about computers. It has to do more with algorithms, logic, and mathematics.
You can study computer science, and produce new knowledge in the field, without ever touching a computer.
This misunderstanding is, I think, part of the reason so many students drop out of CompSci. They head into it thinking it's about programming, and are startled to find that computation and programming are not equivalent.
That's why the Compilers course at PSU is considered the "filter" which kills all the students who aren't really interested in computer science. They really need to spin off a seperate "Software engineering" school for these students, since what they really want to study is programming.
A fully validated WHOIS database would make it trivial to enforce punishment against people who use spammers to promote the websites and scams on said websites registered to them.
A fully validated WHOIS database would also make it trivial to enforce punishment against those who express politically dissident views. It would no longer be possible to create a domain for political discussion without the government knowing who you are and where you are.
But I guess you're okay with that scenario as long as it stops spam, right?
The responses on the thread just go to prove something I've always suspected:
If Jesus Christ were born in our time, you'd all be calling him a fucking idiot.
Jesus did things without caring if he would be appreciated for them. We view this behavior as God-like in that context. But in today's world, a person trying to do the same is treated like an imbecile, or accused of being naive. "A fool and his money..." Maybe the fool understands something about money that you don't. Maybe you're the fool.
Our individualism in this country is a strength, but simultaneously it makes us all assholes.
How about you all shut up? He's a far better man than any of you will ever be. That is, if he's serious.
I say this and I'm not even Christian.
Oh, and don't try to deflect my point by saying he's investing in technology that will soon be obsolete. You'd be railing on him equally hard if he'd simply decided to divvy up the $7k amongst his neighbors.
The U.S. has this bullheaded, senseless addiction to a system of measurement where temperature is graded between the lowest and highest reproducible temperatures in a 19th century lab
You're trying to say they couldn't boil water? In order to boil water you need something hotter than 212 degrees Fahrenheit. So clearly they could produce temperatures in excess of 212. How hot exactly do you think fire is?
and where linear measurement is based on the length of some long-gone king's thumb, foot and arm.
Actually these days, the inch is defined in terms of the metric system: an inch is precisely 2.54 centimeters, thus a foot is exactly 0.3048 meters. It's not based on some long-dead king. At least, not anymore.
Why are there three teaspoons in a tablespoon, but only two tablespoons in an ounce, but then 8 ounces in a cup but only two cups in a pint, and two pints in a quart, but then 4 quarts in a gallon?
Because those were convenient quantities at the time they were standardized. You could make everything powers of ten (perhaps powers of two would make more sense) but then you'd have to use more fractions to describe commonly-used quantities. The system was arrived at through centuries of trial and error.
You'll notice that the multiples in the U.S. system have many factors: a foot is evenly divisible into sixths, quarters, thirds, and halfs, all without using fractional numbers of inches. In contrast, a 10 centimeter length can only be divided by 2, 5, or 10. You don't even get quarters out of the deal.
The term "quart" should make it easy to remember that there are four of them in a gallon:-)
I strongly support the use of metric in the sciences, but for everyday measurement, why not let people use whatever is convenient for them? Maybe you have trouble remembering the multiples, but other people don't. And when's the last time you needed to measure in tablespoons and you weren't cooking something? Knowing all the cooking measures is a standard of pride among chefs.
I've thought of doing this, except with sound instead of light. Instead of a "system monitor" that takes up a large chunk of my screen real estate, what I really want is a monitor that uses sound to tell me what's going on.
System load could be signified by clicks, with the frequency of the clicks increasing as system load increases.
Each new TCP connection would make some kind of "boing" sound, with the frequency again depending on what service I'm connecting to (http would go boioioioing, ssh would go beeeerooooing, etc)
Memory usage would be signified by a double-beep, "beeee-beep," with the "duty cycle" indicating the percentage of memory usage. Two short beeps means lots of memory is free. One long "beeeep" means I'm swapping to disk.
Disk seek activity would be a series of random bleeping sounds, like Brownian motion across frequencies.
Basically, I would like an irritating cacophony of sound to emanate from my workstation, which only I can interpret:-)
In a way it does. It firsts asks for a username, and then a password. If one of them is incorrect, you don't get access. But SSH doesn't tell you which one was incorrect.
This reminds me of a cgi driven website I visited a loooong time ago (1996?)
