That's exactly the same argument that can be used for outsourcing IT jobs. You can't have it both ways people! You can't have your cheap consumer economy in the US, and still want your jobs protected. Why not complain about the poor music industry jobs that are being "outsourced" to Russia?
NO, this is not like outsourcing.
What this is doing is using the industry's geographical price discrimitation against them.
They might charge $10 for a widget in the US and only $4 in Cambodia, so what's happening here it that the same goods are still being purchased from the same company, it's just the geographic price discrimination is being avoided.
Incidentally price controls like this are illegal in the US, it's just that nobody exists to deal with them on an internaional level. Thus, you can ship a DVD that won't play in Korea, but not one which won't play in Kentucky.
Have you ever seen the cars that people bring to compete in A MOD? A formula one car would demolish the entire field at most autocrosses. 1000 lbs, 900 hp and slicks go a long way towards victory. Even the most stripped miata is still going to be about 1800 lbs with driver. And that assumes no rollcage for autox use only.
And yes, obviously a formula one car will not generate 5g of skidpad at 30 mph, but you shouldnt write off a car just because it generates only 1.5 or 2 g of skidpad at low speeds. Still worlds faster than any car I have seen at an autocross.
Perhaps, but you understand my point. In an autocross, you're be better off without the big wings and insane horsepower in favor of a lighter car. There's not really a "perfect car" out there.
Adding downforce increases the car's "weight" for purposes of calculating the grip of the tires on the road, but doesnt increase the mass of the car that they have to change the direction of.
This is why the "ideal" race car is a stick figure formula 1 type car with a giant engine and huge wings.
What you didn't mention is that all that extra horsepower and downforce is useless at low speeds. (I believe downforce is proportional velocity cubed, so it falls off rather quickly.)
A Formula 1 race car is an idea car for a Formula One race, but I wouldn't be suprised to see one get its ass kicked by a race-prepped Miata at an autocross event with a bunch of hairpin turns.
In the end, the "perfect car" depends on where you're going to be driving it.
Re:Most of the criticisms...
on
The Bugatti Veyron
·
· Score: 2, Interesting
As far as the driving like assholes, that only happens with idiot rockstars and the like who just got a million dollar paycheck. The vast majority of the people driving this car will be doctors, investment bankers and the like. These guys drive carefully.
Hmmmm, where do I begin. The people who are in the market to buy this automobile are not going to drive it every day. In fact, they will probably transport it in its own trailer if they take it to any concours, etc.
Yes, but a car says something about your taste as well. It's like the guy who chooses the most expensive wine in the restraurant vs. the guy who chooses the best. The Veyron has a lot of power, but power isn't free.
Excerpt from an earlier post of mine: This actually has the interesting implication that if you know the fastest you ever want to go and can sustain 1G acceleration up to that point, any additional horsepower is a waste of weight which will detract from the vehicle's braking and cornering performance. (Tire frictional force is nonlinear WRT weight.)
For me, it's about taste. I'd rather have a less powerful car with much better handling than 1000 "look how big my penis is" horsepower. For a true automotive enthusiast, there are other factors that must be considered besides straight line performance.
A good car is a real balancing act. One has to weigh horsepower, aerodynamics, weight balance, moment of intertia, suspension geometry, unsprung mass, etc.
I like to joke, that with enough money you could make a picnic table run a 10 sec. quarter mile, but in the end you've just got a really fast picnic table.
Personally, I'd rather have Bill Gate's Porsche 959 than this monstrosity. It's like the difference between wearing huge gawdy gold chains around your neck vs. wearing a classic and valuable wristwatch.
Apparently the acceleratory (is that a word?) force of this car is so immense, that at full bore you are pulling the same kinds of G-Forces as you do on the vertical drops of a roller coaster ride.
The actual G forces really aren't much more than you can accomplish in a typical street car, it's just that high horsepower cars are able to sustain 1G acceleration for much longer than your typical commuter car.
With the same tires, my car would probably stay neck and neck with this thing up to about 30 MPH, but then my ability to accelerate starts being limited by horsepower instead of tire traction.
This actually has the interesting implication that if you know the fastest you ever want to go and can sustain 1G acceleration up to that point, any additional horsepower is a waste of weight which will detract from the vehicle's braking and cornering performance. (Tire frictional force is nonlinear WRT weight.)
(Note: I'm ignoring downforce.)
Re:Small engine, fast cars but what about airplane
on
The Bugatti Veyron
·
· Score: 1
The question from the orginal conversation was "has anyone used a wenkel rotary (it has a low weight to power ratio) in a plane?" Why/Whynot . ..
It may have only cost you $20 to print it out, but someone also had to license the book, so the total cost is more than $20, although probably not the $100 of a hard-cover textboox.
The situation being discussed was one in which someone already (legally) had a copy of the ebook. Thus, they can legally make one hardcopy.
The MBA program I attended used electronic versions of books a lot. I hated it. A lot of times I wanted to highlight a section or makes notes in the margin. You just can't do this without a real book. Some people printed theirs out. The cost of doing this is ridiculous versus just buying the book in the first place.
What they heck are you using to print them?
A 600 page college textbook is usually around $100.
