Re:How long before people start gaming the system?
on
IBM vs. Content Chaos
·
· Score: 1
First, the "hardness" needs to be measured against the value of the benefit obtained from gaming. If it's large, more effort will be thrown at it.
Second, you seem to have missed the implications of my carefully-chosen word simulate. You don't need to replicate the algorithm, just create something that mostly works in most of the situations that you care about. (Both "mosts" are important.) This is a significantly lower bar then "complete replication", and is one of the reasons it's so hard to combat this; while you need to get to 99.9% accuracy to be a "good search engine", the engine gamers can quite often get by on a simulation that is hardly correct at all... I'd guess the Google spammers understand at most 1% of the Google algorithm... but that's plenty to game the system, often because they can make up the difference with "overwhelming force". (The google spammers aren't going to sweat the difference between a ranking of 5.554 and 5.556 when they can just create another 20 sites that link to the desired site; "overwhelming force".) The search engines are fighting on a very unbalanced-in-the-gamer's-favor battlefield.
Again, don't overread my claim. I'm not saying this is inevitable; Google largely wins overall. But the possibility should not be waved away; it's a big mistake to do so. (Especially for the implementers; if you think your algorithm is perfect you're likely to learn how wrong you are in a big way as you bet too much on it, whereas I doubt the people at Google ever thought they'd ever completely "win", so they make much more reasonable choices.)
No, the body will flush it through normal processes, just like gum doesn't sit in the body for years and years. Your digestive system is a pipe; there's nowhere for something to just "hang out" for years.
This is true and it's easy to prove: A penny is much larger, completely indigestable, and will invariably come out. The only danger of eating a single penny is that some part of the "pipe" may be too small to accomodate it, in which case it could be fatal as it clogs the whole system up. As styrofoam has much more give, this does not apply. If you can handle a penny, you can handle syrofoam.
Shit-eating is not just "not fun", it's actively and supremely dangerous, with a good possibility of death without (and possibly with) medical attention. Among other nasty organisms, it tends to harbor e coli in dangerous quantities.
This is why you have not seen a "shit eating" contest on "Fear Factor"; it's not safe at all.
I doubt Styrofoam would cause much more then mild discomfort, if that. Shit stands a good chance to kill you. I know which I'd choose... Styrofoam would be a better bet, by a long shot.
"Natural" is a meaningless term in determining toxicity; botulism toxin is natural, cobra venom is natural, all toxic houseplants are natural, for that matter e coli and ebola are 100% all natural too. I'd rather eat 100 grams 100% artificial Vitamin C then 100 grams of botulism toxin.
Re:How long before people start gaming the system?
on
IBM vs. Content Chaos
·
· Score: 2, Informative
It's important not to underestimate people's ability to game systems, regardless of the thought put into them. The simple algorithm
Reconstruct algorithm.
Simulate algorithm and play with the inputs until the outputs match what you want.
Bring those inputs about.
is extremely powerful, and note that as a "meta-algorithm" there's absolutely no way to completely shut it down.
You have only four basic defenses against this:
Keep changing the algorithm (expensive and large changes may not be possible if stability is desirable, which for search results it generally is),
make the input-gaming process more expensive then the value of the output to the attacker (as you become more valuable you're a more enticing target),
make the outputs desired by the attacked impossible (generally not possible in the general case, but in certain limited ways it is; it is probably not possible to be the #1 google result for all possible search terms, for instance, despite the desirability of such a result),
or have a human monitoring attacks and shut them down manually (only possible if you can out-staff the attackers)
There are some other possibilities but a lot of them don't apply in the real world, like "make it impossible to reverse the inputs necessary for some output" (like MD5); this is not applicable to a real-world application like a search engine because there has to be some obvious human-sensible logic to the placement or the search engine is just returning random results, which is not even a "search engine", let alone a useful one. Not even all four can be brought to bear in a given situation; #2 probably doesn't apply in this case since the benefits could be in the millions of dollars in theory.
I'm not saying that WebFountain is hosed; Google has trouble but it is handlable. But it is worth talking about; certain basic algorithms will have certain effects as people try to game them, and it may be the case that some clever, useful algorithm is so easily gamed and so difficult to create countermeasures for that it will never be possible in the real world in the general case.
(I doubt this is the case, but there's only one way to find out, and that's try it and see what happens.)
I neither insulted you nor hinted that you don't know what you are talking about.
Surely not, since this is the first I've posted to the thread.
"You're wrong, you're talking nonsense, and you don't even know it" isn't condenscending if it's true.
If it's possible, then we can do it.
So what? Penetrating the Heisenberg Uncertainty Principle isn't possible, even in principle.
I'm more a student of humanity than science. And I've seen it more than once, the common belief that we can't do something, only to find out later ( days, years, centuries ) that it is possible.
Unfortunately, your misconception is not neutral, which is why I post. (And remember, you're not the only reader; I realize I don't stand a chance in hell of convincing you, because you're so certain you're right you have the confidence to post on the topic. Yes, the equivalent logic applies equally to me.) Unrealistic expectations fuel Luddite tendencies and fuel unrealistic extrapolations into the future.
Unfortunately, when acted on, these cause suboptimal results, sometimes even killing people. Consider the effects of unrealistic expectations on the environmentalism issues; what if we act as if you're right and always assume there's a way to clean up our environmental messes before they kill us, when the science says we're wrong?
And I've seen it more than once, the common belief that we can't do something, only to find out later ( days, years, centuries ) that it is possible.
Somewhat ironically, it shows that you're a student of ancient history but have not been keeping up to date on recent history. History doesn't support the idea that we're going to find a way past the Heisenberg Uncertainty Principle. History has in fact shown the increasing understanding and refinement of boundaries and limits, ever since the first ones were discovered.
You should not be surprised when people don't buy your Argument from Ignorance, or call it for what it is. The arrogance lies in your thinking that your knowlege, despite its nonexistance (I note you could not possibly have studied the Halting Problem or Godel's Incompleteness Theorum in the time period since your last post), is so much greater then scientists and mathematicians that you can contradict them on the fundamentals of the universe. That is arrogance, that is inflated self-assessment.