I was creating a user account, and was using the password "beelzebub". However, the system refused to let me create the account. It displayed a page which stated "That password is invalid: It is being used by another user. Please select a unique password."
Apparently, some genius thought it was good security to ensure that no two users had the same password. I hope you can see the intrinsic flaw in this:-)
This adds a layer of obscurity to a security policy. It can't substitute for security, but it certainly can help.
An analogy would be a military base with a ten-foot-thick steel blast door. This is like having a door that teleports around at random, which can only be frozen in one spot by speaking some magic word. Even if you know the word, you still don't have the key to the door. But if you do have the key, you still can't get in without the magic word because the door keeps teleporting around.
Obscurity is great, if it is part of a layered security policy which is ultimately based on strong cryptography. This is a really cool idea!
...the average computer user isn't going to start becoming computer-savy anytime soon. Even this generation of children are woefully ignorant for the most part.
I think you're completely wrong. Either you don't hang around little children, or the ones you hang around with are stunted somehow...
I've seen an 18 month old sit on her mother's lap and play Flash games. She can't read, but she knows how to navigate a hierarchical menu going only off the color and shape coding.
This same child just turned three. Let me tell you just some of the other things I've seen her do:
1. Hitting the DEL key to get into the BIOS, then proceeding to set everything to random settings.
2. Change screen resolution. She only had to be shown twice.
3. Ctrl-Alt-Del when shit ain't working.
4. Insert, eject, rewind, and fast forward a VCR. She can use slow motion, also. She knows how to switch to "Line In" when she wants to play a video game.
We've learned not to let her observe us doing something "technical" because she will try to reproduce it, usually with disasterous results. Things like, changing the video RAM clock rate. I'm serious.
Pretty much any three year old I've met, whose parents aren't fucking idiots who damage their brains, has been an absolute genious. Something, I'm not sure quite what, turns them into idiots around age 6.
Is it parents failing to provide stimulus? Is it the retarding effect of the American public school system? Probably a little of both, but I assure you, children have absolutely no problem with technology. Stupidity has to be forced on them, they are naturally extremely intelligent.
That way, you don't need to be technically elite, you just need to choose someone who is.
Without being technically elite, you have little way of judging who else is technically elite.
Look at the "expert" computer witnesses that are often called to testify in court. Many of these people are ignorant morons. How did they ever get into a courtroom in the first place? Because an attorney or a judge who wasn't technically inclined decided who was an expert witness and who wasn't.
If a judge cannot correctly figure out who knows what he's talking about and who doesn't, how can you reasonably expect Joe Average to be able to choose a web hosting company that fits your criteria for technological eliteness?
The internet is about social revolution, not technology. By requiring people to become technically savvy in order to participate in it, you are depriving millions of people from participating in the most radical social change that has ever occurred on this planet. And why, all because they don't know the difference between a megabyte and a megahertz?
I'm not attacking the guy for trying to build a better filter, but because he was all like "it doesn't take long to manually categorise spam".
Did you even understand the point I was trying to make? I'm not saying that spam isn't a problem because we (as humans) can filter it, I'm saying that because humans can filter it, there is an upper bound on how confusing spam can be to an automated filter. I was making a (semi-)mathematical argument, not telling you you delete your spam manually. Sheesh.
You're right, my terminology was a little sloppy. I was of course referring to the "Bayesian" algorithm that readers of Slashdot are commonly familiar with (which isn't even the same as the unigram Bayesian algorithm commonly discussed in introductory NLP texts), not the entire framework of Bayesian statistics.
My prime point was that the algorithm referred to as "Bayesian filtering" is by no means the best thing out there. I'll try to strike a better balance between clarity and technical accuracy next time.
What, exactly, do you mean by "not feasible to do on PC hardware"?
We're talking about O(n^3) algorithms here. If you double the amount of data being fed to the algorithm, the run time goes up 8 times. At least it's not exponential, I guess.
And we're not talking about keeping an entire mainframe working hard -- we're talking keeping a Beowulf cluster of mainframes working hard. Imagine a lexicon of 100,000 symbols (not unreasonable), where the run time is O(n^3) in the lexicon size. That's a quadrillion algorithmic steps to be performed -- assuming you can do each step in a nanosecond (a total impossibility) it will still take you 11 days to finish! And that's just one iteration -- you need to keep iterating until the algorithm converges!