The 600 page course packet (textbook) for one of my courses cost $20 (the cost of duplication at a local copy shop).
Even with color laser I bet it would be less than $100.
Re:Holy "read the quote you used," Robin
on
Videogames as Art
·
· Score: 0, Troll
To which you reply: "The medium does not decide if something is or is not art." Huh?
LOOK AT WHERE YOU CHOSE TO END THE QUOTE!
Here's the whole section I quoted:
"I'm sorry, but this is contemporary art. As much as I agree with the author's premise that video games is art, his writing style bears no reality on the current status of art as a discipline and offers about as much insight as my grandma would on the state of open source in the computing disciple. Comparing it to hollywood, which has it's own artistic foundation totally removed from the authors writing, is grotesque."
Note how IN ITS ENTIRETY it takes on a much more caustic tone. This guy just points at a few things he likes, states an opinion and starts trashing people and entire industries.
My point is that all Hollywood movies aren't bad and all videogames aren't great. The medium really doesn't decide the quality of the art.
The reason I chose to label this guy a "snob" is because of his willingness to insult others.
The parent's saying this ain't much of a show, not that its premise is wrong.
No he's saying: "his writing style bears no reality on the current status of art as a discipline and offers about as much insight as my grandma would on the state of open source".
A) Don't you think that's a bit over the top?
B) Having a poor "writing style" don't necessarily say that the logical content is bad. It would have been a lot more respectable if he had actually said something RE the points the author was actually trying to make.
There was WAYYY too much venom in that post for me.
I'm sorry, but this is contemporary art.
As much as I agree with the author's premise that video games is art, his writing style bears no reality on the current status of art as a discipline and offers about as much insight as my grandma would on the state of open source in the computing disciple.
Comparing it to hollywood, which has it's own artistic foundation totally removed from the authors writing, is grotesque.
Holy frickin art snobbery batman!
Really, there is a shitload of good art in videogames. In fact ANYTHING CAN BE ART.
What really makes something "art" is the effort and thought put into it. That's why some buildings are "art" while others are not.
A cheap, pre-fabbed home is not art (typically) but somethingdesigned by a guy like IM Pei is.
The medium does not decide if something is or is not art.
The Holocaust Museum in Berlin is an amazing work of art*, but a trailer park is not. The medium is the same, the difference is all in the effort and mastery that was put into their aesthetics.
*So much so that the museum was actually shown before there was even any art in it.
That's the point of registering a trademark, so that no-one can use your name to associate their similar product with a competitor's brand.
That's not the point at all. The point is so that no one else can sell a product with your name on it. It doesn't stop me from "associating" facial tissue with "kleenex", it just stops me from selling my facial tissue AS "kleenex".
I "assosciate" all sorts of trademarked products with their equivalents. If I got to a bar and ask for a Murphy's and they say "How about a Guinness?" they're not violating somebody's trademark.
All the bartender is recognizing what I'm asking for and offering me an alternative. In this case, Google is doing the same.
If google lose this case it would be illegal for me (as a hypothetical Guiness rep.) to pay a bartender to offer someone a Guiness instead, every time they ask for a Murphy's. Clearly this is not what trademark law is about. It exists to protect you from others using you name, not to prevent others from making recommendations when your name is mentioned.
See, I have been earning money by building large scale mail systems for about four years now. I have tried this, with more than one user. It works. It works, because language is redundant. As long as I am sending to people speaking the same language, there is enough vocabulary that is neutral or close to neutral so that I can formulate my message in a way that has a Bayes score in uncritical regions for the vast majority of all users.
1st off my original point was that providing feeback is bad because it makes it possible to tune "Bayes Writers". I never meant to get into an arguement about their pros and cons.
Anyways, what you have to say is intersting, but I'll need a better explaination than redundancy of vocabulary to buy it. I mean sure a world like "has" gets used a lot, but from my current filter:
word spamprob #ham #spam
'has' 0.3033 63 110
It's just not one of those words that pushes a message either way.
Meanwhile, something like my proper zipcode is two orders of manitude away.
word spamprob #ham #spam
'XXXXX' 0.00672646 33 0
It seems to me, that you can bombard my spam filter with "redundant language" and it will work for a small number of messages, but eventually this redundant language with get ranked neutrally. In the end my filter starts working again, because the weight of something like the name of the town I'm in has a weight of 0.00672646 Meanwhile, the weight of a word that's rarely used in my normal correspondence, but is a typical engilsh world like 'extensive' shoots up to 0.762245
Your Bayes works, but it does so only because Spammers are not using a Bayes Writer, and do not yet evaluate Bayes databases on a large scale.
I am by no means claiming that bayesian schemes are perfect, but an attacker is stuck trying to guess which "redundant" english words will get their message ranked well, but if they happen to guess 'internet' it has a ranking of 0.980211 at the moment.
In the end, for a large set of sufficiently populated real filters, I would be quite suprised if you could really get through to 95% of people, and do so consistently. After all, the filters are going to be learning from what you send them.
If you can find a magic message that can get though everybody's filter once, it doesn't really matter. Spammers rely on being able to crank out huge volumes of messages with very little effort. If they need a statistician to work for a week on every spam they send out, it becomes a much more expensive proposition.