Of course that doesn't mean they're always right; but they don't have to be to still be right about certain limits.
This is your cue (especially due to my previous parenthetical note) to now go find some cursory overview of Godel's Theorem, read it in two minutes, and explain why one of the greatest mathematical acheivements in the history of man that proves certain things are impossible doesn't prove that certain things are impossible. Here, let me help with that first part. OK, this is a little condenscending, but can I just say "been here, done this?" Have fun.
I Am Not A Materials Scientist... but doesn't plastic's structure take permenent damage from deformation on this level? Even the strongest plastic key takes damage from each other key it destroys, so eventually it too will die.
Thus, there is no strongest plastic key; they will all be destroyed in short order. In fact, while you might superstitiously keep your "superkey" that broke 20 others, statistically speaking the ideal strategy (in terms of maximal wins) is to use a key precisely once and throw it away. Your superkey is more and more likely to break on each try.
If I'm wrong please correct me, or if you know the terminology please let me know.
This is perhaps the single most pointless fad ever, and elitist comments to the contrary, I would expect that the general public, even the young kids general public, is too smart to be interested in this as more then a passing fad.
Strict weather prediction will never happen; see the sibling post to your own. QM actually prevents it, believe it or not.
What could conceivably happen is that we start manipulating weather on a large scale, and we might learn how to bend weather to our will. We'd need essentially random corrections due to the forces of chaos, but conceivably with enough control, we could say "It will rain 3 inches on this site three years in the future" (with the implicit assumption the weather control grid will still be working, i.e., no major nuclear war, no nearby supernovas, etc.).
But that's not prediction, that's control, and there's a big difference. The unpredictability of the system would still manifest itself as a complete inability to predict in advance what inputs to the system would be necessary to maintain the states we desire; we'd have to correct dynamically and in the short-term. So, even this doesn't solve the "predictability" problem, it just pushes it out one meta-level; the fundamental unpredictability remains.
Seriously though, we may not be able to imagine how it will work, or the solutions we can imagine don't work at all, but I'm confident it will happen, both for earthquakes and weather and anything else overly complex. Note that I did not say sometime soon, although I would like to see that too, I understand the technology and science we need just isn't up to par yet.
"Science" has proven that it can't work. Making those things work requires that the impossible be done. Arguments of the form "If an impossible thing happens, another impossible thing can happen" are trivially logically true, but not relevant in the real world.
Before you continue to assert how I will eventually be "proven wrong by the unbounded and unimaginable progress of humanity!!!1!!", please study the computer science concept of reduction; any solution to the weather prediction problem reduces to a method to penetrate the Heisenberg Uncertainty Principle fog, which would cause the complete collapse of particle physics as we know them (and remember, advances historically speaking refine past theories, not destroy them). If you still believe at that point that we might get past it, at least then you'll have some vague glimmering of the magnitude of power you are claiming we can obtain; I get no sense that you realize how scientifically and mathematically silly you're being from your current messages.
While you're at it, might want to study Godel's Incompleteness Theorum too, and the Halting Problem; there are just some limits we aren't going to go past, and as science gets more refined it can define them more and more completely.
This is a compromise I can live with even if I don't fully approve of it: Use the protection when the games comes out and patch it out of existence later.
The main problem with anti-piracy measures is there is this sick idea that they should be 100% effective, so it's worth annoying the customer, crippling the game, etc. They don't need to be 100% effective, just "greater then the value lost by the company", and since the value curve is very steep (downhill) in computer games, it's worth removing the pirate protection later.
The fact of the matter is that as of this moment, nobody is financially hurt in any significant way if I 'pirate' Quake 1 right now. (Could conceivably change if they were to re-release it as is, but that is unlikely.) Why worry about it so? Sure, it's illegal and arguably unethical, but why should the companies care to sweat it?
I value a handful of concepts from Post-Modernism. In particular, I have made repeated use of the concept of "narrative", in ways that saved me a lot of words that would have still failed to concisely capture what I mean.
For instance, I have made the claim that the reason the modern news media looks biased is that the editors are constructing a narrative using the raw material of accounts of events. I am not now making that claim and debate would be out of line; my point is that using the concept of 'narrative' makes that a very powerful sentence, whether it is right or wrong.
However, we are not criticizing post-modernism per se. We are criticising the claimed leaders of the field, and the claimed scholarship of the field. These I unashamedly call "shit". I do not have a problem seperating some of the root concepts from the people theoretically promoting them, but the valuable aspects of post-modernism are mostly uncontroversial observations about the nature of language, completely incapable of supporting the airy academic castles that have been built on top of them.
Looking over a lot of the replies here, I think "Engineer's Disease" is a pretty accurate assessment--essentially, the belief that any field that deals in the unquantifiable is of lesser value than fields that deal with the quantifiable.
That is not the claim. The claim is that the output of the academic self-appointed "practitioners" of the field are bereft of value, not merely lesser value.
I'm a musician in my spare time. I value the contributions of theory to the field, even as I recognize that it is soft, not capable of "proof" or "disproof", and really only "useful" inasmuch as it is used to produce music. But there is true value there, true scholarship, true attempts to formulate theories of what sounds will have what effect on listening humans. Is it "more" or "less" valuable then materials science? I don't even consider the question meaningful.
But I still say that postmodern academic discourse is without meaning, and I say it without further qualification or shame. I am not a music expert but I recognize that there is value there. If despite true effort, I can not even discern meaning in the post-modern writing, then I feel fully qualified to call them on it, being rather educated and practiced in the art of extracting meaning from symbolic representations, both in theory and in practice.
Science and engineering "toolkits" are not applicable to the task of analyzing stories.
I am not analysing stories, I am analysing scholarship, a very particular brand of it.
No, he made the argument and he meant it. Under their rules it should not have bothered them that he did not mean it.
If the Social Texteditors find my arguments convincing, then why should they be disconcerted simply because I don't?
