Really? I would think that being able to benefit from your labor and creativity is a strong incentive. Strong enough that the person who does it best gets rewarded accordingly, and only indirectly (though substantially) does the public benefit. The public benefit is frosting on the cake. Protection of an individual's claim to their own work is the heart of it.
Not even slightly. Since he said it better than I could, I'll quote Thomas Babington Macauley (he's talking about copyright, but the principle is the same):
The advantages arising from a system of copyright are obvious. It is desirable that we should have a supply of good books; we cannot have such a supply unless men of letters are liberally remunerated; and the least objectionable way of remunerating them is by means of copyright.[...]
I believe, Sir, that I may with safety take it for granted that the effect of monopoly generally is to make articles scarce, to make them dear, and to make them bad. [...] It is good that authors should be remunerated; and the least exceptionable way of remunerating them is by a monopoly. Yet monopoly is an evil. For the sake of the good we must submit to the evil; but the evil ought not to last a day longer than is necessary for the purpose of securing the good.
Patents are a contract between society and inventors: Keep inventing good things, and we'll protect your claims long enough that you may profit. There is no "right" to profit solely from your ideas; it is a privilege society grants in exchange for something of sufficient value. This privilege is being abused by a fucked-in-the-head system and understandably avaricious opportunists.
Good show. You got to most of the dreadfully incisive commentary I was going to make first. This does seem more like an attempt to use/.'s well-known proclivities to evoke pressure against Mr. de Joode.
I want to add my own voice to the din, though, and say quite strongly that I find Mr. Bertrand's position in thie matter untenable at best, quite despicable at worst. And to add to Mr. de Joode has published his his own stance on the subject.
Domains are a first-come/first-served biz. Even if Mr. Bertrand's claim that Mr. de Joode grabbed the.org domain before they registered the.com (which seems flaky in light of the information returned by whois) is factual, he loses out. He does not have some sort of Goddess-granted right to the domain. Next case.
I'm not that big a fan of the idea of expanding it to 500 questions. IMO, that's just starts to drag. But I do have suggestions, being mainly that you cut out or rephrase those questions the exemplify Slashdot-pure behavior (i.e., having to check your User Info page to know your Karma? Bah.) -- and rework some of the overly specific questions. Others have pointed out the issue with the IANAL cluster.
Mostly, though, it's a pretty good start. Oh, and because I said on the test I would but didn't say I'd cheated on it, I am apparently a JonKatz Wannabe at 74.5%. I'm offended.
I can't say that I agree at all with S@H's position on this subject. True, there are some right bastards out there who need to feed their egos so very much that they would falsify data, but that risk has already been taken. The important thing now is to find a way to minimize said risk.
S@H seems to think that this is best accomplished by keeping the source code to themselves. Not so. For instance, analyzing the communications protocol employed by S@H clients does not require the least bit of knowledge of how the client works internally. No, the solution to the bad-data problem that immediately occurs to me is redundancy -- having multiple independent computers run through the same dataset and compare their results. A mismatch would be handled by reissuing the same dataset to a different set of independent computers -- until one iteration produces agreement.
This would slow down traversal of the dataspace some, but that is far preferable to the introduction of rogue data -- which, I repeat, is already occuring.
Besides, S@H is running out of dataspace anyway.
Making the client sources available would almost certainly improve the efficiency of the client and would quiet those who object so strongly to running an unchecked program on their system. It would, it seems to me, make the entire effort more worthwhile and more interesting. Security through obscurity simply doesn't work.
I am very strongly opposed to this suggestion, for one simple reason: I've tutored students who started with bad habits. I think I frightened them, actually (I can grow very emphatic when disgusted by seeing yet another random jump or meaningless variable name). Pascal (or maybe Scheme) is still my prime choice for a starting language -- sure, it sucks to actually use it for anything, but I've never met a language that teaches better habits. And it's easy enough for ye students to get a grasp on. Since I don't think any languages should be taught the first year anyway (I like the idea of simply working with pseudocode for a year, see below), I'm particularly opposed to teaching them something that will take years to unlearn if they decide to get serious about programming.
Since I've gotten sucked into this discussion anyway, here's my thought. Learning to program, first of all, is a general skill which actually has applications in other fields. It requires rigorous structured thinking, and does a world of good towards sharpening critical thinking skills. For that reason and the fact that the world is rapidly turning more digital, I'm strongly in favor of making programming concepts a required course.
Note that I said concepts. The language taught in the first year should be nothing more and nothing less than a vehicle for the concepts being conveyed. I don't know enough about Python (it's on my TODO list, but you know how that goes) to be able to comment on its suitability. I would be happier if it were made practical to not teach using a language at all for at least the first semester. The thing is, picking up a new language once you know how to program is easy. Want to learn perl so you can do cgis? Read some manpages. Want to learn Java? Go read Sun's web pages for about a week. No sweat. But if you start somebody off with bad concepts, everything else will be harder for them.
This is also why I'm opposed to OOP as part of the first concept-course. OOP is wonderful. When I first discovered the concept, I was stunned by the power thereby implied. Oddly enough, it leaves first-year students cold (I will not go into my rant about CMU at this point, I'll stick to the rant I'm on). The thing is, it's an overwhelming set of ideas to start with. In my mental map of programming stuff, it fits into a cubby marked 'extensions to basic programming concepts'. First, a student learning OOP after learning algorithmic programming will have some idea of what he is seeing and why it is worth learning. This is good. Second, a stundet who starts on OOP will be screwed in a non-OOP environment. Bad. Third, well, I just think the sheer initial learning curve I associate with OOP, added to the learning curve of programming in general, would be evil and dissuasive. So you teach it to second-year students in parallel with their C classes.