You might even be able to come up with a method of pulling my filter around by blasting it with TONS of messages, but every time you did I'd be retraining my filter, and you would probably need an order of magnitude more messages to try the same thing again.
Anyways, I'm not necessarily ruling anything out, and if you have some info on a method which can actually crack 1000 differnt filters with two sigma success I'd be interested.
Sure they are. How much you want to bet that my last name (not part of my email address) has a different ranking in my filter than in yours? Or the proper name of the company I work for? Or the product I work on? All these things are going to be ranked highly in my filter, and are very likely to be used in legitimate email.
I have access to quite a large corpus of messages past and current, and if you use that as a base for a bayesian database, if is actually quite easy to formulate a message that won't score above 5.
For you. The thing about bayesian filtering is that it makes it had to formulate a message that will get through for EVERYONE, even given a source of ham.
And a nice thing about the filtering is that I get to provide feedback. You might be able to get a message though (they do get through ever one in a while), but then I mark it as spam and my filter gets better.
I had a period a while ago where I was getting lots of spam with random sections of prose in the body. Initially, it got through, but I marked it as spam and the filter re-trained itself.
I think you've got this wrong. They are suing because Google is selling their name as an Adword, not that their competitor comes up in the search portion of the page. It seems like there could be a good case that the competitor (and Google) is making money by trading on their good name. I'm not saying that I agree, but I don't think it is an open-and-shut case, especially when the name isn't a common english or french word.
I don't really think AXA has a chance.
This like me paying google so that when you search for "dell", an ad for my computer store comes up on the side.
I'm not claiming to be Dell, but I sell a competing product. All I'm doing is recognizing that by searching for "dell" you may be looking to buy a computer and taking advantage of that knowedge.
It seems pretty clear to me that AXA is trying to abuse trademark law to do something that it wasn't meant to do. Trademarks are supposed to prevent someone from claiming to be you. I can't go sell "Lee" jeans. They do not prevent me from talking about "Lee" jeans, even if I'm a direct competitior.
I initially thought their case might have merit, but after think about it I realized that: Google isn't banking on AXA's name, they're using the correlation between the AXA name and a particular type of product.
While the search term itself is trademarked, Google is not using the word in anything but an informational sense. In essence they're saying:
This name links to this type of product, so a maker of that type of product has paid us to be assosciated with it in our database.
If I type "kleenex" in a box on a webpage, there's no good reason the results can't include the generic equivalent, just also long as the generic equivlent doesn't actually claim to be "kleenex". I have no obligation to provide you contact information for "kleenex", just because they have a tradmark on it. So just as I don't claim to be making "kleenex", I'm in the clear.
That report is interesting but also makes my point.
The guy doing the attack had access to the "ham" not just the "spam".
He provides a list of major spam cues and major ham cues. They key thing is that these are different for everyone.
What makes it through your filter may not make it through mine, one of the key things about bayesian filtering is that everybody's filter is unique.
I might have a friend named "Sally" or I might not. I might do a lot fo work with optics, so the word "refraction" might be a big help.
The fact the you do not know the words in my filter and do not get feedback is what makes the bayesian concept work. It has other weaknesses, but making your ham and spam public or providing feedback totally kills the concept.
The bottom line is that the study you linked to doesn't really show much of a problem with bayesian filtering, and kind of proves my point. He was able, with a significant amount of work, to get one message through one specfic filter. He was able to do it and know he was successful because he had feedback availible.
(My point is that giving feedback makes it easy to defeat bayesian filtering. I'm not really sure what you're trying to say since that article does more to reinforce my point than refute it.)
The filter is public anyway. If you send spam professionally, you are expected to know SpamAssassin and tune your spam so that it scores low.
The filter is NOT public, particularly any bayesian component of it. For example, right now there is no way for you to know what ranking my copy of spambayes is giving the word "foobar".
If I start giving you feedback, you can use this to being correlating words WRT accept vs. reject.
And even if an ID is forged, as long as it is expensive to forge, most criminals will have few of them, and losing or exposing one of their IDs will be a heavy loss for them. A forged ID may to reveal the identity of a criminal, but it will still create a traceable and linkable trail. Which is what really counts when you try to catch such people.
Except that in the US, because of our stupid alcohol laws, there is an absoluetly HUGE market for fake ids. There's serious money to be made on any college campus. In essence our silly 21 drinking age is subsidizing fake ids for terrorists.
If the drinking age were reduced to a more reasonable age (like say the same age that I had to send in my damned DRAFT CARD), the market for fake IDs would be vastly decreased, making it much harder to get them. Ages below 18 would also have significantly less money to spend.
Right now, the setup costs for your fake ID machine can be amortized across hundreds of people on any given college campus.
In essence, the US has broken it's own ID system, by creating too much incentive to do so.
The problem with SpamAssassin is that it mixes up predelivery checks and postdelivery checks. It would be worth the effort to extract all predelivery checks from SpamAssassin (DNSBL checks, mostly), throw in Milter Sender like checks [snert.com] and create a predelivery milter-sender Spamasssassin which would catch most of the Spam in transit and reject it with fivehundreds.