The interesting work currently being done is in figuring out how to reduce the run times of these algorithms to at least O(n^2). There are some linear algorithms (to be precise, they are O(l*m*n) where l and m are significantly smaller than n), in fact, I use one of these algorithms in my own filter. We could do very amazing things if we had the power to run some of these O(n^3) algorithms over realistic data sets. Unfortunately we can't do it on any currently existing hardware.
It's sort of the same situation with computer graphics. All the cool special effects you see in movies these days were understood in theory way back in the late '70s, early '80s. It wasn't until 1995 or thereabout that computers actually became fast enough to implement those ideas.
Misleading subject lines make it harder to bulk delete stuff.
If in doubt, it's spam. Simple rule, really. If some company really wants to report that my "Payment is past due" they'll send me a letter. Besides, most corporate contacts aren't going to be named "kaislyais." It takes a split second to make this judgment.
Maybe you are smarter than the average spammer, but you need to prove it. When you produce your magical filter, I'll consider your intellectual prowess.
You make it sound like I was claiming to be the world's statistical NLP expert. Yes, I have my own ideas, along with hundreds of other people. In general I think it's a safe bet to say we're smarter than spammers. I'm not really interested in proving anything, since I never intended this to be an IQ competition. I don't think it's big-headed to assume I'm smarter than a slimeball.
As for the magical filter, it isn't magical, and it already exists. As I said, PC hardware severely limits the kinds of algorithms that we can use without making the user impatient. If you'd like to run it, I'll gladly send it to you, but don't expect me to document anything, and don't complain about the run time:-)
Currently I use a hybrid filter which uses a word clustering algorithm to concentrate the information which flows into a feed forward neural network. Currently I'm examining the use of SOMs instead of feed forward networks to automatically generate document classes. I'm also looking into ways to use the neural network as feedback to fine-tune the clustering algorithm.
I believe these blends of statistical and information-theory techniques with "sloppier" systems like neural networks will become more and more valuable as people continue to research them.
As I said though, it's a research project which means it has no documentation and the training process is laborious and not yet automated. It's not the kind of thing you install on a mail server and just forget about. I'm more interested in the filter than the mechanics of getting it integrated into mail systems.
The stuff you're talking about is all fine, but it will fail because the spammers will evolve to defeat it.
I think you overestimate the intelligence of these creeps. The fact that spammers are using more and more of these garbage terms, randomizers, and other hacks to get around the filters actually encourages me -- it demonstrates that they really don't have the slightest clue how statistical content based filtering actually works. Currently, they are taking advantage of the extremely bad decision to assign a 0.4 score to unknown words. The spammers are exploiting a crack in the armor, which means the armor needs to be fixed.
A human can filter spam. A spammer can't weasel his way around human intelligence, so this sets an upper bound on how advanced the spammer techniques can get. All we have to do is get document classification up to the point of competitiveness with human performance, and the problem is solved. And research into these directions isn't wasted, because the motivation for the research is for actual important document organization tasks. The effect of stomping out spam will be a cool side effect.
If a spammer was ever actually intelligent enough to get around serious, well-constructed classifiers, I highly doubt he would be in the business of spamming. To suggest that spammers could intellectually compete with people whose have spent years specializing in statistical language processing is a tad bit ridiculous.
At some point, to sell something, the spammer has to say something intelligible which is an advertisement. They can't hide this. Techniques which are foiled by bogus terms at the bottom of the email are broken. It's not a valid reason to believe that spammers are actually getting smart.
People just have to realise that filtering based on content doesn't work, and will never work, until perhaps we have strong AI.
That's an overly strong statement to make, and even a little bit irritating to people like myself who actually implement statistical content filters, natural language systems, etc.
If you are equating "content based filtering" to "Bayesian filtering" then you really only understand 1% of the current state of document classification. Bayesian filtering is a rage right now because it's a linear time algorithm (i.e., implementable on PC hardware). There are document classification schemes that will eat Bayesian for lunch, which are not appropriate for email filtering at this time because of their computational cost. But with continual progress on the algorithms, new methods for reducing search spaces via extremely clever sense-similarity heuristics, and with computers doubling in speed every 18 months, it's closer than you think.