Okay, enough ranting for just now.
Most of the ideas expressed in this post are its author's; some of them arose out of a discussion on the subject. Any errors in logic or application are my fault, not theirs.
Slashdot Mirror
Not even slightly. Since he said it better than I could, I'll quote Thomas Babington Macauley (he's talking about copyright, but the principle is the same): Patents are a contract between society and inventors: Keep inventing good things, and we'll protect your claims long enough that you may profit. There is no "right" to profit solely from your ideas; it is a privilege society grants in exchange for something of sufficient value. This privilege is being abused by a fucked-in-the-head system and understandably avaricious opportunists.
Good show. You got to most of the dreadfully incisive commentary I was going to make first. This does seem more like an attempt to use /.'s well-known proclivities to evoke pressure against Mr. de Joode.
I want to add my own voice to the din, though, and say quite strongly that I find Mr. Bertrand's position in thie matter untenable at best, quite despicable at worst. And to add to Mr. de Joode has published his his own stance on the subject.
Domains are a first-come/first-served biz. Even if Mr. Bertrand's claim that Mr. de Joode grabbed the .org domain before they registered the .com (which seems flaky in light of the information returned by whois) is factual, he loses out. He does not have some sort of Goddess-granted right to the domain. Next case.
I'm not that big a fan of the idea of expanding it to 500 questions. IMO, that's just starts to drag. But I do have suggestions, being mainly that you cut out or rephrase those questions the exemplify Slashdot-pure behavior (i.e., having to check your User Info page to know your Karma? Bah.) -- and rework some of the overly specific questions. Others have pointed out the issue with the IANAL cluster.
Mostly, though, it's a pretty good start. Oh, and because I said on the test I would but didn't say I'd cheated on it, I am apparently a JonKatz Wannabe at 74.5%. I'm offended.
I can't say that I agree at all with S@H's position on this subject. True, there are some right bastards out there who need to feed their egos so very much that they would falsify data, but that risk has already been taken. The important thing now is to find a way to minimize said risk.
S@H seems to think that this is best accomplished by keeping the source code to themselves. Not so. For instance, analyzing the communications protocol employed by S@H clients does not require the least bit of knowledge of how the client works internally. No, the solution to the bad-data problem that immediately occurs to me is redundancy -- having multiple independent computers run through the same dataset and compare their results. A mismatch would be handled by reissuing the same dataset to a different set of independent computers -- until one iteration produces agreement.
This would slow down traversal of the dataspace some, but that is far preferable to the introduction of rogue data -- which, I repeat, is already occuring.
Besides, S@H is running out of dataspace anyway.
Making the client sources available would almost certainly improve the efficiency of the client and would quiet those who object so strongly to running an unchecked program on their system. It would, it seems to me, make the entire effort more worthwhile and more interesting. Security through obscurity simply doesn't work.
I am very strongly opposed to this suggestion, for one simple reason: I've tutored students who started with bad habits. I think I frightened them, actually (I can grow very emphatic when disgusted by seeing yet another random jump or meaningless variable name). Pascal (or maybe Scheme) is still my prime choice for a starting language -- sure, it sucks to actually use it for anything, but I've never met a language that teaches better habits. And it's easy enough for ye students to get a grasp on. Since I don't think any languages should be taught the first year anyway (I like the idea of simply working with pseudocode for a year, see below), I'm particularly opposed to teaching them something that will take years to unlearn if they decide to get serious about programming.
Since I've gotten sucked into this discussion anyway, here's my thought. Learning to program, first of all, is a general skill which actually has applications in other fields. It requires rigorous structured thinking, and does a world of good towards sharpening critical thinking skills. For that reason and the fact that the world is rapidly turning more digital, I'm strongly in favor of making programming concepts a required course.
Note that I said concepts. The language taught in the first year should be nothing more and nothing less than a vehicle for the concepts being conveyed. I don't know enough about Python (it's on my TODO list, but you know how that goes) to be able to comment on its suitability. I would be happier if it were made practical to not teach using a language at all for at least the first semester. The thing is, picking up a new language once you know how to program is easy. Want to learn perl so you can do cgis? Read some manpages. Want to learn Java? Go read Sun's web pages for about a week. No sweat. But if you start somebody off with bad concepts, everything else will be harder for them.
This is also why I'm opposed to OOP as part of the first concept-course. OOP is wonderful. When I first discovered the concept, I was stunned by the power thereby implied. Oddly enough, it leaves first-year students cold (I will not go into my rant about CMU at this point, I'll stick to the rant I'm on). The thing is, it's an overwhelming set of ideas to start with. In my mental map of programming stuff, it fits into a cubby marked 'extensions to basic programming concepts'. First, a student learning OOP after learning algorithmic programming will have some idea of what he is seeing and why it is worth learning. This is good. Second, a stundet who starts on OOP will be screwed in a non-OOP environment. Bad. Third, well, I just think the sheer initial learning curve I associate with OOP, added to the learning curve of programming in general, would be evil and dissuasive. So you teach it to second-year students in parallel with their C classes.
Okay, enough ranting for just now.
Most of the ideas expressed in this post are its author's; some of them arose out of a discussion on the subject. Any errors in logic or application are my fault, not theirs.
P.S.: Has anybody considered interCal?