I know I'm an old fart, but for future reference, writing in capital letters is considered equivalent to shouting, and is therefore rude. When writing to express an opinion, especially as part of an official consultation, its best to provide emphasis *like* *this*, rather than LIKE THIS. If you have the option of using HTML or other rich formats, you can similarly do this or this.
Nice post. As a matter of interest, whats your take on Crusoe's VLIW instractions ? To what extent can the problems with VLIW be circumvented by running the compiler at run-time ?
My guess would be that you can do it to some extent, provided your compiler is sophisticated enough, and you're prepared to compile several versions of the same source for different input data.
Of course I have a choice about whether to install a freenet node. Thats not the question. The question is: Freenet, and other technologies, if used widely, have consequences even for those who do not use it, in that, for instance, copyright will become untenable. Is it therefore OK that those who introduce such things do so unhindered by social constraints ?
Spouting naive egoism doesn't answer every (or possibly any) question. People don't live in hermetically sealed boxes, and therefore don't always get a choice when the "rules" start changing.
Sorry to be picky, but all modern CPU evaluate many instructions simultaneously, using multiple-issue systems for many similar functional units of eack kind. The difference with VLIW is that several instructions are held in the same instruction word.
In conventional VLIW the instructions combined in the same word must not have any dependencies on one another. The Crusoe native instruction set is VLIW in this sense. The Itanium system - which they call EPIC - is more complex, allowing instructions that interfere with one another to be combined in the same word. This ought to be somewhat easier for the compiler, but it seems in fact that elements of it are causing problems.
Just thinking out loud here, but: is there any precedent concerning the distributing of decompiled or reverse engineered code for proprietary software. Its not clear to me that to do this is illegal. Consider:
1. The license agreement is a contract of adhesion, and therefore invalid in common law (except possibly, oddly, in Scotland). The software and its media are considered "yours" under conventional copyright law.
2. The original source code, and the object code are considered copyright of the publisher. Additionally the original source is a trade secret.
3. The reverse engineered or translated source, however, is your work, or the work of your software, albeit derived from the work of the original publisher.
It seems the situation is analogous to a translation of a book into another language, or a "cover" of a piece of music by a different artist. Whats the legal situation of that ?
I think this whole "write in languages that are C, but easier" movement that's been going on for decades is a little weird.
There are only four languages, C++, Java, Objective-C and C# that even come close to meeting this description. In the case of C++, its a mixed level language - parts are very low level, like C, and parts are fairly high leve. In this it is like many older, larger languages, like Pascal, Modula, Ada, PL/1, etc. Objective-C is similar in nature, but its high level parts are higher level - its a kind of wierd hybrid of Smalltalk and C. These two languages are as they are because they want to retain *compatibility* with C, which is widely used, whilst getting some or all of the percieved advantages of OO.
Java and C# are as they are because people are familiar with C-like syntax. The underlying mechanism of the languages relate more closely to Smalltalk - for instance they are dynamically linked, and can support reflection. Thus they are not really "like C" except in their syntax.
For once I agree with you. One of the things conventional approaches to Object Oriented Development tend to miss is that in the real world requirements change, and in cases where reuse is wanted, requirements often change drastically.
Real companies - especially product companies -cannot afford to set solid requirements at the start of a project. Scope for change - and reuse - has to be allowed in the design.
This, bluntly, is where *good* design and implementation come into play. Inevitably code must be changed, and inevitably the original design will have certain rigidities in it. To pick examples from different styles of programming, someone will have decided to use a 4 bit field for that thing that really has more than 16 different variants, or someone will have decided to use inheritance heavily to represent variation on some trivial dimension where you really need it for some unanticipated case that also includes all the same variation.
The only true way to avoid these problems is good engineering. Precisely what that means is hard to say, but I guess two elements of OO design *help* (and here, no doubt, we differ).
Firstly design patterns and abstraction should be used appropriatly in the design to abstract away the kind of variation between objects that might cause problems. As far as possible, public interfaces should be in places that you can be more certain of, and private ones in places that are more likely to vary.