It did bother them, thus it is valid to speculate about why. (My call is "hypocrisy", but that's just one interpretation. I can't prove it without access to the editors I don't have.)
The problem is that as scientists/engineers/rational people, the concept of "gibberish" has meaning. If "gibberish" is given the same standing in a journal like this as claimed "meaningful writing", then the logical conclusion is that there is no distinguishing between "gibberish" and "meaningful writing", and as a consequence there is no such thing as "meaningful writing" going on if it's all logically equivalent to "gibberish".
Thus, to be consistent, the po-mos must act as you say, but as scientists/engineers/rational people, we're not going to buy it and we're going to conclude they're full of crap. This bothers them, and again, it shouldn't since they despise our worldview so. This is what I believe Sokal was saying, and what I thought you were getting at. There's no inconsistency here from Sokal or the engineering camp, there's just inconsistency from the po-mos. (And hypocrisy, in the sense that by their own logic this isn't supposed to bother them, but it clearly does. As a scientist/engineer/rational person, I would claim this is because no matter how hard you try, you can only disconnect from reality so far...)
The list of benefits coming from using Java is too long to take the speed-only view seriously.
The problem there is if you start taking a truly open look around, Java stops looking so cool either.
Its single greatest feature that might justifiably be called unique is the fact that it has a lot of library support you can buy, and if those things help enough they might tip the balance.
But in every other way, there's a language that meets or beats Java on almost every level.
To prevent this from becoming too confrontational, I won't try to name the languages. But Java really isn't the king the pro-Java zealots like to pretend it is; it's just the current Big Kahuna in certain domains, just like C++.
Perhaps it would be more correct to say that OOP is a SB when realized by a language like Python (it definitely isn't w/ C++ and Java).
Actually, over the past year or so I've come to the point where I also believe this. The advantages of OO promised in the 80's are just now materializing...... but only in very small handful of languages, mostly Python and C++ in the hands of a true expert. (I'd bet ruby qualifies but I've never used it, and Perl can qualify but you end up needing to implement good object semantics on your own, which is both good and bad. Of these languages Python probably has the lowest barrier to entry for the maximal OO power.)
This fact is largely going unnoticed, but it'll probably penetrate the mainstream here in another five years. I sure hope so;-), programming in Python is so much less frustrating then other langauges. A lot of people have a lot of unlearning to do before they can realize the benefits.
Python with good OO in the hands of a competent person magnifies their power almost beyond belief. I'm building an innovative application in Python, have been over the last year or so, and I wouldn't even be a 10% as far as I am now in C++. I am easily matching the output of a full-time 3 or 4 man C++ team in my spare time with Python. Once this really cranks up in the mainstream I expect to start seeing some real differences in programs.
It is not the engineers thinking they understand everything; it is the engineers demonstrating that the lit crits understand nothing technical, and arguing by extension that the odds of them understanding anything in their own field, absent evidence to the contrary, seem to be very, very low.
It is also a disease to think that you must be an "expert" in something to have any sort of valid opinion. The fact that dedicated attempts by smart people with no agenda and an honest intent to find meaning in postmodern lit-crit have failed is a strong indictment that can not be waved away with "Oh, they're not experts", and merely provides more fuel for the idea that lit-crit is full of crap and rather then demonstrate some sort of meaning, they must resort to ad-hominem to defend it.
Note that engineering disciplines can indeed meet that criteria; even if you don't understand math you can see there's something there. Even if you don't understand structural engineering, you can see buildings that stand vs. those that don't. (Consider the recent earthquakes in California and Iran, with similar magnitudes but vastly different outcomes.)
These people aren't making appeals to authority, they aren't standing on their "engineering creds", they're demonstrating pointlessness. It would be a fallacy to claim "I am an engineer, therefore their writing is pointless." But they are not doing that; they are saying "After a serious attempt to find meaning in these writings, we have failed and have been forced to conclude there is none." You want to prove otherwise, you need to produce the unabiguous "meanings" for these sentences; "our" side has done its work (as I am with these guys).
Indeed, if anyone is guilty of assuming competence, it is these lit folks, literally attempting to re-write the world so that engineering and science are just "another point of view", all the while willfully failing to understand why they are different.
The engineering and science toolkits are perfectly applicable to the task of analyzing lit-crit. The failures of the analysis must be laid at the feet of the lit-crit, not the techniques.
Grow up, branch out. Experts are just people who have studied something for a while, and they may yet be wrong. Nothing prevents an engineer from being an expert in something else, too. Stop pigeonholing people and stop suffering from "expert disease", OK? It's not good for any of us, because you can vote.
I was going to RTFA you, but I note Sokal's followup has not been linked yet. In it, he makes the following statement:
Of course, I'm not oblivious to the ethical issues involved in my rather unorthodox experiment. Professional communities operate largely on trust; deception undercuts that trust. But it is important to understand exactly what I did. My article is a theoretical essay based entirely on publicly available sources, all of which I have meticulously footnoted. All works cited are real, and all quotations are rigorously accurate; none are invented. Now, it's true that the author doesn't believe his own argument. But why should that matter? The editors' duty as scholars is to judge the validity and interest of ideas, without regard for their provenance. (That is why many scholarly journals practice blind refereeing.) If the Social Texteditors find my arguments convincing, then why should they be disconcerted simply because I don't? Or are they more deferent to the so-called ``cultural authority of technoscience'' than they would care to admit?
Perhaps this will resolve your misunderstanding on why "no on seems to have made that argument"?
My apologies; I was unclear. I did not mean "kicking out a high bandwidth users", though I understand why you may have thought that on re-reading my original post.
What I did mean is that something is going to happen to those users to cut their usage down, and I did not mean to specify or imply exactly "what" at all.
But I do guarentee they won't like it and many of them will bitch. There's no way to make them happy, so that's a bit of a lost cause. (You weren't necessarily claiming otherwise but it's worth pointing out;-) )
In fact, a typical distribution is indeed exponential, so in real life many fewer then half their customers will be above average. The cited example is very close to reality.
