In any event, why do you think Mr. Garbus should have known the answers to these questions before he asked them?
I think effective lawyers, when they are questioning a witness, always know the answer they would like to get back. It's their job to know, because they are trying to present a case which is a legal argument for an audience, the judge and/or jury. Of course, this is a deposition, so it may be that the lawyer is actually trying to discover information rather than make a legal argument.
Seriously though... why does he object to the form of every other question? Just goes to show that the MPAA doesn't feel that they have too great of a case and have to win it on legal nitpicking. Bastards.
Almost without exception, when Gold objects to the form of a question, he is usually trying to get the question clarified so that Schumann can answer it meaningfully. You could hardly call that nitpicking. I would like to see the MPAA lose this case, but I think Gold's objections are totally justified. In fact, I think Gold performs considerably better than either Garbus or Hernstadt in this transcript.
In fact, that exchange was in my original submission for the story... I understand that Mr. Garbus is what... in his 50s? And Linux etc is a pretty new thing for him. He's worked really hard to try to understand what's going on... but reading this deposition has a "comedy of errors" feeling to it. Lots of things that show that they really don't quite know what's going on.
It may be the case that Garbus et al. are not very well-informed w.r.t. Linux, etc., but I don't believe that that is necessarily the reason they are asking the deponent what you may regard as banal questions. I believe a lawyer's objective in questioning is, in part, to establish the scope of knowledge of the deponent, and also to establish for the record what is "best practice", etc. in his expert opinion (if he is an expert).
For example, Garbus (or Hernstadt?) asked if Schumann (the deponent) knew what CSS-cat is. Maybe Garbus knows, or maybe he doesn't. But the point is that it matters whether or not Schumann knows, because it bears directly on whether he can be credible on the subject.
Also, I think, a lawyer needs to raise certain issues in testimonies using only the Q&A method. They can't just state them outright themselves, especially if they are technical subjects; they need to be introduced according to the methods that are available in the legal proceedings. I think that lawyers don't usually address the judge or jury directly when presenting some piece of evidence, except in the opening and closing arguments.
You can kind of understand how the media ends up using the word HACKED since there are MANY pages out there (see: attrition.org cracked pages archive), that specifically say "YOU HAVE BEEN HACKED BY" or just the ever rampant "hacked".gif (my apologies Unisys) staring you in the face when you look at it.
Actually, I think the vogue is now, "you have been owned by...", followed by some 1337-speak. I agree that a "cracker" is someone who breaks through a barrier, like copy protection or security, and DoS attacks are a little different, so maybe we should call such "crackers" "owners" instead.:)
Excuse me for flaming, but this has got to be the most inane set of interview questions I have seen on Slashdot.
Q: CS people are clueless. I hate you ivory tower morons. Don't you think you should allow Joe Random Programmer into your school? I mean, if he knows Perl and Linux, he is a match for any roomful of educated people.
A: Yes, I deeply regret getting my Ph.D., and now worship and grovel before my very own students. Please don't hurt me...
*gag*
Q: You require SAT scores to get in? Don't you think they are a poor indicator of real skills? I mean, I had terrible SATs but I still managed to get a job!
A: Yeah, you're right. Every other American college offers SATs only because it introduces an additional stumbling block which lets us filter out lazy malcontents who think they have skills but only like to constantly complain that they are not getting "recognized" for their original contributions to science, like that Perl wrapper they wrote last week.
Ugh.
Q: I don't want to slight you, and this isn't a troll or anything, but isn't Ars Digita just a veil for presenting your own views on software architecture and design. I mean, isn't this going to be kind of, well, I dunno, biased?
A: Absolutely not. I was planning to sublimate my entire weltanschauung and teach only ideas that other people invented, and with which I vehemently disagree.
*wimper*
Q: Hi, Phil. I'm a high-placed executive at a large company. Our profits rose by 30% in the last quarter. Here, check out these margins! Don't you think you would profit by making a partnership with us?
A: Yeah, Oracle and Sun have given me tens of thousands of dollars of free stuff; they've served their use. I guess I will drop them now and try to make a deal with their competitors...
Sad. Slashdot garners some pretty interesting people for interviews. I hate to see these chances go to waste with such absurd and vaccuous questions for material.