Secondly, as programs, requirements and functionality evolve around your code, it must be constantly redesigned - not merely modified - in order to accomodate the changes. The minuite you spot that the design is wrong, you must change it. Its never appropriate to create "working code" now, and decide to "do it right later", because no-one, ever, ever does.
A lot of the discussion in this thread has focused on heavy, up-front, object-oriented design by "architects" who don't actually write any of the code, they just produce diagrams. Its my experience that this style of working is destructive. It deprives the actual developers of the power to go in an change the design which they will inevitably need, and it tends to produce designs which are, as you say, brittle, in that the factoring into classes is naive to the point where changes to the code will eventually totally destroy the abstraction.
There really are no compelling arguments against libertarianism that I''ve seen.
Sigh. Well this article is so old, noone will ever read this, but I cannot let you get away with saying this without challenge.
There are a great many arguments against libertarianism that hold some water, but if you reason entirely from libertarian principles its possible to condemn just about all of them as authoritarian. I recommened Mike Hubens excellent Site of critiques of libertarianism including his Non-libertarian FAQ for many different views.
The problem is that libertarianism reasoning is a hermetically closed loop of logic, which is in itself free of inconsistencies, but in itself that proves nothing. Nothin about consistency gaurantees good governance, and nothing about the way the principle of libertarianism are derived does either. This is why, to be completely blunt, libertarians are so dogmatic (the statement "there are no good arguments..." is dogmatic, and arrogant. If you disagree, check a dictionary), and those who try to argue with them get so frustrated.
The logic goes like this: The highest value is freeedom. We must maximise everyone's freedom. This is meant in the negative sense of freedom: a man is free if noone prevents him doing as he pleases with himself and his property if he interferes with noone else. Thus the sole role of the state is the prevention of coercion and fraud.
The first thing to note is that this is a very unusal use of the word "freedom". In general usage people are happy to talk about their freedom at work, or in their families, or to change suppliers for some good. Libertarians, however, assert that if you sign a contract obliging you to do something you have acted freely, and thus if I complain about my lack of freedom to, say, take bathroom breaks at work, this is mere whinging and my freedom has not been affected. After all, I can quite, can't I ? and I signed the contract with my employer in the first place.
To see this, if its not sufficiently clear, consider an employer who sets up separate "whites only" and "blacks only" drinking fountains at work, and fires employess who disregard the separation. Most of us would consider such behaviour abhorent, and most people would not object to a law against it. Libertarians, however, assert that the employer is quite within his rights. I admit I'm pressing emotional buttons to make a point here. I don't imply that libertarianism implies racism, or that libertarians would condone such employment politicies. I do, however, assert that the libertarian idea of freedom is not very close to the common use of the word. IMHO this problem derives from treating property as an extension of the person and essentially absolute. It is better - in my view - to see property as a social phenomenon, a compromise, whose use must be regulated.
Secondly, and along similar lines, libertarianism is not an adequate moral system, as should be clear from the above. At best all it offers is a minimal framework for law. Nothing in fundamental libertarian philosophy prevents one from selling oneself into slavery, for instance.
The only possible justification for libertarianism other than that it is moral is that it is efficient - that it provides the greatest degree of social or economic good of any social system. The arguments here are economic, but they are at the very least inconclusive. You have to believe in the perfection of the unregulated free market to accept them.
First I should point out that I am a libertarian, with a small l, in that my primary interest in politics is to ensure maximum freedom for everyone, but I have a rather broader idea of that than the LP program does, and thus could never support them, even if I were an American.
I'm just posting to spew a little random political philosophy. Your argument that we should first assume that new laws are unecessary until proven otherwise rests on totally libertarian assumptions, and this sounds totally reasonable. However, there are other logical (and possibly moral) political positions to take.
Its possible to argue that society comes before the individual, both historically and morally, and this argument is sound insofar as there never was a "state of nature" with individuals fending for themselves. Rather, humans have always lived in groups and those groups have always been hierarchical. The state it merely the latest in a long line of forms of social organisation which are hierarchical in that some individuals are acknowledged as leading the rest (within some constraints).