The problem with that is it allows me to get feedback on what your system is and is not rejecting. This allows me to tune my spam generation filter to counter your spam detection filter.
original spam=>spam generation filter=>spam detection filter=>you inbox
The messages are the feedback signal that I could use to adjust my filter until I have the inverse of yours. (Within limits. Blacklists would still work, but bayesian filterting wouldn't.)
Again, I say: just don't buy them. Ripping them off only proves that 1) you want the music and 2) you're not willing to pay for it.
Or 3) You're perfectly willing to pay for music, but you know that by giving the RIAA money, you're making the situation worse for both artists and fans.
On that note (oh, i'm so punny..), I'm washing my hands of this moronic discussion.
It wouldn't be so moronic if you would actually read and comprehend what I'm saying. Rather than attempting to argue with option 3, you just ignore it and claim it doesn't exist. That doesn't leave you much room to say anything insightful. It's like a little kid sticking their fingers in their ears and going LALALALA!
nothing quite like legitimizing your enemy's corrupt activity by participating in your own corrupt activity
It's more like Chris Rock's sketch where he says: "I'm not saying he killed her, but I understand."
Funny thing about monopolies with a product nobody needs... it's still a totally voluntary purchase.
Nobody NEEDS music. It's not a basic need. Neither is that computer you're sitting in front of.
Despite the fact that you are totally clueless (shouldn't you be watching American Idol reruns about now?), I don't think even YOUR vapid, pointless existance (how else could you explain it if you think that living without a Foghat CD is too painful to even consider) wouldn't come to an end if you just helped those of us who actually THOUGHT about it put them out of business by NOT BUYING FROM THEM.
Sounds like I've got a true intellectual on my hands. READ
Oops, you still haven't overcome that one funky little problem that blows away all of your claims that your activity is anything more than the boring antics of a common criminal: you could just stop buying them.
And you could just stop using your computer. You don't NEED it.
What an idiot I am for suggesting that the real problem is dumb consumers willing to pay a hyperinflated market price.
Yes. You have a choice between blaming a market-raping corporate cartel and 13 year-old Brittney-listeners.
Why, yes. I am slinging personal insults at you now, mainly because I think you're just a yipping little kid who doesn't want to play fair anymore than the RIAA does.
Because the only thing that's fair is playing by rules the RIAA has bought and paid for?
In college I remember taking the final and 10 guys spent all of 10 minutes taking it. They had programmed their calculators with every possible question
This reminds me of something I always wanted to do back in college:
Walk into an extremly difficult test for a course I'm not in (Advanced Thermodynamics or some such thing), sit down, doodle on the test for ten minutes, and walk our with a really smug look on my face. (Maybe even cough and say "easy" under my breath.)
Everyone would be going: "Damn! Who is that guy!"
Not knowing that I didn't have a damn clue what I was doing on the test.
They'd all hate me, and for some reason that sounds kinda fun.
I use to be head proctor for the placement exams for the Engineering school at Cornell. The year calculators were added to the Calculus AP, we saw a statistically significant drop in scores. However, when I complain about these problems, I get called a technophobe.
As a recent CU alum (ECE '03), I'm pretty much amazed that calculators are allowed on the placement test. They aren't allowed on the tests for any of the math courses the students are placing into.
God, how I hate these things.
As an engineer or a student, I love them. Sure I could have done all the math by hand, but my TI-89 saved me hours every week when I was a student.
The reality is, students will have access to calulators out in the "real world".... and that's a good thing. I like the way Cornell does things in their actual courses:
-For pure math courses (Math 191,192,293,294) calculators are allowed on the homework but not the test. This ensures I actually learn what the integral of sin(x) is.
-For engineering courses calculators are allowed (Why should students be trying to row reduce matrices during a 40 minute test that's supposed to be testing their knowedge of something else?)
-Nobody ever cared what calculator I had, and you picked up how to use it on your own.
Also, it's worth noting that the drop in scores may not have to do with the use of calculators. NYS has been yanking tests and requirements all over the place in recent years. Personally, I scored fairly poorly when I took my placement test and it meant zilch. I went right into Math 192 anyways and did just fine. I was just a little out of practice when I took the test.
One thing I wish Cornell would change is the "magic question" on all their math exams. Every exam was pretty much the same format:
Five questions: Four sensible questions, and a fifth question designed spefically to keep the mean down.
This question was always beyond the scope of the material covered in the course up to that point, and the only way you could solve it was a sudden burst of inspiriation, or learning it outside the course. I remember one specfic occasion where this "magic question" was actually coved in a section of the textbook that we were specfically told we didn't have to read.
It was a dirty trick, plain and simple and based on the silly belief that even though you're teaching an intro calc. course to students whose abilities are in the top percentiles of students, the mean should be a B.
By adding that question, the kept the scores where they wanted them.
Slashdot Mirror
That's exactly the same argument that can be used for outsourcing IT jobs. You can't have it both ways people! You can't have your cheap consumer economy in the US, and still want your jobs protected. Why not complain about the poor music industry jobs that are being "outsourced" to Russia?
NO, this is not like outsourcing.
What this is doing is using the industry's geographical price discrimitation against them.