The spam/ham problem is what data mining researchers would call a "toy problem." You want us to classify documents into only two classifications? Only two? Piece of cake. The problem is, you want us to do it on PC hardware where it isn't feasible to run O(n^2) or O(n^3) machine learning algorithms.
Let the researchers continue what they're doing. People are just now starting to apply SVMs and other cool techniques to the problem of spam filtering. You'd be amazed at how many of the well-known data mining and statistical NLP researchers have not even thought of using their arsenal against spam.
You're confused. A point will always be 1/72nd of an inch. Yes, even in Microsoft land.
The number 96 you seem to be remembering is the assumed dots-per-inch resolution of a CRT output device. That is, in the absence of information to the contrary (such as info from the video driver) Windows assumes that the CRT display device is a 96 DPI device.
It's quite irritating that the word "point" is used to refer both to a physical unit (1/72 inch) and a logical unit (a pixel on a screen).
Right now, my understanding is that cryogenics is a crock because the freezing process causes the cells in the body to, for lack of a better word, explode.
More like they pop, as in a popping balloon. It isn't the expansion of water inside the cell which bursts it (the membrane is elastic), it is the fact that ice crystallizes and forms very sharp crystal edges which cut through the cell membrane like a knife edge.
The idea behind cryogenic flash freezing is that by freezing the tissue extremely quickly, these ice crystals don't have a chance to form and the water instead gels into a more amorphous structure where ice crystals are small, or perhaps not even present.
That being said, I don't think I would want my body frozen while there's still a chance of getting it fixed in this century:-) It's the kind of thing I might consider if I knew that I would certainly die soon anyway.
Fluorine (you misspelled it, argh), chlorine, bromine, and iodine, and don't forget astatine all end in 'ine' because they are all halogens.
Argon, xenon, radon, and also neon and krypton all end in 'on' because they are noble gases.
The other oddballs you mention: hydrogen, oxygen, boron, carbon, silicon, nitrogen, were all named back when chemistry was a little less organized than it is today. However, there is still structure in their names: hydrogen, oxygen, and nitrogen are all gases, and the 'gen' implies that they are involved in the creation of some other substance. In the case of hydrogen, water. In the case of oxygen, acid (although this turned out to be incorrect -- oxygen has nothing to do with acidity).
Boron, carbon, and silicon are all solid, nonmetallic elements.
You'll notice that all the metals end in 'ium', except for those which have been known far before the advent of chemistry (gold, silver, iron, nickel, copper, etc.)
The vast majority of elements end in 'ium' because the vast majority of elements are metallic in nature.
The sad part is there is no alternative on campus because the university gets funded by Coke. Most students don't have vehicals to go else ware for soft drinks and so pay the gouging prices.
That's what happens when you like something that a corporation makes. You sit there and complain about Coke yet continue to drink their product. Why not just drink water?
The difference between what you're talking about, and the textbook situation, is that in many cases you have no option but to buy the book. You can stop drinking Coke at any time. It makes sense to whine about the price of textbooks -- not the price of Coke.
Since experts supposedly write the books because of the "public or perish" system in academia, they aren't really doing it for the money.
Why do the authors of textbooks continue to insist on going through publishers? Why do they not produce computerized works instead of printed books? If they really are making a pittance on the royalties, they shouldn't care whether the book is in print or in the form of a PDF (or some other doc format) -- the point is to get a textbook pub out.
Is this purely because of the editorial facilities of publishing companies? Is it really that hard to edit and typeset a document yourself? People do it themselves all the time in academic publications, why not in textbooks?
The bug does exist in other browsers, specifically Mozilla. Try it for yourself
I'm using the version of Mozilla that came STOCK with Red Hat 8.0 (that is, it's an older version). When I hover over the wacky link, I see "http://www.microsoft.com#" (where the '#' is actually a little box outline).
However, when I click the link, what displays in the address bar is "http://www.microsoft.com%01%00@secunia.com/intern et_explorer_address_bar_spoofing_test/".
So the hacker can spoof the status bar. Big fucking deal -- he can do that already with JavaScript. And even if you DO click the link, all you have to do is glance at the address bar to see you've been fooled.
In other words, there's a bug somewhat like the IE bug in my version of Mozilla, but it isn't the same bug, it has different symptoms, and it seems much less dangerous -- a funny little box appears when you hover over the link, this is clue 1, and the address bar shows exactly where you've really gone to, this is clue 2.