Thus the states use of force, in this view, is not the anwarranted and unjustifiable coercion it is in the libertarian view, but rather a necessary (but regrettable) part of the social structure.
Not that I agree with this. It simply needs to be pointed out that enlightenment liberal ideas are a recent innovation, and while they may *seem* obvious they are not.
Random points: Its "Anarchy, State, and Utopia", you got a superfluous "the" in there. Its surprisingly readable for academic political philosophy. Its also (IMNAAHO) not so much mistaken as inadequate to the task it sets out to acheive. Nozick has since changed his mind about quite a lot (though not all) of it, and has spent many recent years trying to devise a just system of inheritance taxation. I disagree strongly (regardless of its inadequacy) with his basic premise that a potential explanation is almost as good as a real one.
Nozick is not only a classmate of Rawls, but a friend and an admirer. "Anarchy, State and Utopia" was intended as a counterpoint to Rawls' infinitely duller "A Theory of Justice" (though I suspect Rawls' is a by far the better reasoned theory). Rawls is emphatically not an apologist for the welfare state (based on entitlement) but an advocate of a kind of modernised liberalism (in the classic sense) based on the idea that a system is just if someone would agree to participate in it even if they had no idea what their position in the society would be.
Nice post (and nice sig). My concern with Libertarianism (as opposed to libertarianism (note capitalisation) to which I would more or less adhere) is that it is - like many forms of collectivism - a highly dogmatic and "rationalist" (as opposed to empiricist) system of thought. I think you allude to this when you say that all abstractions will have bugs when implemented by humans.
In some cases libertarianism seems to have become as much a tool for making yourself more equal than others as Marxism became. Take, for example, the (actually objectivist, but hey) claims made elsewhere in the comments for this article that its OK to polute, make noise, and generally behave in an obnoxious manner, because "noone should tell me how to live my life" (and incidentally the LP seems to support this viewpoint). As the previous-but-one poster pointed our, in these cases one must suspect that libertarianism is merely an excuse to behave as one pleases with no regard for others.
Well thats an interesting theory. Personally I'm inclined to think that libertarianism/objectivism is attractive to the tech crowd (and has been attractive to me, though not much now), because it offers a consistent (if you don't look to hard) and clear system for moral behaviour.
I think many techies are disturbed by the woolly and complex nature of most people's moral ideas, and tend to resort to libertarianism (in personal life and politics) because it offers a safe harbour from that wooliness and a clear response to allegations of wrongdoing.
Many libertarians become irate to the point of appearing to panic when their ideas are challenged, especially by someone coming from a logical but more socialistic or conservative framework of ideas. This implies to me that libertarianism is really very important in their worldview, and I suspect this is a tool for cutting away the large swathes of fuzzy, illogical human concern which the more technical mind finds disturbing (I know I do) that libertarianism (and moreso objectivism) says are irrelevant.
If that sort fo stuff crashes the JVM, then its Well stuffed. Get a different one. Array out of bounds errors should throw an exception, and Java initialises *all* data to defaults (though for objects, this is nil, which will cause exceptions to be thrown).
Unless you use JNI, or some other kind of native code, a correctly written VM should never crash (though of course, it might *stop*).
This seems to be heresy at the moment, but: why does wireless matter ?
At the moment, the wireless technology in Europe consists of cellphone access to email and the web. Email I can see, but the web ? why ? Who wants to book railway tickets from the bus, or order books from the middle lane of the M25 ? Let alone read some random person's ramblings on their home page on a screen smaller than my fingernails.
I'm not saying wireless is useless. Wireless web access on devices with large enoungh screens to be usable, or access from cellphones to services that would be useful while on the move would both be handy. Cellphone applications especially, but it would rely on having phones with positioning technology.
Applications range from a map showing my current position, to the timetable for the railway station I'm currently standing in, to the location of the nearest McDonalds, or a list of movies showing in town tonight.