They might charge $10 for a widget in the US and only $4 in Cambodia, so what's happening here it that the same goods are still being purchased from the same company, it's just the geographic price discrimination is being avoided.
Incidentally price controls like this are illegal in the US, it's just that nobody exists to deal with them on an internaional level. Thus, you can ship a DVD that won't play in Korea, but not one which won't play in Kentucky.
Have you ever seen the cars that people bring to compete in A MOD? A formula one car would demolish the entire field at most autocrosses. 1000 lbs, 900 hp and slicks go a long way towards victory. Even the most stripped miata is still going to be about 1800 lbs with driver. And that assumes no rollcage for autox use only. And yes, obviously a formula one car will not generate 5g of skidpad at 30 mph, but you shouldnt write off a car just because it generates only 1.5 or 2 g of skidpad at low speeds. Still worlds faster than any car I have seen at an autocross.
Perhaps, but you understand my point. In an autocross, you're be better off without the big wings and insane horsepower in favor of a lighter car. There's not really a "perfect car" out there.
Adding downforce increases the car's "weight" for purposes of calculating the grip of the tires on the road, but doesnt increase the mass of the car that they have to change the direction of.
This is why the "ideal" race car is a stick figure formula 1 type car with a giant engine and huge wings.
What you didn't mention is that all that extra horsepower and downforce is useless at low speeds. (I believe downforce is proportional velocity cubed, so it falls off rather quickly.)
A Formula 1 race car is an idea car for a Formula One race, but I wouldn't be suprised to see one get its ass kicked by a race-prepped Miata at an autocross event with a bunch of hairpin turns.
In the end, the "perfect car" depends on where you're going to be driving it.
As far as the driving like assholes, that only happens with idiot rockstars and the like who just got a million dollar paycheck. The vast majority of the people driving this car will be doctors, investment bankers and the like. These guys drive carefully.
You mean like Bill Gates was doing when he got arrested for reckless driving?
Hmmmm, where do I begin. The people who are in the market to buy this automobile are not going to drive it every day. In fact, they will probably transport it in its own trailer if they take it to any concours, etc.
Yes, but a car says something about your taste as well. It's like the guy who chooses the most expensive wine in the restraurant vs. the guy who chooses the best. The Veyron has a lot of power, but power isn't free.
Excerpt from an earlier post of mine:
This actually has the interesting implication that if you know the fastest you ever want to go and can sustain 1G acceleration up to that point, any additional horsepower is a waste of weight which will detract from the vehicle's braking and cornering performance. (Tire frictional force is nonlinear WRT weight.)
For me, it's about taste. I'd rather have a less powerful car with much better handling than 1000 "look how big my penis is" horsepower. For a true automotive enthusiast, there are other factors that must be considered besides straight line performance.
A good car is a real balancing act. One has to weigh horsepower, aerodynamics, weight balance, moment of intertia, suspension geometry, unsprung mass, etc.
I like to joke, that with enough money you could make a picnic table run a 10 sec. quarter mile, but in the end you've just got a really fast picnic table.
Personally, I'd rather have Bill Gate's Porsche 959 than this monstrosity. It's like the difference between wearing huge gawdy gold chains around your neck vs. wearing a classic and valuable wristwatch.
Apparently the acceleratory (is that a word?) force of this car is so immense, that at full bore you are pulling the same kinds of G-Forces as you do on the vertical drops of a roller coaster ride.
The actual G forces really aren't much more than you can accomplish in a typical street car, it's just that high horsepower cars are able to sustain 1G acceleration for much longer than your typical commuter car.
With the same tires, my car would probably stay neck and neck with this thing up to about 30 MPH, but then my ability to accelerate starts being limited by horsepower instead of tire traction.
This actually has the interesting implication that if you know the fastest you ever want to go and can sustain 1G acceleration up to that point, any additional horsepower is a waste of weight which will detract from the vehicle's braking and cornering performance. (Tire frictional force is nonlinear WRT weight.)
(Note: I'm ignoring downforce.)
The question from the orginal conversation was "has anyone used a wenkel rotary (it has a low weight to power ratio) in a plane?" Why/Whynot . . .
They have.
OT: My '86 RX-7 is a blast to drive. I wish they still made them.
It may have only cost you $20 to print it out, but someone also had to license the book, so the total cost is more than $20, although probably not the $100 of a hard-cover textboox.
The situation being discussed was one in which someone already (legally) had a copy of the ebook. Thus, they can legally make one hardcopy.
The MBA program I attended used electronic versions of books a lot. I hated it. A lot of times I wanted to highlight a section or makes notes in the margin. You just can't do this without a real book. Some people printed theirs out. The cost of doing this is ridiculous versus just buying the book in the first place.
What they heck are you using to print them?
A 600 page college textbook is usually around $100.
The 600 page course packet (textbook) for one of my courses cost $20 (the cost of duplication at a local copy shop).
Even with color laser I bet it would be less than $100.
To which you reply: "The medium does not decide if something is or is not art." Huh?
LOOK AT WHERE YOU CHOSE TO END THE QUOTE!