Slashdot Mirror
1. Imperative
-- 1a. Procedural (Pascal/C/BASIC)
-- 1b. Object-Oriented (Eiffel/Smalltalk/Java/C++)
-- 1c. Assembly language
2. Functional-Type
-- 2a. Pseudo-functional (Scheme/Lisp)
-- 2b. Pure functional (Haskell/ML/Pure lambda calculus)
3. Declarative (Prolog)
Imperative languages are based on the execution of individual commands. Fundamentally they are based on the concept of assignment -- moving data from one place to another. Functional languages are based on the evaluation of expressions and the absence of side-effects. Pseudo-functional languages have variables, loops, and side-effects but are mainly based on functional concepts. Declarative languages are based on the concept of goals, and the recursive description of how those goals should be achieved, or the definition of what constitutes achievement of the goals.
I'm not sure why you consider Forth a declarative language. To me it seems more like an imperative language with an unusual syntax.
But an engineer doesn't take a whole physics workload. He/she takes the basic classes, and then some classes more specific to their engineering field. A mechanical engineer would probably take a class on the physics of vibration, for example.
Similarly, software engineering would be based in computer science but would focus on engineering itself -- designing large systems, fault tolerance, working as a team, and things of that sort. I can attest that many incredibly intelligent computer scientists don't have any idea how to write a coherent large program! Just checking for error return codes seems too much of a bother for some :-)
I say all this as a person who considers himself foremost a "computer scientist" but who is stuck doing software engineering for the moment.
I think it's a little weird to call this "Learning Computer Science via Assembly Language." It's programming, not computer science. Computer science is really only marginally about computers. It has to do more with algorithms, logic, and mathematics.
You can study computer science, and produce new knowledge in the field, without ever touching a computer.
This misunderstanding is, I think, part of the reason so many students drop out of CompSci. They head into it thinking it's about programming, and are startled to find that computation and programming are not equivalent.
That's why the Compilers course at PSU is considered the "filter" which kills all the students who aren't really interested in computer science. They really need to spin off a seperate "Software engineering" school for these students, since what they really want to study is programming.
A fully validated WHOIS database would also make it trivial to enforce punishment against those who express politically dissident views. It would no longer be possible to create a domain for political discussion without the government knowing who you are and where you are.
But I guess you're okay with that scenario as long as it stops spam, right?
If Jesus Christ were born in our time, you'd all be calling him a fucking idiot.
Jesus did things without caring if he would be appreciated for them. We view this behavior as God-like in that context. But in today's world, a person trying to do the same is treated like an imbecile, or accused of being naive. "A fool and his money..." Maybe the fool understands something about money that you don't. Maybe you're the fool.
Our individualism in this country is a strength, but simultaneously it makes us all assholes.
How about you all shut up? He's a far better man than any of you will ever be. That is, if he's serious.
I say this and I'm not even Christian.
Oh, and don't try to deflect my point by saying he's investing in technology that will soon be obsolete. You'd be railing on him equally hard if he'd simply decided to divvy up the $7k amongst his neighbors.
You're trying to say they couldn't boil water? In order to boil water you need something hotter than 212 degrees Fahrenheit. So clearly they could produce temperatures in excess of 212. How hot exactly do you think fire is?
and where linear measurement is based on the length of some long-gone king's thumb, foot and arm.
Actually these days, the inch is defined in terms of the metric system: an inch is precisely 2.54 centimeters, thus a foot is exactly 0.3048 meters. It's not based on some long-dead king. At least, not anymore.
Why are there three teaspoons in a tablespoon, but only two tablespoons in an ounce, but then 8 ounces in a cup but only two cups in a pint, and two pints in a quart, but then 4 quarts in a gallon?
Because those were convenient quantities at the time they were standardized. You could make everything powers of ten (perhaps powers of two would make more sense) but then you'd have to use more fractions to describe commonly-used quantities. The system was arrived at through centuries of trial and error.
You'll notice that the multiples in the U.S. system have many factors: a foot is evenly divisible into sixths, quarters, thirds, and halfs, all without using fractional numbers of inches. In contrast, a 10 centimeter length can only be divided by 2, 5, or 10. You don't even get quarters out of the deal.