The fallacy of course lies in the implicit idea that if you get rid of all of the "above average" customers, you won't still have "above average" customers.;-)
Still, eliminating or significantly reducing the bandwidth used by as few as 50 or 100 people can significantly improve the performance of the system for many, many thousands of others. (Without going into details I will claim without evidence that I've seen the numbers in a real life example to back this up.) If those thousands of others are experiencing difficulties and complaining (and subsequently terminating service), guess who's gonna get it?
It may suck if you're one of the 50 or 100 people, but if you look at it abstractly, there's nothing else an ISP can possible do. Not even increase the bandwidth, since things like Gnutella and Bittorrent can grow their bandwidth use to match the expansion. Sooner or later, the top folks need to curb their use, and for better or for worse, the ISP folks will have to be the heavies.
FWIW, they don't necessarily enjoy it, it's just the way life is.
Similar to the if (0 == c) trick in C, I've been trying to train my fingers to type DELETE WHERE whenever I mean to type "DELETE". Then, I fill out the WHERE clause and only then go back to say what table to delete from.
This also gives you time to ponder the wisdom of first running a SELECT statement with the same WHERE clause and comtemplate whether you want to do this.
Technically, this would not be "the recording industry" but "the manufacturing industry". I don't know but I'd bet significant money that physical CD production is outsourced by all RIAA members; it makes zero economic sense for them to do otherwise.
If returns become a problem, be assured that that information will wend its way back to the CD manufacturers eventually, with direct economic consequences.
About two years ago [now more like three], I took Statistics in college. The second most interesting part of the course was the industrial manufacturing focus, specifically quality control.
Even more specifically, reliable ways to determine how much quality you can take out of the product and still meet some specification with some good probability....
On the other hand, these are the techniques used to reliably manufacture a device that will fail in 2 years, plus or minus 3 months, with 50% probability. This is the source of the flood of cheap garbage that has really only been gaining steam in the past four or five years; yes, in the 80s and 90s people were decrying "consumerism" but it's gotten several times worse as some of these techniques become refined and universally applied.
The upshot of all of this is yes, the quality of consumer electronics has been steadily declining for a decade or two now, along with everything else that comes off the factory floor, and the better the statistical techniques get, the closer to the consumer rejection threshold this stuff will get.
(New emphasis; as I'm quoting myself "emphasis mine" doesn't make too much sense;-) )
My lecturer in Distributed Communications said that "increasin bandwith will just result in software makers letting their software use more bandwith", which off course brings us back to where we started.
I do not find this argument anywhere near as compelling as the (more-or-less proven true) argument that as processors increase, programmers will use more processor time.
Programmers use more processor time because we can trade off processor time for human development time. If you're doing "normal programming" and using C when you could instead be using Perl or Python, you're a fool and it's costing you lots of money. For most programs, the users will never know the difference, but you're taking more of your preciou$ time to develop a worse product.
What's the equivalent with bandwidth? We generally already send down the entire dataset for a given app; when you view a web page, the browser greedily loads it, even if it's 20 pages long and there's an image at the bottom you won't look at for another ten minutes. Speculative downloading of the next page only costs 2x or 3x more of the bandwidth.
Decentralized P2P programs aren't "wasting" the bandwidth, though they could (and should) get a little more efficient, they're using it. Bigger pipes might actually increase their efficiency, though they might be encouraged to be more wasteful then. Still, this is only one app that might conceivably grow the bandwidth usage to match bandwidth, a trend that does not make.
Sure, we'll ship around DVDs, but those are new apps; nobody bitches that SETI@Home consumes all of my brand new processor, the app needs all the power it can get. New apps that weren't feasible previously don't really that much more power.
I just don't see what software is suddenly going to waste 10x-100x-1000x more bandwidth just because it can, the way our programs are easily 100x slower then they used to be (but they are written many, many, many times faster then they used to be; certain types of programs are easily 1000x easier to write then they were in the days things were running 100x "faster"). What do the programmers get for this? Handling more data is generally harder, and it's not like the data we're throwing around the net is already especially compressed, or that we're going to stop compressing it anytime soon. (As long as processors remain umpteen times faster then bandwidth, it's worth it to compress.)
I can see why your Distibuted Processing professor would claim this, but most of us aren't using it, aren't about to use it, and have no interest in it. Bandwidth used in a lab, full of distributed processing computers, isn't an interesting metric. Even then, it takes a lot of distributed processing value to use much bandwidth; distributed processing is only useful if you can offload a problem and get an answer more quickly then you can compute it yourself, and that typically only applies in certain heavy-duty computation cases that aren't normal. Mere message-passing is going to look much the same as the bandwidth we already use; I'm subscribed to about 40 RSS files, and we could pass messages instead of my software polling them, but the bandwidth use will be about the same either way (almost nil vs. almost nil).
I've tried multiple times to do the Google search for the computations I've seen, but I always get swamped by other, irrelevant results. Here's as close as I can get, a link to a quote from Bruce Schneier, which quotes him as saying:
"...the annual energy output of our sun is about 1.21*10^41 ergs. This
is enough to power about 2.7*2^56 single bit changes on our ideal
computer; enough state changes to put a 187-bit counter through all its
values. If we built a Dyson sphere around the sun and captured all of
its energy for 32 years, without any loss, we could power a computer to
count up to 2^192. Of course, it wouldn't have the energy left over to
perform any useful calculations with this counter.
"A typical supernova releases something like 10^51 ergs... If all of
this energy could be channeled into a single orgy of computation, a
219-bit counter could be cycled through all of its states.
"These numbers have nothing to do with the technology of the devices;
they are the maximums that thermodynamics will allow. And they strongly
imply that brute-force attacks against 256-bit keys will be infeasible
until computers are built from something other than matter and occupy
something other than space."
(Note I don't know what algorithm that is a 256-bit key for.)
Of course this trivially extends into more bits, even if you consume all the power in the universe. Conservatively estimating the power you could obtain at, say, a dectillion supernovas (10 ^ 33 times the supernova), that only buys you another 110 bits. And as he says, that's just for the counter; actual computation will strip several bits from you. 512 bits isn't even concievably brute-force-able in this universe and that's being really, really conservative.