Here's the whole section I quoted:
"I'm sorry, but this is contemporary art. As much as I agree with the author's premise that video games is art, his writing style bears no reality on the current status of art as a discipline and offers about as much insight as my grandma would on the state of open source in the computing disciple. Comparing it to hollywood, which has it's own artistic foundation totally removed from the authors writing, is grotesque."
Note how IN ITS ENTIRETY it takes on a much more caustic tone. This guy just points at a few things he likes, states an opinion and starts trashing people and entire industries.
My point is that all Hollywood movies aren't bad and all videogames aren't great. The medium really doesn't decide the quality of the art.
The reason I chose to label this guy a "snob" is because of his willingness to insult others.
The parent's saying this ain't much of a show, not that its premise is wrong.
No he's saying: "his writing style bears no reality on the current status of art as a discipline and offers about as much insight as my grandma would on the state of open source".
A) Don't you think that's a bit over the top?
B) Having a poor "writing style" don't necessarily say that the logical content is bad. It would have been a lot more respectable if he had actually said something RE the points the author was actually trying to make.
There was WAYYY too much venom in that post for me.
I'm sorry, but this is contemporary art. As much as I agree with the author's premise that video games is art, his writing style bears no reality on the current status of art as a discipline and offers about as much insight as my grandma would on the state of open source in the computing disciple. Comparing it to hollywood, which has it's own artistic foundation totally removed from the authors writing, is grotesque.
Holy frickin art snobbery batman!
Really, there is a shitload of good art in videogames. In fact ANYTHING CAN BE ART.
What really makes something "art" is the effort and thought put into it. That's why some buildings are "art" while others are not.
A cheap, pre-fabbed home is not art (typically) but something designed by a guy like IM Pei is.
The medium does not decide if something is or is not art.
The Holocaust Museum in Berlin is an amazing work of art*, but a trailer park is not. The medium is the same, the difference is all in the effort and mastery that was put into their aesthetics.
*So much so that the museum was actually shown before there was even any art in it.
That's the point of registering a trademark, so that no-one can use your name to associate their similar product with a competitor's brand.
That's not the point at all. The point is so that no one else can sell a product with your name on it. It doesn't stop me from "associating" facial tissue with "kleenex", it just stops me from selling my facial tissue AS "kleenex".
I "assosciate" all sorts of trademarked products with their equivalents. If I got to a bar and ask for a Murphy's and they say "How about a Guinness?" they're not violating somebody's trademark.
All the bartender is recognizing what I'm asking for and offering me an alternative. In this case, Google is doing the same.
If google lose this case it would be illegal for me (as a hypothetical Guiness rep.) to pay a bartender to offer someone a Guiness instead, every time they ask for a Murphy's.
Clearly this is not what trademark law is about. It exists to protect you from others using you name, not to prevent others from making recommendations when your name is mentioned.
1st off my original point was that providing feeback is bad because it makes it possible to tune "Bayes Writers". I never meant to get into an arguement about their pros and cons.
Anyways, what you have to say is intersting, but I'll need a better explaination than redundancy of vocabulary to buy it. I mean sure a world like "has" gets used a lot, but from my current filter:
It's just not one of those words that pushes a message either way.
Meanwhile, something like my proper zipcode is two orders of manitude away.
It seems to me, that you can bombard my spam filter with "redundant language" and it will work for a small number of messages, but eventually this redundant language with get ranked neutrally. In the end my filter starts working again, because the weight of something like the name of the town I'm in has a weight of 0.00672646 Meanwhile, the weight of a word that's rarely used in my normal correspondence, but is a typical engilsh world like 'extensive' shoots up to 0.762245
Your Bayes works, but it does so only because Spammers are not using a Bayes Writer, and do not yet evaluate Bayes databases on a large scale.
I am by no means claiming that bayesian schemes are perfect, but an attacker is stuck trying to guess which "redundant" english words will get their message ranked well, but if they happen to guess 'internet' it has a ranking of 0.980211 at the moment.
In the end, for a large set of sufficiently populated real filters, I would be quite suprised if you could really get through to 95% of people, and do so consistently. After all, the filters are going to be learning from what you send them.
If you can find a magic message that can get though everybody's filter once, it doesn't really matter. Spammers rely on being able to crank out huge volumes of messages with very little effort. If they need a statistician to work for a week on every spam they send out, it becomes a much more expensive proposition.
You might even be able to come up with a method of pulling my filter around by blasting it with TONS of messages, but every time you did I'd be retraining my filter, and you would probably need an order of magnitude more messages to try the same thing again.
Anyways, I'm not necessarily ruling anything out, and if you have some info on a method which can actually crack 1000 differnt filters with two sigma success I'd be interested.
Actually, they aren't that different.
Sure they are. How much you want to bet that my last name (not part of my email address) has a different ranking in my filter than in yours? Or the proper name of the company I work for? Or the product I work on? All these things are going to be ranked highly in my filter, and are very likely to be used in legitimate email.
I have access to quite a large corpus of messages past and current, and if you use that as a base for a bayesian database, if is actually quite easy to formulate a message that won't score above 5.
For you. The thing about bayesian filtering is that it makes it had to formulate a message that will get through for EVERYONE, even given a source of ham.