The term "quart" should make it easy to remember that there are four of them in a gallon :-)
I strongly support the use of metric in the sciences, but for everyday measurement, why not let people use whatever is convenient for them? Maybe you have trouble remembering the multiples, but other people don't. And when's the last time you needed to measure in tablespoons and you weren't cooking something? Knowing all the cooking measures is a standard of pride among chefs.
System load could be signified by clicks, with the frequency of the clicks increasing as system load increases.
Each new TCP connection would make some kind of "boing" sound, with the frequency again depending on what service I'm connecting to (http would go boioioioing, ssh would go beeeerooooing, etc)
Memory usage would be signified by a double-beep, "beeee-beep," with the "duty cycle" indicating the percentage of memory usage. Two short beeps means lots of memory is free. One long "beeeep" means I'm swapping to disk.
Disk seek activity would be a series of random bleeping sounds, like Brownian motion across frequencies.
Basically, I would like an irritating cacophony of sound to emanate from my workstation, which only I can interpret :-)
This reminds me of a cgi driven website I visited a loooong time ago (1996?)
I was creating a user account, and was using the password "beelzebub". However, the system refused to let me create the account. It displayed a page which stated "That password is invalid: It is being used by another user. Please select a unique password."
Apparently, some genius thought it was good security to ensure that no two users had the same password. I hope you can see the intrinsic flaw in this :-)
An analogy would be a military base with a ten-foot-thick steel blast door. This is like having a door that teleports around at random, which can only be frozen in one spot by speaking some magic word. Even if you know the word, you still don't have the key to the door. But if you do have the key, you still can't get in without the magic word because the door keeps teleporting around.
Obscurity is great, if it is part of a layered security policy which is ultimately based on strong cryptography. This is a really cool idea!
I think you're completely wrong. Either you don't hang around little children, or the ones you hang around with are stunted somehow...
I've seen an 18 month old sit on her mother's lap and play Flash games. She can't read, but she knows how to navigate a hierarchical menu going only off the color and shape coding.
This same child just turned three. Let me tell you just some of the other things I've seen her do:
1. Hitting the DEL key to get into the BIOS, then proceeding to set everything to random settings.
2. Change screen resolution. She only had to be shown twice.
3. Ctrl-Alt-Del when shit ain't working.
4. Insert, eject, rewind, and fast forward a VCR. She can use slow motion, also. She knows how to switch to "Line In" when she wants to play a video game.
We've learned not to let her observe us doing something "technical" because she will try to reproduce it, usually with disasterous results. Things like, changing the video RAM clock rate. I'm serious.
Pretty much any three year old I've met, whose parents aren't fucking idiots who damage their brains, has been an absolute genious. Something, I'm not sure quite what, turns them into idiots around age 6.
Is it parents failing to provide stimulus? Is it the retarding effect of the American public school system? Probably a little of both, but I assure you, children have absolutely no problem with technology. Stupidity has to be forced on them, they are naturally extremely intelligent.
Without being technically elite, you have little way of judging who else is technically elite.
Look at the "expert" computer witnesses that are often called to testify in court. Many of these people are ignorant morons. How did they ever get into a courtroom in the first place? Because an attorney or a judge who wasn't technically inclined decided who was an expert witness and who wasn't.
If a judge cannot correctly figure out who knows what he's talking about and who doesn't, how can you reasonably expect Joe Average to be able to choose a web hosting company that fits your criteria for technological eliteness?
The internet is about social revolution, not technology. By requiring people to become technically savvy in order to participate in it, you are depriving millions of people from participating in the most radical social change that has ever occurred on this planet. And why, all because they don't know the difference between a megabyte and a megahertz?
Did you even understand the point I was trying to make? I'm not saying that spam isn't a problem because we (as humans) can filter it, I'm saying that because humans can filter it, there is an upper bound on how confusing spam can be to an automated filter. I was making a (semi-)mathematical argument, not telling you you delete your spam manually. Sheesh.
My prime point was that the algorithm referred to as "Bayesian filtering" is by no means the best thing out there. I'll try to strike a better balance between clarity and technical accuracy next time.
We're talking about O(n^3) algorithms here. If you double the amount of data being fed to the algorithm, the run time goes up 8 times. At least it's not exponential, I guess.