Your message says "Assuming the impossible happens, an impossible thing can happen."
Yeah. Sure. Believe that computational fairies will pop out of particle physics that will carry us off to computational heaven. But don't sit there and tell me the impossible is actually going to happen.
When you figure out how to move anything, let alone a computer, into another dimension, drop me a line. I'd be interested in investing in your company. In the meantime, you're believing in the impossible, and it's on you to prove otherwise, preferably with something a bit more solid then "proof by assertion".
Good luck; you're a lock for the Nobel Prize in Physics if you succeed.
I was not arguing that 128-bit encryption will stand forever. I was arguing that there is a (very reasonable) level of encryption that will.
Again, "We will progress forever!" is a fallacy; it contradicts everything we know about physics. Is it absolutely impossible? No, but it's so unreasonable a stance as to be effectively so. There is an upper limit to the amount of computation a given amount of matter can sustain. It's huge. But encryption gets hard exponentially, you can only linearly increase your computation power once you hit the limit by throwing more matter at it. Exponentially rapidly beats linear.
To claim otherwise is to essentially believe in magic. Real magic, not just advanced technology.
Slashdot Mirror
First, the "hardness" needs to be measured against the value of the benefit obtained from gaming. If it's large, more effort will be thrown at it.
Second, you seem to have missed the implications of my carefully-chosen word simulate. You don't need to replicate the algorithm, just create something that mostly works in most of the situations that you care about. (Both "mosts" are important.) This is a significantly lower bar then "complete replication", and is one of the reasons it's so hard to combat this; while you need to get to 99.9% accuracy to be a "good search engine", the engine gamers can quite often get by on a simulation that is hardly correct at all... I'd guess the Google spammers understand at most 1% of the Google algorithm... but that's plenty to game the system, often because they can make up the difference with "overwhelming force". (The google spammers aren't going to sweat the difference between a ranking of 5.554 and 5.556 when they can just create another 20 sites that link to the desired site; "overwhelming force".) The search engines are fighting on a very unbalanced-in-the-gamer's-favor battlefield.
Again, don't overread my claim. I'm not saying this is inevitable; Google largely wins overall. But the possibility should not be waved away; it's a big mistake to do so. (Especially for the implementers; if you think your algorithm is perfect you're likely to learn how wrong you are in a big way as you bet too much on it, whereas I doubt the people at Google ever thought they'd ever completely "win", so they make much more reasonable choices.)
No, the body will flush it through normal processes, just like gum doesn't sit in the body for years and years. Your digestive system is a pipe; there's nowhere for something to just "hang out" for years.
This is true and it's easy to prove: A penny is much larger, completely indigestable, and will invariably come out. The only danger of eating a single penny is that some part of the "pipe" may be too small to accomodate it, in which case it could be fatal as it clogs the whole system up. As styrofoam has much more give, this does not apply. If you can handle a penny, you can handle syrofoam.
Shit-eating is not just "not fun", it's actively and supremely dangerous, with a good possibility of death without (and possibly with) medical attention. Among other nasty organisms, it tends to harbor e coli in dangerous quantities.
This is why you have not seen a "shit eating" contest on "Fear Factor"; it's not safe at all.
I doubt Styrofoam would cause much more then mild discomfort, if that. Shit stands a good chance to kill you. I know which I'd choose... Styrofoam would be a better bet, by a long shot.
"Natural" is a meaningless term in determining toxicity; botulism toxin is natural, cobra venom is natural, all toxic houseplants are natural, for that matter e coli and ebola are 100% all natural too. I'd rather eat 100 grams 100% artificial Vitamin C then 100 grams of botulism toxin.
- Reconstruct algorithm.
- Simulate algorithm and play with the inputs until the outputs match what you want.
- Bring those inputs about.
is extremely powerful, and note that as a "meta-algorithm" there's absolutely no way to completely shut it down.You have only four basic defenses against this:
- Keep changing the algorithm (expensive and large changes may not be possible if stability is desirable, which for search results it generally is),
- make the input-gaming process more expensive then the value of the output to the attacker (as you become more valuable you're a more enticing target),
- make the outputs desired by the attacked impossible (generally not possible in the general case, but in certain limited ways it is; it is probably not possible to be the #1 google result for all possible search terms, for instance, despite the desirability of such a result),
- or have a human monitoring attacks and shut them down manually (only possible if you can out-staff the attackers)
There are some other possibilities but a lot of them don't apply in the real world, like "make it impossible to reverse the inputs necessary for some output" (like MD5); this is not applicable to a real-world application like a search engine because there has to be some obvious human-sensible logic to the placement or the search engine is just returning random results, which is not even a "search engine", let alone a useful one. Not even all four can be brought to bear in a given situation; #2 probably doesn't apply in this case since the benefits could be in the millions of dollars in theory.I'm not saying that WebFountain is hosed; Google has trouble but it is handlable. But it is worth talking about; certain basic algorithms will have certain effects as people try to game them, and it may be the case that some clever, useful algorithm is so easily gamed and so difficult to create countermeasures for that it will never be possible in the real world in the general case.
(I doubt this is the case, but there's only one way to find out, and that's try it and see what happens.)
I neither insulted you nor hinted that you don't know what you are talking about.
Surely not, since this is the first I've posted to the thread.
"You're wrong, you're talking nonsense, and you don't even know it" isn't condenscending if it's true.
If it's possible, then we can do it.
So what? Penetrating the Heisenberg Uncertainty Principle isn't possible, even in principle.
I'm more a student of humanity than science. And I've seen it more than once, the common belief that we can't do something, only to find out later ( days, years, centuries ) that it is possible.
Unfortunately, your misconception is not neutral, which is why I post. (And remember, you're not the only reader; I realize I don't stand a chance in hell of convincing you, because you're so certain you're right you have the confidence to post on the topic. Yes, the equivalent logic applies equally to me.) Unrealistic expectations fuel Luddite tendencies and fuel unrealistic extrapolations into the future.