Metallica is not the only group of artists who have suffered grievous financial hardship and borderline privation due to the diabolical Napster client. Don't you think that other artists such as Sting, Janet Jackson and Oasis should also be recompensed for their losses?
Well, I do; that's why I created the paysting.com, payjanet.com and payoasis.com sites, to solicit money from all you stupid suckers^H^H^Hwonderful fans who only downloaded the MP3s in a moment of weakness. I have no affiliation whatsoever with the artists, their agents or their respective labels, but I promise all the proceeds will be handed over to the artists in full.
Honest.
(Translation: "I can't believe no one has mentioned this yet: what assurance does anybody have that the donations will actually reach Metallica at all?")
Of course I know little or nothing about paleology, but if the meteor did indeed block out a large degree of sunlight, it seems likely that a lot of marine life would also die off. For example, sea algae, I believe, needs sunlight just like any other plant. If the sea algae die off, then a major source of food for some plant-eating marine animals would disappear. Then those animals die off, and so a food source for some other animals disappears. And so on.
The ecological balance is a delicate thing...
Really, it is hard to imagine a catastrophic event like a major meteor impact not having significant and deadly repercussions on every ecological niche.
AFAIK, the GPL has never been challenged in court. Do you think it should be? Of course I'm not saying we should "pick a fight", but do you think it may be better to establish at least one precedent to demonstrate that the GPL actually holds up in the legal arena?
Of course, the best thing for free software adovcates is if everyone just follows the terms without any conflict, but it has occasionally occurred to me that confidence in the GPL could be undermined by even just one incident which impinges even indirectly on the legal validity of the GPL. If that happens, it would not only encourage people to stop using the GPL, but might destabilize existing free software houses like Red Hat, or even encourage a family of companies to lead a legal assault on the GPL which the FSF could not withstand, and which would render all of your and our (the free source community) work on free software unprotected.
Open source (and sometimes free software) has been hailed as the answer to every problem known to modern programmers. It makes programs reliable. It makes programs secure. It makes programs more efficient. It speeds development. It encourages plurality and increases your options as a developer.
I think that there is a kernel of truth to all of these things, but also that it has been blown greatly out of proportion. One may hold on principle, as you do, that we should only use free software; in that case any benefits derived from the OSS methodology are a pleasant side-effect.
Others use the benefits of OSS themselves as the incentive for pursuing it. OK, fair enough. Still, despite the claims, I'm not aware of any scientifically conducted and peer-reviewed surveys (not case studies---these are only the beginnings of a statistical sampling) that substantiate them. If you know of any please provide pointers.
And then there are many zealots and marketing weenies who have seized on the notion with little thought other than jumping on the bandwagon, and now hold it up as a panacea. It reminds me an awful lot of the mania that surrounded Java and XML when they first gained prominence.
Personally I believe all this hype mania only weakens our position in making arguments for free software that we can substantiate.
What is your response to all this frenzy? Are we over the hump?
In a press release announcing the suit, publicists for the band and music companies even threw in a statement from Metallica drummer Lars Ulrich, who said it is "sickening to know that our art is being traded like a commodity rather than the art that it is."
Lars, aren't music sales and promotion already commoditizing your art?
To the author of the Infinite Monkey Protocol Suite (IMPS), SteQven M. Christey:
Thank you for your work toward proving that, given a finite amount of time, an infinite amount of monkeys can produce the complete works of Shakespeare. It seems that you are already well on your way to proving an important related hypothesis, namely that a single monkey given an infinite amount of time can produce an RFC which is a complete waste of work.
I encourage you to continue your worthy research; I am confident it will bear fruit and rewards for your efforts. In fact, here, have a banana.
I think all this furor over cybersquatting is ridiculous. A domain name is just a heuristic, a convenient way to designate a network address. If you really want to find a site, you don't start guessing *.com names and seeing if they lead somewhere—you enter some keywords into a search engine, and then you bookmark them when you find them!
I agree that cybersquatting is a problem. But what it should be indicating to us is not that we need the government to step in and protect some arbitrary naming system which happens to collide with commercial trademarks and what-have-you, but rather that the technology, i.e., the global namespace system, is faulty (or obsolete)!