And a nice thing about the filtering is that I get to provide feedback. You might be able to get a message though (they do get through ever one in a while), but then I mark it as spam and my filter gets better.
I had a period a while ago where I was getting lots of spam with random sections of prose in the body. Initially, it got through, but I marked it as spam and the filter re-trained itself.
Lawyers don't sue people. People sue people.
There's a saying that goes:
One laywer in a small town can go broke, but two lawyers in a small town can make a killing.
One of the key sentiments you're supposed to take away from that is lawyers (at least financially) do not have the public's best interest at heart.
They only profit when there's conflict, whether it's warranted or not.
I think you've got this wrong. They are suing because Google is selling their name as an Adword, not that their competitor comes up in the search portion of the page. It seems like there could be a good case that the competitor (and Google) is making money by trading on their good name. I'm not saying that I agree, but I don't think it is an open-and-shut case, especially when the name isn't a common english or french word.
I don't really think AXA has a chance.
This like me paying google so that when you search for "dell", an ad for my computer store comes up on the side.
I'm not claiming to be Dell, but I sell a competing product. All I'm doing is recognizing that by searching for "dell" you may be looking to buy a computer and taking advantage of that knowedge.
It seems pretty clear to me that AXA is trying to abuse trademark law to do something that it wasn't meant to do. Trademarks are supposed to prevent someone from claiming to be you. I can't go sell "Lee" jeans. They do not prevent me from talking about "Lee" jeans, even if I'm a direct competitior.
I initially thought their case might have merit, but after think about it I realized that:
Google isn't banking on AXA's name, they're using the correlation between the AXA name and a particular type of product.
While the search term itself is trademarked, Google is not using the word in anything but an informational sense. In essence they're saying:
This name links to this type of product, so a maker of that type of product has paid us to be assosciated with it in our database.
If I type "kleenex" in a box on a webpage, there's no good reason the results can't include the generic equivalent, just also long as the generic equivlent doesn't actually claim to be "kleenex". I have no obligation to provide you contact information for "kleenex", just because they have a tradmark on it. So just as I don't claim to be making "kleenex", I'm in the clear.
That report is interesting but also makes my point.
The guy doing the attack had access to the "ham" not just the "spam".
He provides a list of major spam cues and major ham cues. They key thing is that these are different for everyone.
What makes it through your filter may not make it through mine, one of the key things about bayesian filtering is that everybody's filter is unique.
I might have a friend named "Sally" or I might not. I might do a lot fo work with optics, so the word "refraction" might be a big help.
The fact the you do not know the words in my filter and do not get feedback is what makes the bayesian concept work. It has other weaknesses, but making your ham and spam public or providing feedback totally kills the concept.
The bottom line is that the study you linked to doesn't really show much of a problem with bayesian filtering, and kind of proves my point. He was able, with a significant amount of work, to get one message through one specfic filter. He was able to do it and know he was successful because he had feedback availible.
(My point is that giving feedback makes it easy to defeat bayesian filtering. I'm not really sure what you're trying to say since that article does more to reinforce my point than refute it.)
The filter is public anyway. If you send spam professionally, you are expected to know SpamAssassin and tune your spam so that it scores low.
The filter is NOT public, particularly any bayesian component of it. For example, right now there is no way for you to know what ranking my copy of spambayes is giving the word "foobar".
If I start giving you feedback, you can use this to being correlating words WRT accept vs. reject.
Name one thing the Government (any government) does well?
Hmmm....I dunno, how about mail?
I've had cheap, reliable mail service for the entire time I've been alive.
Oh....wait, and roads too. I drove on roads to work today.
Oh.....and what about the FRICKIN INTERNET!
The government does tons of things right every day, it's just that you don't notice.
I'm not saying the the gov't doesn't mess up or has the solution for this particular problem, buy saying the gov't is worthless is just plain stupid.
I don't know about you, but I'd rather not have to walk around with a gun all day and grow my own food. Anarchy sucks.
And even if an ID is forged, as long as it is expensive to forge, most criminals will have few of them, and losing or exposing one of their IDs will be a heavy loss for them. A forged ID may to reveal the identity of a criminal, but it will still create a traceable and linkable trail. Which is what really counts when you try to catch such people.
Except that in the US, because of our stupid alcohol laws, there is an absoluetly HUGE market for fake ids. There's serious money to be made on any college campus. In essence our silly 21 drinking age is subsidizing fake ids for terrorists.
If the drinking age were reduced to a more reasonable age (like say the same age that I had to send in my damned DRAFT CARD), the market for fake IDs would be vastly decreased, making it much harder to get them. Ages below 18 would also have significantly less money to spend.
Right now, the setup costs for your fake ID machine can be amortized across hundreds of people on any given college campus.
In essence, the US has broken it's own ID system, by creating too much incentive to do so.
The problem with SpamAssassin is that it mixes up predelivery checks and postdelivery checks. It would be worth the effort to extract all predelivery checks from SpamAssassin (DNSBL checks, mostly), throw in Milter Sender like checks [snert.com] and create a predelivery milter-sender Spamasssassin which would catch most of the Spam in transit and reject it with fivehundreds.