Unfortunately, when acted on, these cause suboptimal results, sometimes even killing people. Consider the effects of unrealistic expectations on the environmentalism issues; what if we act as if you're right and always assume there's a way to clean up our environmental messes before they kill us, when the science says we're wrong?
And I've seen it more than once, the common belief that we can't do something, only to find out later ( days, years, centuries ) that it is possible.
Somewhat ironically, it shows that you're a student of ancient history but have not been keeping up to date on recent history. History doesn't support the idea that we're going to find a way past the Heisenberg Uncertainty Principle. History has in fact shown the increasing understanding and refinement of boundaries and limits, ever since the first ones were discovered.
You should not be surprised when people don't buy your Argument from Ignorance, or call it for what it is. The arrogance lies in your thinking that your knowlege, despite its nonexistance (I note you could not possibly have studied the Halting Problem or Godel's Incompleteness Theorum in the time period since your last post), is so much greater then scientists and mathematicians that you can contradict them on the fundamentals of the universe. That is arrogance, that is inflated self-assessment.
Of course that doesn't mean they're always right; but they don't have to be to still be right about certain limits.
This is your cue (especially due to my previous parenthetical note) to now go find some cursory overview of Godel's Theorem, read it in two minutes, and explain why one of the greatest mathematical acheivements in the history of man that proves certain things are impossible doesn't prove that certain things are impossible. Here, let me help with that first part. OK, this is a little condenscending, but can I just say "been here, done this?" Have fun.
I Am Not A Materials Scientist... but doesn't plastic's structure take permenent damage from deformation on this level? Even the strongest plastic key takes damage from each other key it destroys, so eventually it too will die.
Thus, there is no strongest plastic key; they will all be destroyed in short order. In fact, while you might superstitiously keep your "superkey" that broke 20 others, statistically speaking the ideal strategy (in terms of maximal wins) is to use a key precisely once and throw it away. Your superkey is more and more likely to break on each try.
If I'm wrong please correct me, or if you know the terminology please let me know.
This is perhaps the single most pointless fad ever, and elitist comments to the contrary, I would expect that the general public, even the young kids general public, is too smart to be interested in this as more then a passing fad.
Strict weather prediction will never happen; see the sibling post to your own. QM actually prevents it, believe it or not.
What could conceivably happen is that we start manipulating weather on a large scale, and we might learn how to bend weather to our will. We'd need essentially random corrections due to the forces of chaos, but conceivably with enough control, we could say "It will rain 3 inches on this site three years in the future" (with the implicit assumption the weather control grid will still be working, i.e., no major nuclear war, no nearby supernovas, etc.).
But that's not prediction, that's control, and there's a big difference. The unpredictability of the system would still manifest itself as a complete inability to predict in advance what inputs to the system would be necessary to maintain the states we desire; we'd have to correct dynamically and in the short-term. So, even this doesn't solve the "predictability" problem, it just pushes it out one meta-level; the fundamental unpredictability remains.
Seriously though, we may not be able to imagine how it will work, or the solutions we can imagine don't work at all, but I'm confident it will happen, both for earthquakes and weather and anything else overly complex. Note that I did not say sometime soon, although I would like to see that too, I understand the technology and science we need just isn't up to par yet.
"Science" has proven that it can't work. Making those things work requires that the impossible be done. Arguments of the form "If an impossible thing happens, another impossible thing can happen" are trivially logically true, but not relevant in the real world.
Before you continue to assert how I will eventually be "proven wrong by the unbounded and unimaginable progress of humanity!!!1!!", please study the computer science concept of reduction; any solution to the weather prediction problem reduces to a method to penetrate the Heisenberg Uncertainty Principle fog, which would cause the complete collapse of particle physics as we know them (and remember, advances historically speaking refine past theories, not destroy them). If you still believe at that point that we might get past it, at least then you'll have some vague glimmering of the magnitude of power you are claiming we can obtain; I get no sense that you realize how scientifically and mathematically silly you're being from your current messages.
While you're at it, might want to study Godel's Incompleteness Theorum too, and the Halting Problem; there are just some limits we aren't going to go past, and as science gets more refined it can define them more and more completely.
This is a compromise I can live with even if I don't fully approve of it: Use the protection when the games comes out and patch it out of existence later.
The main problem with anti-piracy measures is there is this sick idea that they should be 100% effective, so it's worth annoying the customer, crippling the game, etc. They don't need to be 100% effective, just "greater then the value lost by the company", and since the value curve is very steep (downhill) in computer games, it's worth removing the pirate protection later.
The fact of the matter is that as of this moment, nobody is financially hurt in any significant way if I 'pirate' Quake 1 right now. (Could conceivably change if they were to re-release it as is, but that is unlikely.) Why worry about it so? Sure, it's illegal and arguably unethical, but why should the companies care to sweat it?
I value a handful of concepts from Post-Modernism. In particular, I have made repeated use of the concept of "narrative", in ways that saved me a lot of words that would have still failed to concisely capture what I mean.
For instance, I have made the claim that the reason the modern news media looks biased is that the editors are constructing a narrative using the raw material of accounts of events. I am not now making that claim and debate would be out of line; my point is that using the concept of 'narrative' makes that a very powerful sentence, whether it is right or wrong.
However, we are not criticizing post-modernism per se. We are criticising the claimed leaders of the field, and the claimed scholarship of the field. These I unashamedly call "shit". I do not have a problem seperating some of the root concepts from the people theoretically promoting them, but the valuable aspects of post-modernism are mostly uncontroversial observations about the nature of language, completely incapable of supporting the airy academic castles that have been built on top of them.
Looking over a lot of the replies here, I think "Engineer's Disease" is a pretty accurate assessment--essentially, the belief that any field that deals in the unquantifiable is of lesser value than fields that deal with the quantifiable.
That is not the claim. The claim is that the output of the academic self-appointed "practitioners" of the field are bereft of value, not merely lesser value.