We should learn a lesson from the financial regulatory community. It used to be that when a bank failed, the government would step in and save it. Then we discovered that banks saved that way would just fail again, because they were not competently managed. So now we let banks fail so that new, better managed banks can emerge, and serve the general populace better and more efficiently.
The same principle of natural selection should hold for technology. If some technology proves inadequate, let it fail! Don't perpetuate it with regulations or laws. We all know the "fail early, fail fast" motto.
"Calculus" is just a term for a formal system, usually equational in nature, which supports some form of deduction based on rewriting. There are easily hundreds, probably thousands, of calculi which appear in the programming language theory and related literature, and few of them if any have anything to do with the differential calculus you are familiar with from high school.
If you are interested in learning about lambda-calculus, you might want to start with untyped lambda-calculus, unless you have a background in logic. There are many books on the subject; I started with Chris Hankin's "Lambda Calculi: A Guide for Computer Scientists", which is IMO quite readable. You can also find many tutorials and lecture notes on the web, for example here, here and here. If you know Scheme, or have read SICP, then you already know mostly what lambda-calculus is, though you probably don't realize why it's so remarkable. (Hankin's book is good for that.)
Surely you jest! Most languages are turing complete and allow non-terminating programs.
I do not jest. (I think you misread my message. I said that ML programs need not terminate.) There are many calculi which are interesting and useful and also terminate. When you say "most languages", I think you mean "most programming languages". Some people take computational adequacy ("Turing completeness") to be the defining difference between a programming calculus and a programming language.
For ML the following would do:
let rec f x = (f x)
Indeed. This makes use of the the fixpoint operator, but there is no fixpoint in what is generally taken as the polymorphic lambda-calculus. That's one of the differences between them, though there are many others.
Perhaps you are confusing with Charity in which all programs terminate?
No, I'm not. Also, Charity is not based on any lambda-calculus. It has a fixpoint operator (four, in fact), but it is a logically sound one; hence one gets termination.
So now this Harel decides that a problem is insoluble? If a team of researchers try to solve a problem,
should they stop because Harel says it can't be done? Who does this guy thinks he is, the All-knowing
deus? Isn't it so that the effort to solve a problem can yield other results? Isn't that what science is
about?
Without having read the book, it seems safe to assume that Harel is discussing computability problems. These are the very foundation of computer science. Harel didn't just "decide" something is unworthy of further research; it is a fundamental theorem that the answers to some questions, such as the Halting Problem (given an arbitrary program, does it terminate?), are not computable. As this is a mathematical fact, not an opinion, trying to write a program which solves the problem in general really is certifiable exercise in futility.
That said, finding special cases, i.e., classes of programs for which termination is decidable, is certainly interesting and useful and has led to some interesting mathematics. For example, typed, polymorphic functional programming languages such as ML are based on a core language (the polymorphic lambda-calculus) in which every program is guaranteed to terminate. Then, to turn that formalism into a general-purpose programming language, one generally extends it with, among other things, a facility for doing recursion, which destroys the termination properties, but still leaves you with a language which satisfies other nice properties and enjoys a relatively simple correspondence with formal mathematics, which is immensely useful for checking partial correctness.
Anyway, it seems to me that what Harel is probably trying to accomplish in this book is not to obstruct research or dissuade researchers from pursuing ideas, but rather to point out to a non-technical audience what we know for a fact that computers are incapable of doing. (I admit I am somewhat aghast at the claim that some computer researchers are not aware of computability limitations, though! It's absolutely fundamental...)
BTW, another good book is Douglas Hofstadter's Pulitzer Prize-winning Goedel, Escher, Bach: An Eternal Golden Braid, which goes into the issue in much detail and has much interesting musing on what artificial intelligence might mean.
Slashdot Mirror
For example, Garbus (or Hernstadt?) asked if Schumann (the deponent) knew what CSS-cat is. Maybe Garbus knows, or maybe he doesn't. But the point is that it matters whether or not Schumann knows, because it bears directly on whether he can be credible on the subject.
Also, I think, a lawyer needs to raise certain issues in testimonies using only the Q&A method. They can't just state them outright themselves, especially if they are technical subjects; they need to be introduced according to the methods that are available in the legal proceedings. I think that lawyers don't usually address the judge or jury directly when presenting some piece of evidence, except in the opening and closing arguments.