None of these things can be done yet. The usefulness of a WML browser on a unadorned cellphone is about nil. You Americans aren't missing anything.
You're quite right that most of an interview depends on getting on with the interviewer. In fact, the importance of every second decreases exponentially as the interview goes on. Many interviewers will have decided whether they'll recommend hiring you or not before you even open your mouth. Indeed, quite a lot of training *to* interview (rather than be interviewed) involves learning to keep an open mind for as long as possible.
However, I do think technical questions at interviews matter. Not necessarily the "How does this work in Java ?" type of questions (which are really just checks that you didn't exaggerate your knowledge on your CV), but problem solving and puzzles. Interviewers for really serious technical posts (not database grinding or HTML hacking) want to see your though processes and problem solving style, both to check that you really can do it, and to check that your "style" is compatible with "how we do it here". This can only be done at interview. On CVs, or over the phone, these things don't come through.
Its definitely true that you get more out of UML the more diagrams you use, and I think we agree that its a reasonable way to communicate a view of a project.
There's an AC post attached to mine that should really have been moderated up, pointing out that Christopher Alexander (on whose architecture work Design Patterns are based) made a distinction between formal and functional description, where the formal description describes what was done, and the functional description says why. I think this is important here: UML is a formal description. It needs to be accompanied by functional description to make useful sense.
My experience with UML is that the diagrams do not serve much useful purpose on their own. They almost always need to be accompanied by a verbal explanation of the problem being solved and the type of solution employed before they start to make sense. Possibly this is why this book does not go across very well. I do not see the addition of colour helping very much.
At best these kinds of diagrams and things are an aid to communication, but they are not a substitute for it, and they certainly are not a substitute for design. There seems to be quite a common belief that the existence of UML diagrams somehow implies that the design they represent is well thought through, or even coherent. It does not.
Copyright grants to the creator of a work the right to control the distribution of his work for profit (or otherwise, as he chooses). Its relevant that this is (in most people's opinion) not a "natural" right, because that makes its scope and form questionable, and up to society (through the state) to control.
Your copyright allows you to make demands of the licensee in exchange for your work. So far, so good. This allows you to impose certain conditions - in the case of the Microsoft EULA, payment, in the case of the GPL/BSDL certain conditions relating to copying and modification.
However, there are limits to what you can do with such a license. In short, you can't deprive the licensee or any other individual of their other rights through it. You can't (to pick an absurd example) require that someone go out and commit a murder in order to use you software.
To get back to the original issue: The question is not whether copyright is good or bad, but what the legitimate scope of a copyright license is. Can you require that only approved devices be used to access the work ? can you require that no copies be made ? Can you override somebody's right to free expression on the grounds that the knowledge they've expressed might be used to undermine your monopoly on players, or copy your work ?
According the doctrine of fair use, which is fundamental to the law in this matter, the answer to the first question is "no", you can't create a monopoly on players in this way, as it has nothing to do with your profiting from your work. To the second "sometimes", you can only control the making of copies where you can reasonably argue that you're being deprived of a sale you'd otherwise have made.
The answer to the third question is clearly the critical matter in this case. To us, the answer is pretty clearly no. The MPAA has no right to maintain its monopoly on players, and no right to restrict information about copying (thats like outlawing the instructions for making a photocopier).
Frankly, I find it disturbing that so many people have such a naive idea of copyright as a form of "ownership" that allows the owners copyright to override everything else.
Its true they have too many people processing for them - or at least that they don't particularly want more, or want to speed up the client. You can check the FAQ on their website There's only a limited amount of data coming from the radio observatories, and they have enough clients to process it.
I don't know of them sending out false data, but they've certainly sent out duplicate data. This, however, is partly a security precaution to catch people who modify the client (either for silly reasons (to get more points), or to speed it up) and break the core algorithms.
I don't think Robin was arguing that employees are not responsible for the actions of their employers. I think he was saying that they *determine* the actions of their employers.