I'm a musician in my spare time. I value the contributions of theory to the field, even as I recognize that it is soft, not capable of "proof" or "disproof", and really only "useful" inasmuch as it is used to produce music. But there is true value there, true scholarship, true attempts to formulate theories of what sounds will have what effect on listening humans. Is it "more" or "less" valuable then materials science? I don't even consider the question meaningful.
But I still say that postmodern academic discourse is without meaning, and I say it without further qualification or shame. I am not a music expert but I recognize that there is value there. If despite true effort, I can not even discern meaning in the post-modern writing, then I feel fully qualified to call them on it, being rather educated and practiced in the art of extracting meaning from symbolic representations, both in theory and in practice.
Science and engineering "toolkits" are not applicable to the task of analyzing stories.
I am not analysing stories, I am analysing scholarship, a very particular brand of it.
The problem is that as scientists/engineers/rational people, the concept of "gibberish" has meaning. If "gibberish" is given the same standing in a journal like this as claimed "meaningful writing", then the logical conclusion is that there is no distinguishing between "gibberish" and "meaningful writing", and as a consequence there is no such thing as "meaningful writing" going on if it's all logically equivalent to "gibberish".
Thus, to be consistent, the po-mos must act as you say, but as scientists/engineers/rational people, we're not going to buy it and we're going to conclude they're full of crap. This bothers them, and again, it shouldn't since they despise our worldview so. This is what I believe Sokal was saying, and what I thought you were getting at. There's no inconsistency here from Sokal or the engineering camp, there's just inconsistency from the po-mos. (And hypocrisy, in the sense that by their own logic this isn't supposed to bother them, but it clearly does. As a scientist/engineer/rational person, I would claim this is because no matter how hard you try, you can only disconnect from reality so far...)
The list of benefits coming from using Java is too long to take the speed-only view seriously.
The problem there is if you start taking a truly open look around, Java stops looking so cool either.
Its single greatest feature that might justifiably be called unique is the fact that it has a lot of library support you can buy, and if those things help enough they might tip the balance.
But in every other way, there's a language that meets or beats Java on almost every level.
To prevent this from becoming too confrontational, I won't try to name the languages. But Java really isn't the king the pro-Java zealots like to pretend it is; it's just the current Big Kahuna in certain domains, just like C++.
Perhaps it would be more correct to say that OOP is a SB when realized by a language like Python (it definitely isn't w/ C++ and Java).
... but only in very small handful of languages, mostly Python and C++ in the hands of a true expert. (I'd bet ruby qualifies but I've never used it, and Perl can qualify but you end up needing to implement good object semantics on your own, which is both good and bad. Of these languages Python probably has the lowest barrier to entry for the maximal OO power.)
;-), programming in Python is so much less frustrating then other langauges. A lot of people have a lot of unlearning to do before they can realize the benefits.
Actually, over the past year or so I've come to the point where I also believe this. The advantages of OO promised in the 80's are just now materializing...
This fact is largely going unnoticed, but it'll probably penetrate the mainstream here in another five years. I sure hope so
Python with good OO in the hands of a competent person magnifies their power almost beyond belief. I'm building an innovative application in Python, have been over the last year or so, and I wouldn't even be a 10% as far as I am now in C++. I am easily matching the output of a full-time 3 or 4 man C++ team in my spare time with Python. Once this really cranks up in the mainstream I expect to start seeing some real differences in programs.
Try the Sokal articles.
It is not the engineers thinking they understand everything; it is the engineers demonstrating that the lit crits understand nothing technical, and arguing by extension that the odds of them understanding anything in their own field, absent evidence to the contrary, seem to be very, very low.
It is also a disease to think that you must be an "expert" in something to have any sort of valid opinion. The fact that dedicated attempts by smart people with no agenda and an honest intent to find meaning in postmodern lit-crit have failed is a strong indictment that can not be waved away with "Oh, they're not experts", and merely provides more fuel for the idea that lit-crit is full of crap and rather then demonstrate some sort of meaning, they must resort to ad-hominem to defend it.
Note that engineering disciplines can indeed meet that criteria; even if you don't understand math you can see there's something there. Even if you don't understand structural engineering, you can see buildings that stand vs. those that don't. (Consider the recent earthquakes in California and Iran, with similar magnitudes but vastly different outcomes.)
These people aren't making appeals to authority, they aren't standing on their "engineering creds", they're demonstrating pointlessness. It would be a fallacy to claim "I am an engineer, therefore their writing is pointless." But they are not doing that; they are saying "After a serious attempt to find meaning in these writings, we have failed and have been forced to conclude there is none." You want to prove otherwise, you need to produce the unabiguous "meanings" for these sentences; "our" side has done its work (as I am with these guys).
Indeed, if anyone is guilty of assuming competence, it is these lit folks, literally attempting to re-write the world so that engineering and science are just "another point of view", all the while willfully failing to understand why they are different.
The engineering and science toolkits are perfectly applicable to the task of analyzing lit-crit. The failures of the analysis must be laid at the feet of the lit-crit, not the techniques.
Grow up, branch out. Experts are just people who have studied something for a while, and they may yet be wrong. Nothing prevents an engineer from being an expert in something else, too. Stop pigeonholing people and stop suffering from "expert disease", OK? It's not good for any of us, because you can vote.
Geez, kids these days.
;-)
sally.au
You old-schoolers know what I'm talkin' about.
Oh, come on, old schooler, get with the program. This is the web, for crying out loud! That should read as:
sally.au
Much better.
Warning, not safe for work
(A class 2 sarcasm warning has been issued for this post.)
Wow, one brush, 300 million people.
And this tripe was modded "Insightful"? Try the exact opposite.
My apologies; I was unclear. I did not mean "kicking out a high bandwidth users", though I understand why you may have thought that on re-reading my original post.
;-) )
What I did mean is that something is going to happen to those users to cut their usage down, and I did not mean to specify or imply exactly "what" at all.