Actually, I think the vogue is now, "you have been owned by...", followed by some 1337-speak. I agree that a "cracker" is someone who breaks through a barrier, like copy protection or security, and DoS attacks are a little different, so maybe we should call such "crackers" "owners" instead. :)
Excuse me for flaming, but this has got to be the most inane set of interview questions I have seen on Slashdot.
*gag*
Ugh.
*wimper*
Sad. Slashdot garners some pretty interesting people for interviews. I hate to see these chances go to waste with such absurd and vaccuous questions for material.
Metallica is not the only group of artists who have suffered grievous financial hardship and borderline privation due to the diabolical Napster client. Don't you think that other artists such as Sting, Janet Jackson and Oasis should also be recompensed for their losses?
Well, I do; that's why I created the paysting.com, payjanet.com and payoasis.com sites, to solicit money from all you stupid suckers^H^H^Hwonderful fans who only downloaded the MP3s in a moment of weakness. I have no affiliation whatsoever with the artists, their agents or their respective labels, but I promise all the proceeds will be handed over to the artists in full.
Honest.
(Translation: "I can't believe no one has mentioned this yet: what assurance does anybody have that the donations will actually reach Metallica at all?")
P.S. All major credit cards are accepted.
Of course I know little or nothing about paleology, but if the meteor did indeed block out a large degree of sunlight, it seems likely that a lot of marine life would also die off. For example, sea algae, I believe, needs sunlight just like any other plant. If the sea algae die off, then a major source of food for some plant-eating marine animals would disappear. Then those animals die off, and so a food source for some other animals disappears. And so on.
The ecological balance is a delicate thing...
Really, it is hard to imagine a catastrophic event like a major meteor impact not having significant and deadly repercussions on every ecological niche.
Well, just my two cents.
AFAIK, the GPL has never been challenged in court. Do you think it should be? Of course I'm not saying we should "pick a fight", but do you think it may be better to establish at least one precedent to demonstrate that the GPL actually holds up in the legal arena?
Of course, the best thing for free software adovcates is if everyone just follows the terms without any conflict, but it has occasionally occurred to me that confidence in the GPL could be undermined by even just one incident which impinges even indirectly on the legal validity of the GPL. If that happens, it would not only encourage people to stop using the GPL, but might destabilize existing free software houses like Red Hat, or even encourage a family of companies to lead a legal assault on the GPL which the FSF could not withstand, and which would render all of your and our (the free source community) work on free software unprotected.
Open source (and sometimes free software) has been hailed as the answer to every problem known to modern programmers. It makes programs reliable. It makes programs secure. It makes programs more efficient. It speeds development. It encourages plurality and increases your options as a developer.
I think that there is a kernel of truth to all of these things, but also that it has been blown greatly out of proportion. One may hold on principle, as you do, that we should only use free software; in that case any benefits derived from the OSS methodology are a pleasant side-effect.
Others use the benefits of OSS themselves as the incentive for pursuing it. OK, fair enough. Still, despite the claims, I'm not aware of any scientifically conducted and peer-reviewed surveys (not case studies---these are only the beginnings of a statistical sampling) that substantiate them. If you know of any please provide pointers.
And then there are many zealots and marketing weenies who have seized on the notion with little thought other than jumping on the bandwagon, and now hold it up as a panacea. It reminds me an awful lot of the mania that surrounded Java and XML when they first gained prominence.
Personally I believe all this hype mania only weakens our position in making arguments for free software that we can substantiate.
What is your response to all this frenzy? Are we over the hump?
Lars, aren't music sales and promotion already commoditizing your art?
To the author of the Infinite Monkey Protocol Suite (IMPS), SteQven M. Christey:
Thank you for your work toward proving that, given a finite amount of time, an infinite amount of monkeys can produce the complete works of Shakespeare. It seems that you are already well on your way to proving an important related hypothesis, namely that a single monkey given an infinite amount of time can produce an RFC which is a complete waste of work.
I encourage you to continue your worthy research; I am confident it will bear fruit and rewards for your efforts. In fact, here, have a banana.