In most large companies, there is a distinct lack of overall strategy coming from the top that actually succeeds in driving the whole organisation. That the point - there are no orders. The actions of Microsoft, or ZDNet, or Slashdot are just the actions of individuals acting more-or-less independently.
Nope. Its only a copyright violation if you copy text verbatim and try to pass it off as your own, or sell it. Summaries, reviews, reports, even properly attributed quotes, are all OK.
Its no wonder, really, that the public is so easily confused about copyright "theft" when even/. readers can't get it straight.
Slashdot Mirror
I know I'm an old fart, but for future reference, writing in capital letters is considered equivalent to shouting, and is therefore rude. When writing to express an opinion, especially as part of an official consultation, its best to provide emphasis *like* *this*, rather than LIKE THIS. If you have the option of using HTML or other rich formats, you can similarly do this or this.
Nice post. As a matter of interest, whats your take on Crusoe's VLIW instractions ? To what extent can the problems with VLIW be circumvented by running the compiler at run-time ?
My guess would be that you can do it to some extent, provided your compiler is sophisticated enough, and you're prepared to compile several versions of the same source for different input data.
Of course I have a choice about whether to install a freenet node. Thats not the question. The question is: Freenet, and other technologies, if used widely, have consequences even for those who do not use it, in that, for instance, copyright will become untenable. Is it therefore OK that those who introduce such things do so unhindered by social constraints ?
Spouting naive egoism doesn't answer every (or possibly any) question. People don't live in hermetically sealed boxes, and therefore don't always get a choice when the "rules" start changing.
As a matter of interest: why ?
PS. Crusoe's native IS is VLIW.
Sorry to be picky, but all modern CPU evaluate many instructions simultaneously, using multiple-issue systems for many similar functional units of eack kind. The difference with VLIW is that several instructions are held in the same instruction word.
In conventional VLIW the instructions combined in the same word must not have any dependencies on one another. The Crusoe native instruction set is VLIW in this sense. The Itanium system - which they call EPIC - is more complex, allowing instructions that interfere with one another to be combined in the same word. This ought to be somewhat easier for the compiler, but it seems in fact that elements of it are causing problems.
Just thinking out loud here, but: is there any precedent concerning the distributing of decompiled or reverse engineered code for proprietary software. Its not clear to me that to do this is illegal. Consider:
1. The license agreement is a contract of adhesion, and therefore invalid in common law (except possibly, oddly, in Scotland). The software and its media are considered "yours" under conventional copyright law.
2. The original source code, and the object code are considered copyright of the publisher. Additionally the original source is a trade secret.
3. The reverse engineered or translated source, however, is your work, or the work of your software, albeit derived from the work of the original publisher.
It seems the situation is analogous to a translation of a book into another language, or a "cover" of a piece of music by a different artist. Whats the legal situation of that ?
I think this whole "write in languages that are C, but easier" movement that's been going on for decades is a little weird.
There are only four languages, C++, Java, Objective-C and C# that even come close to meeting this description. In the case of C++, its a mixed level language - parts are very low level, like C, and parts are fairly high leve. In this it is like many older, larger languages, like Pascal, Modula, Ada, PL/1, etc. Objective-C is similar in nature, but its high level parts are higher level - its a kind of wierd hybrid of Smalltalk and C. These two languages are as they are because they want to retain *compatibility* with C, which is widely used, whilst getting some or all of the percieved advantages of OO.
Java and C# are as they are because people are familiar with C-like syntax. The underlying mechanism of the languages relate more closely to Smalltalk - for instance they are dynamically linked, and can support reflection. Thus they are not really "like C" except in their syntax.
For once I agree with you. One of the things conventional approaches to Object Oriented Development tend to miss is that in the real world requirements change, and in cases where reuse is wanted, requirements often change drastically.
Real companies - especially product companies -cannot afford to set solid requirements at the start of a project. Scope for change - and reuse - has to be allowed in the design.