The problem with that is it allows me to get feedback on what your system is and is not rejecting. This allows me to tune my spam generation filter to counter your spam detection filter.
original spam=>spam generation filter=>spam detection filter=>you inbox
The messages are the feedback signal that I could use to adjust my filter until I have the inverse of yours. (Within limits. Blacklists would still work, but bayesian filterting wouldn't.)
Okey... I'm home now and wanted to read few stories. Could everyone please not visit any articles for the next hour or so?
Thanks in advance, Gunnar
Well, I don't think you're going to be able to get people not to visit the articles, but I think it's a fairly safe bet they won't read them.
Funny how that works, isn't it? It seems like we can bring a website to its knees, but still no one will have RTFA.
Maybe everyone posts and then reads the article?.....[scratches head]
Again, I say: just don't buy them. Ripping them off only proves that 1) you want the music and 2) you're not willing to pay for it.
Or 3) You're perfectly willing to pay for music, but you know that by giving the RIAA money, you're making the situation worse for both artists and fans.
On that note (oh, i'm so punny..), I'm washing my hands of this moronic discussion.
It wouldn't be so moronic if you would actually read and comprehend what I'm saying. Rather than attempting to argue with option 3, you just ignore it and claim it doesn't exist. That doesn't leave you much room to say anything insightful. It's like a little kid sticking their fingers in their ears and going LALALALA!
nothing quite like legitimizing your enemy's corrupt activity by participating in your own corrupt activity
It's more like Chris Rock's sketch where he says:
"I'm not saying he killed her, but I understand."
Funny thing about monopolies with a product nobody needs... it's still a totally voluntary purchase.
Nobody NEEDS music. It's not a basic need. Neither is that computer you're sitting in front of.
Despite the fact that you are totally clueless (shouldn't you be watching American Idol reruns about now?), I don't think even YOUR vapid, pointless existance (how else could you explain it if you think that living without a Foghat CD is too painful to even consider) wouldn't come to an end if you just helped those of us who actually THOUGHT about it put them out of business by NOT BUYING FROM THEM.
Sounds like I've got a true intellectual on my hands. READ
Oops, you still haven't overcome that one funky little problem that blows away all of your claims that your activity is anything more than the boring antics of a common criminal: you could just stop buying them.
And you could just stop using your computer. You don't NEED it.
What an idiot I am for suggesting that the real problem is dumb consumers willing to pay a hyperinflated market price.
Yes. You have a choice between blaming a market-raping corporate cartel and 13 year-old Brittney-listeners.
Why, yes. I am slinging personal insults at you now, mainly because I think you're just a yipping little kid who doesn't want to play fair anymore than the RIAA does.
Because the only thing that's fair is playing by rules the RIAA has bought and paid for?
"The people's good is the highest law." -Cicero
In college I remember taking the final and 10 guys spent all of 10 minutes taking it. They had programmed their calculators with every possible question
This reminds me of something I always wanted to do back in college:
Walk into an extremly difficult test for a course I'm not in (Advanced Thermodynamics or some such thing), sit down, doodle on the test for ten minutes, and walk our with a really smug look on my face. (Maybe even cough and say "easy" under my breath.)
Everyone would be going: "Damn! Who is that guy!"
Not knowing that I didn't have a damn clue what I was doing on the test.
They'd all hate me, and for some reason that sounds kinda fun.
I use to be head proctor for the placement exams for the Engineering school at Cornell. The year calculators were added to the Calculus AP, we saw a statistically significant drop in scores. However, when I complain about these problems, I get called a technophobe.
As a recent CU alum (ECE '03), I'm pretty much amazed that calculators are allowed on the placement test. They aren't allowed on the tests for any of the math courses the students are placing into.
God, how I hate these things.
As an engineer or a student, I love them. Sure I could have done all the math by hand, but my TI-89 saved me hours every week when I was a student.
The reality is, students will have access to calulators out in the "real world".... and that's a good thing. I like the way Cornell does things in their actual courses:
-For pure math courses (Math 191,192,293,294) calculators are allowed on the homework but not the test. This ensures I actually learn what the integral of sin(x) is.
-For engineering courses calculators are allowed (Why should students be trying to row reduce matrices during a 40 minute test that's supposed to be testing their knowedge of something else?)
-Nobody ever cared what calculator I had, and you picked up how to use it on your own.
Also, it's worth noting that the drop in scores may not have to do with the use of calculators. NYS has been yanking tests and requirements all over the place in recent years. Personally, I scored fairly poorly when I took my placement test and it meant zilch. I went right into Math 192 anyways and did just fine. I was just a little out of practice when I took the test.
One thing I wish Cornell would change is the "magic question" on all their math exams. Every exam was pretty much the same format:
Five questions: Four sensible questions, and a fifth question designed spefically to keep the mean down.
This question was always beyond the scope of the material covered in the course up to that point, and the only way you could solve it was a sudden burst of inspiriation, or learning it outside the course. I remember one specfic occasion where this "magic question" was actually coved in a section of the textbook that we were specfically told we didn't have to read.
It was a dirty trick, plain and simple and based on the silly belief that even though you're teaching an intro calc. course to students whose abilities are in the top percentiles of students, the mean should be a B.
By adding that question, the kept the scores where they wanted them.