And we're not talking about keeping an entire mainframe working hard -- we're talking keeping a Beowulf cluster of mainframes working hard. Imagine a lexicon of 100,000 symbols (not unreasonable), where the run time is O(n^3) in the lexicon size. That's a quadrillion algorithmic steps to be performed -- assuming you can do each step in a nanosecond (a total impossibility) it will still take you 11 days to finish! And that's just one iteration -- you need to keep iterating until the algorithm converges!
The interesting work currently being done is in figuring out how to reduce the run times of these algorithms to at least O(n^2). There are some linear algorithms (to be precise, they are O(l*m*n) where l and m are significantly smaller than n), in fact, I use one of these algorithms in my own filter. We could do very amazing things if we had the power to run some of these O(n^3) algorithms over realistic data sets. Unfortunately we can't do it on any currently existing hardware.
It's sort of the same situation with computer graphics. All the cool special effects you see in movies these days were understood in theory way back in the late '70s, early '80s. It wasn't until 1995 or thereabout that computers actually became fast enough to implement those ideas.
If in doubt, it's spam. Simple rule, really. If some company really wants to report that my "Payment is past due" they'll send me a letter. Besides, most corporate contacts aren't going to be named "kaislyais." It takes a split second to make this judgment.
Maybe you are smarter than the average spammer, but you need to prove it. When you produce your magical filter, I'll consider your intellectual prowess.
You make it sound like I was claiming to be the world's statistical NLP expert. Yes, I have my own ideas, along with hundreds of other people. In general I think it's a safe bet to say we're smarter than spammers. I'm not really interested in proving anything, since I never intended this to be an IQ competition. I don't think it's big-headed to assume I'm smarter than a slimeball.
As for the magical filter, it isn't magical, and it already exists. As I said, PC hardware severely limits the kinds of algorithms that we can use without making the user impatient. If you'd like to run it, I'll gladly send it to you, but don't expect me to document anything, and don't complain about the run time :-)
Currently I use a hybrid filter which uses a word clustering algorithm to concentrate the information which flows into a feed forward neural network. Currently I'm examining the use of SOMs instead of feed forward networks to automatically generate document classes. I'm also looking into ways to use the neural network as feedback to fine-tune the clustering algorithm.
I believe these blends of statistical and information-theory techniques with "sloppier" systems like neural networks will become more and more valuable as people continue to research them.
As I said though, it's a research project which means it has no documentation and the training process is laborious and not yet automated. It's not the kind of thing you install on a mail server and just forget about. I'm more interested in the filter than the mechanics of getting it integrated into mail systems.
I think you overestimate the intelligence of these creeps. The fact that spammers are using more and more of these garbage terms, randomizers, and other hacks to get around the filters actually encourages me -- it demonstrates that they really don't have the slightest clue how statistical content based filtering actually works. Currently, they are taking advantage of the extremely bad decision to assign a 0.4 score to unknown words. The spammers are exploiting a crack in the armor, which means the armor needs to be fixed.
A human can filter spam. A spammer can't weasel his way around human intelligence, so this sets an upper bound on how advanced the spammer techniques can get. All we have to do is get document classification up to the point of competitiveness with human performance, and the problem is solved. And research into these directions isn't wasted, because the motivation for the research is for actual important document organization tasks. The effect of stomping out spam will be a cool side effect.
If a spammer was ever actually intelligent enough to get around serious, well-constructed classifiers, I highly doubt he would be in the business of spamming. To suggest that spammers could intellectually compete with people whose have spent years specializing in statistical language processing is a tad bit ridiculous.
At some point, to sell something, the spammer has to say something intelligible which is an advertisement. They can't hide this. Techniques which are foiled by bogus terms at the bottom of the email are broken. It's not a valid reason to believe that spammers are actually getting smart.
That's an overly strong statement to make, and even a little bit irritating to people like myself who actually implement statistical content filters, natural language systems, etc.
If you are equating "content based filtering" to "Bayesian filtering" then you really only understand 1% of the current state of document classification. Bayesian filtering is a rage right now because it's a linear time algorithm (i.e., implementable on PC hardware). There are document classification schemes that will eat Bayesian for lunch, which are not appropriate for email filtering at this time because of their computational cost. But with continual progress on the algorithms, new methods for reducing search spaces via extremely clever sense-similarity heuristics, and with computers doubling in speed every 18 months, it's closer than you think.