So would somebody please tell us all what was so great about the Amiga? I remember it had nice graphics for its time. That's about it.
I think all this furor over cybersquatting is ridiculous. A domain name is just a heuristic, a convenient way to designate a network address. If you really want to find a site, you don't start guessing *.com names and seeing if they lead somewhere—you enter some keywords into a search engine, and then you bookmark them when you find them!
I agree that cybersquatting is a problem. But what it should be indicating to us is not that we need the government to step in and protect some arbitrary naming system which happens to collide with commercial trademarks and what-have-you, but rather that the technology, i.e., the global namespace system, is faulty (or obsolete)!
We should learn a lesson from the financial regulatory community. It used to be that when a bank failed, the government would step in and save it. Then we discovered that banks saved that way would just fail again, because they were not competently managed. So now we let banks fail so that new, better managed banks can emerge, and serve the general populace better and more efficiently.
The same principle of natural selection should hold for technology. If some technology proves inadequate, let it fail! Don't perpetuate it with regulations or laws. We all know the "fail early, fail fast" motto.
"Calculus" is just a term for a formal system, usually equational in nature, which supports some form of deduction based on rewriting. There are easily hundreds, probably thousands, of calculi which appear in the programming language theory and related literature, and few of them if any have anything to do with the differential calculus you are familiar with from high school.
If you are interested in learning about lambda-calculus, you might want to start with untyped lambda-calculus, unless you have a background in logic. There are many books on the subject; I started with Chris Hankin's "Lambda Calculi: A Guide for Computer Scientists", which is IMO quite readable. You can also find many tutorials and lecture notes on the web, for example here, here and here. If you know Scheme, or have read SICP, then you already know mostly what lambda-calculus is, though you probably don't realize why it's so remarkable. (Hankin's book is good for that.)
I do not jest. (I think you misread my message. I said that ML programs need not terminate.) There are many calculi which are interesting and useful and also terminate. When you say "most languages", I think you mean "most programming languages". Some people take computational adequacy ("Turing completeness") to be the defining difference between a programming calculus and a programming language.
For ML the following would do:
Indeed. This makes use of the the fixpoint operator, but there is no fixpoint in what is generally taken as the polymorphic lambda-calculus. That's one of the differences between them, though there are many others.
Perhaps you are confusing with Charity in which all programs terminate?
No, I'm not. Also, Charity is not based on any lambda-calculus. It has a fixpoint operator (four, in fact), but it is a logically sound one; hence one gets termination.
So now this Harel decides that a problem is insoluble? If a team of researchers try to solve a problem, should they stop because Harel says it can't be done? Who does this guy thinks he is, the All-knowing deus? Isn't it so that the effort to solve a problem can yield other results? Isn't that what science is about?
Without having read the book, it seems safe to assume that Harel is discussing computability problems. These are the very foundation of computer science. Harel didn't just "decide" something is unworthy of further research; it is a fundamental theorem that the answers to some questions, such as the Halting Problem (given an arbitrary program, does it terminate?), are not computable. As this is a mathematical fact, not an opinion, trying to write a program which solves the problem in general really is certifiable exercise in futility.
That said, finding special cases, i.e., classes of programs for which termination is decidable, is certainly interesting and useful and has led to some interesting mathematics. For example, typed, polymorphic functional programming languages such as ML are based on a core language (the polymorphic lambda-calculus) in which every program is guaranteed to terminate. Then, to turn that formalism into a general-purpose programming language, one generally extends it with, among other things, a facility for doing recursion, which destroys the termination properties, but still leaves you with a language which satisfies other nice properties and enjoys a relatively simple correspondence with formal mathematics, which is immensely useful for checking partial correctness.
Anyway, it seems to me that what Harel is probably trying to accomplish in this book is not to obstruct research or dissuade researchers from pursuing ideas, but rather to point out to a non-technical audience what we know for a fact that computers are incapable of doing. (I admit I am somewhat aghast at the claim that some computer researchers are not aware of computability limitations, though! It's absolutely fundamental...)
BTW, another good book is Douglas Hofstadter's Pulitzer Prize-winning Goedel, Escher, Bach: An Eternal Golden Braid, which goes into the issue in much detail and has much interesting musing on what artificial intelligence might mean.