This, bluntly, is where *good* design and implementation come into play. Inevitably code must be changed, and inevitably the original design will have certain rigidities in it. To pick examples from different styles of programming, someone will have decided to use a 4 bit field for that thing that really has more than 16 different variants, or someone will have decided to use inheritance heavily to represent variation on some trivial dimension where you really need it for some unanticipated case that also includes all the same variation.
The only true way to avoid these problems is good engineering. Precisely what that means is hard to say, but I guess two elements of OO design *help* (and here, no doubt, we differ).
Firstly design patterns and abstraction should be used appropriatly in the design to abstract away the kind of variation between objects that might cause problems. As far as possible, public interfaces should be in places that you can be more certain of, and private ones in places that are more likely to vary.
Secondly, as programs, requirements and functionality evolve around your code, it must be constantly redesigned - not merely modified - in order to accomodate the changes. The minuite you spot that the design is wrong, you must change it. Its never appropriate to create "working code" now, and decide to "do it right later", because no-one, ever, ever does.
A lot of the discussion in this thread has focused on heavy, up-front, object-oriented design by "architects" who don't actually write any of the code, they just produce diagrams. Its my experience that this style of working is destructive. It deprives the actual developers of the power to go in an change the design which they will inevitably need, and it tends to produce designs which are, as you say, brittle, in that the factoring into classes is naive to the point where changes to the code will eventually totally destroy the abstraction.
This isn't +3 insightful, its -1 Troll. And well below Mr So-called-lawyer's usual high standards as well.
He hasn't even read the article for god's sake.
Idiots.
Simon - who originally moderated this down, and wants the points back since the battle's clearly lost.
There really are no compelling arguments against libertarianism that I''ve seen.
Sigh. Well this article is so old, noone will ever read this, but I cannot let you get away with saying this without challenge.
There are a great many arguments against libertarianism that hold some water, but if you reason entirely from libertarian principles its possible to condemn just about all of them as authoritarian. I recommened Mike Hubens excellent Site of critiques of libertarianism including his Non-libertarian FAQ for many different views.
The problem is that libertarianism reasoning is a hermetically closed loop of logic, which is in itself free of inconsistencies, but in itself that proves nothing. Nothin about consistency gaurantees good governance, and nothing about the way the principle of libertarianism are derived does either. This is why, to be completely blunt, libertarians are so dogmatic (the statement "there are no good arguments ..." is dogmatic, and arrogant. If you disagree, check a dictionary), and those who try to argue with them get so frustrated.
The logic goes like this: The highest value is freeedom. We must maximise everyone's freedom. This is meant in the negative sense of freedom: a man is free if noone prevents him doing as he pleases with himself and his property if he interferes with noone else. Thus the sole role of the state is the prevention of coercion and fraud.
The first thing to note is that this is a very unusal use of the word "freedom". In general usage people are happy to talk about their freedom at work, or in their families, or to change suppliers for some good. Libertarians, however, assert that if you sign a contract obliging you to do something you have acted freely, and thus if I complain about my lack of freedom to, say, take bathroom breaks at work, this is mere whinging and my freedom has not been affected. After all, I can quite, can't I ? and I signed the contract with my employer in the first place.
To see this, if its not sufficiently clear, consider an employer who sets up separate "whites only" and "blacks only" drinking fountains at work, and fires employess who disregard the separation. Most of us would consider such behaviour abhorent, and most people would not object to a law against it. Libertarians, however, assert that the employer is quite within his rights. I admit I'm pressing emotional buttons to make a point here. I don't imply that libertarianism implies racism, or that libertarians would condone such employment politicies. I do, however, assert that the libertarian idea of freedom is not very close to the common use of the word. IMHO this problem derives from treating property as an extension of the person and essentially absolute. It is better - in my view - to see property as a social phenomenon, a compromise, whose use must be regulated.
Secondly, and along similar lines, libertarianism is not an adequate moral system, as should be clear from the above. At best all it offers is a minimal framework for law. Nothing in fundamental libertarian philosophy prevents one from selling oneself into slavery, for instance.