The spam/ham problem is what data mining researchers would call a "toy problem." You want us to classify documents into only two classifications? Only two? Piece of cake. The problem is, you want us to do it on PC hardware where it isn't feasible to run O(n^2) or O(n^3) machine learning algorithms.
Let the researchers continue what they're doing. People are just now starting to apply SVMs and other cool techniques to the problem of spam filtering. You'd be amazed at how many of the well-known data mining and statistical NLP researchers have not even thought of using their arsenal against spam.
It's coming, please be patient.
The number 96 you seem to be remembering is the assumed dots-per-inch resolution of a CRT output device. That is, in the absence of information to the contrary (such as info from the video driver) Windows assumes that the CRT display device is a 96 DPI device.
It's quite irritating that the word "point" is used to refer both to a physical unit (1/72 inch) and a logical unit (a pixel on a screen).
More like they pop, as in a popping balloon. It isn't the expansion of water inside the cell which bursts it (the membrane is elastic), it is the fact that ice crystallizes and forms very sharp crystal edges which cut through the cell membrane like a knife edge.
The idea behind cryogenic flash freezing is that by freezing the tissue extremely quickly, these ice crystals don't have a chance to form and the water instead gels into a more amorphous structure where ice crystals are small, or perhaps not even present.
That being said, I don't think I would want my body frozen while there's still a chance of getting it fixed in this century :-) It's the kind of thing I might consider if I knew that I would certainly die soon anyway.
Fluorine (you misspelled it, argh), chlorine, bromine, and iodine, and don't forget astatine all end in 'ine' because they are all halogens.
Argon, xenon, radon, and also neon and krypton all end in 'on' because they are noble gases.
The other oddballs you mention: hydrogen, oxygen, boron, carbon, silicon, nitrogen, were all named back when chemistry was a little less organized than it is today. However, there is still structure in their names: hydrogen, oxygen, and nitrogen are all gases, and the 'gen' implies that they are involved in the creation of some other substance. In the case of hydrogen, water. In the case of oxygen, acid (although this turned out to be incorrect -- oxygen has nothing to do with acidity).
Boron, carbon, and silicon are all solid, nonmetallic elements.
You'll notice that all the metals end in 'ium', except for those which have been known far before the advent of chemistry (gold, silver, iron, nickel, copper, etc.)
The vast majority of elements end in 'ium' because the vast majority of elements are metallic in nature.
That's what happens when you like something that a corporation makes. You sit there and complain about Coke yet continue to drink their product. Why not just drink water?
The difference between what you're talking about, and the textbook situation, is that in many cases you have no option but to buy the book. You can stop drinking Coke at any time. It makes sense to whine about the price of textbooks -- not the price of Coke.
Why do the authors of textbooks continue to insist on going through publishers? Why do they not produce computerized works instead of printed books? If they really are making a pittance on the royalties, they shouldn't care whether the book is in print or in the form of a PDF (or some other doc format) -- the point is to get a textbook pub out.
Is this purely because of the editorial facilities of publishing companies? Is it really that hard to edit and typeset a document yourself? People do it themselves all the time in academic publications, why not in textbooks?
Take bandwidth away from idiots who download copyrighted movies and music 24/7. Give this bandwidth to people who have legitimate uses for it.
Simple, no?
I'm using the version of Mozilla that came STOCK with Red Hat 8.0 (that is, it's an older version). When I hover over the wacky link, I see "http://www.microsoft.com#" (where the '#' is actually a little box outline).
However, when I click the link, what displays in the address bar is "http://www.microsoft.com%01%00@secunia.com/intern et_explorer_address_bar_spoofing_test/".
So the hacker can spoof the status bar. Big fucking deal -- he can do that already with JavaScript. And even if you DO click the link, all you have to do is glance at the address bar to see you've been fooled.
In other words, there's a bug somewhat like the IE bug in my version of Mozilla, but it isn't the same bug, it has different symptoms, and it seems much less dangerous -- a funny little box appears when you hover over the link, this is clue 1, and the address bar shows exactly where you've really gone to, this is clue 2.
What the hell would it matter anyway? Evil spammers probably also use toothpaste. Does that make everyone who uses toothpaste evil?
The fallacious logic here astounds me. Wait, no it doesn't.