But I do guarentee they won't like it and many of them will bitch. There's no way to make them happy, so that's a bit of a lost cause. (You weren't necessarily claiming otherwise but it's worth pointing out
In fact, a typical distribution is indeed exponential, so in real life many fewer then half their customers will be above average. The cited example is very close to reality.
;-)
The fallacy of course lies in the implicit idea that if you get rid of all of the "above average" customers, you won't still have "above average" customers.
Still, eliminating or significantly reducing the bandwidth used by as few as 50 or 100 people can significantly improve the performance of the system for many, many thousands of others. (Without going into details I will claim without evidence that I've seen the numbers in a real life example to back this up.) If those thousands of others are experiencing difficulties and complaining (and subsequently terminating service), guess who's gonna get it?
It may suck if you're one of the 50 or 100 people, but if you look at it abstractly, there's nothing else an ISP can possible do. Not even increase the bandwidth, since things like Gnutella and Bittorrent can grow their bandwidth use to match the expansion. Sooner or later, the top folks need to curb their use, and for better or for worse, the ISP folks will have to be the heavies.
FWIW, they don't necessarily enjoy it, it's just the way life is.
Similar to the if (0 == c) trick in C, I've been trying to train my fingers to type DELETE WHERE whenever I mean to type "DELETE". Then, I fill out the WHERE clause and only then go back to say what table to delete from.
This also gives you time to ponder the wisdom of first running a SELECT statement with the same WHERE clause and comtemplate whether you want to do this.
If returns become a problem, be assured that that information will wend its way back to the CD manufacturers eventually, with direct economic consequences.
Direct economic consequences is why this occurs; I posted about this on my weblog in relation to a similar Ask Slashdot regarding hardware. The CD manufacturers will be using the exact same statistical techniques I mention in my post on CDs that electronics manufacturers use on their stuff.(New emphasis; as I'm quoting myself "emphasis mine" doesn't make too much sense
I wonder if they're looking into giant robot anteaters as an alternative to costly bunker-buster bombs?
Licking terrorists to death is probably SOME violation of the Geneva Convention!
My lecturer in Distributed Communications said that "increasin bandwith will just result in software makers letting their software use more bandwith", which off course brings us back to where we started.
I do not find this argument anywhere near as compelling as the (more-or-less proven true) argument that as processors increase, programmers will use more processor time.
Programmers use more processor time because we can trade off processor time for human development time. If you're doing "normal programming" and using C when you could instead be using Perl or Python, you're a fool and it's costing you lots of money. For most programs, the users will never know the difference, but you're taking more of your preciou$ time to develop a worse product.
What's the equivalent with bandwidth? We generally already send down the entire dataset for a given app; when you view a web page, the browser greedily loads it, even if it's 20 pages long and there's an image at the bottom you won't look at for another ten minutes. Speculative downloading of the next page only costs 2x or 3x more of the bandwidth.
Decentralized P2P programs aren't "wasting" the bandwidth, though they could (and should) get a little more efficient, they're using it. Bigger pipes might actually increase their efficiency, though they might be encouraged to be more wasteful then. Still, this is only one app that might conceivably grow the bandwidth usage to match bandwidth, a trend that does not make.
Sure, we'll ship around DVDs, but those are new apps; nobody bitches that SETI@Home consumes all of my brand new processor, the app needs all the power it can get. New apps that weren't feasible previously don't really that much more power.
I just don't see what software is suddenly going to waste 10x-100x-1000x more bandwidth just because it can, the way our programs are easily 100x slower then they used to be (but they are written many, many, many times faster then they used to be; certain types of programs are easily 1000x easier to write then they were in the days things were running 100x "faster"). What do the programmers get for this? Handling more data is generally harder, and it's not like the data we're throwing around the net is already especially compressed, or that we're going to stop compressing it anytime soon. (As long as processors remain umpteen times faster then bandwidth, it's worth it to compress.)
I can see why your Distibuted Processing professor would claim this, but most of us aren't using it, aren't about to use it, and have no interest in it. Bandwidth used in a lab, full of distributed processing computers, isn't an interesting metric. Even then, it takes a lot of distributed processing value to use much bandwidth; distributed processing is only useful if you can offload a problem and get an answer more quickly then you can compute it yourself, and that typically only applies in certain heavy-duty computation cases that aren't normal. Mere message-passing is going to look much the same as the bandwidth we already use; I'm subscribed to about 40 RSS files, and we could pass messages instead of my software polling them, but the bandwidth use will be about the same either way (almost nil vs. almost nil).
(Note I don't know what algorithm that is a 256-bit key for.)
Of course this trivially extends into more bits, even if you consume all the power in the universe. Conservatively estimating the power you could obtain at, say, a dectillion supernovas (10 ^ 33 times the supernova), that only buys you another 110 bits. And as he says, that's just for the counter; actual computation will strip several bits from you. 512 bits isn't even concievably brute-force-able in this universe and that's being really, really conservative.
Your message says "Assuming the impossible happens, an impossible thing can happen."
Yeah. Sure. Believe that computational fairies will pop out of particle physics that will carry us off to computational heaven. But don't sit there and tell me the impossible is actually going to happen.
When you figure out how to move anything, let alone a computer, into another dimension, drop me a line. I'd be interested in investing in your company. In the meantime, you're believing in the impossible, and it's on you to prove otherwise, preferably with something a bit more solid then "proof by assertion".
Good luck; you're a lock for the Nobel Prize in Physics if you succeed.
I was not arguing that 128-bit encryption will stand forever. I was arguing that there is a (very reasonable) level of encryption that will.
Again, "We will progress forever!" is a fallacy; it contradicts everything we know about physics. Is it absolutely impossible? No, but it's so unreasonable a stance as to be effectively so. There is an upper limit to the amount of computation a given amount of matter can sustain. It's huge. But encryption gets hard exponentially, you can only linearly increase your computation power once you hit the limit by throwing more matter at it. Exponentially rapidly beats linear.
To claim otherwise is to essentially believe in magic. Real magic, not just advanced technology.
Still a fallacy. If you don't realize why, you don't understand quantum computing.