The only possible justification for libertarianism other than that it is moral is that it is efficient - that it provides the greatest degree of social or economic good of any social system. The arguments here are economic, but they are at the very least inconclusive. You have to believe in the perfection of the unregulated free market to accept them.
First I should point out that I am a libertarian, with a small l, in that my primary interest in politics is to ensure maximum freedom for everyone, but I have a rather broader idea of that than the LP program does, and thus could never support them, even if I were an American. I'm just posting to spew a little random political philosophy. Your argument that we should first assume that new laws are unecessary until proven otherwise rests on totally libertarian assumptions, and this sounds totally reasonable. However, there are other logical (and possibly moral) political positions to take. Its possible to argue that society comes before the individual, both historically and morally, and this argument is sound insofar as there never was a "state of nature" with individuals fending for themselves. Rather, humans have always lived in groups and those groups have always been hierarchical. The state it merely the latest in a long line of forms of social organisation which are hierarchical in that some individuals are acknowledged as leading the rest (within some constraints). Thus the states use of force, in this view, is not the anwarranted and unjustifiable coercion it is in the libertarian view, but rather a necessary (but regrettable) part of the social structure. Not that I agree with this. It simply needs to be pointed out that enlightenment liberal ideas are a recent innovation, and while they may *seem* obvious they are not.
Random points: Its "Anarchy, State, and Utopia", you got a superfluous "the" in there. Its surprisingly readable for academic political philosophy. Its also (IMNAAHO) not so much mistaken as inadequate to the task it sets out to acheive. Nozick has since changed his mind about quite a lot (though not all) of it, and has spent many recent years trying to devise a just system of inheritance taxation. I disagree strongly (regardless of its inadequacy) with his basic premise that a potential explanation is almost as good as a real one.
Nozick is not only a classmate of Rawls, but a friend and an admirer. "Anarchy, State and Utopia" was intended as a counterpoint to Rawls' infinitely duller "A Theory of Justice" (though I suspect Rawls' is a by far the better reasoned theory). Rawls is emphatically not an apologist for the welfare state (based on entitlement) but an advocate of a kind of modernised liberalism (in the classic sense) based on the idea that a system is just if someone would agree to participate in it even if they had no idea what their position in the society would be.
Nice post (and nice sig). My concern with Libertarianism (as opposed to libertarianism (note capitalisation) to which I would more or less adhere) is that it is - like many forms of collectivism - a highly dogmatic and "rationalist" (as opposed to empiricist) system of thought. I think you allude to this when you say that all abstractions will have bugs when implemented by humans.
In some cases libertarianism seems to have become as much a tool for making yourself more equal than others as Marxism became. Take, for example, the (actually objectivist, but hey) claims made elsewhere in the comments for this article that its OK to polute, make noise, and generally behave in an obnoxious manner, because "noone should tell me how to live my life" (and incidentally the LP seems to support this viewpoint). As the previous-but-one poster pointed our, in these cases one must suspect that libertarianism is merely an excuse to behave as one pleases with no regard for others.
Well thats an interesting theory. Personally I'm inclined to think that libertarianism/objectivism is attractive to the tech crowd (and has been attractive to me, though not much now), because it offers a consistent (if you don't look to hard) and clear system for moral behaviour.
I think many techies are disturbed by the woolly and complex nature of most people's moral ideas, and tend to resort to libertarianism (in personal life and politics) because it offers a safe harbour from that wooliness and a clear response to allegations of wrongdoing.
Many libertarians become irate to the point of appearing to panic when their ideas are challenged, especially by someone coming from a logical but more socialistic or conservative framework of ideas. This implies to me that libertarianism is really very important in their worldview, and I suspect this is a tool for cutting away the large swathes of fuzzy, illogical human concern which the more technical mind finds disturbing (I know I do) that libertarianism (and moreso objectivism) says are irrelevant.
And then you go and scatter your code with lclint garbage ? No thanks. Put it in the language.
If that sort fo stuff crashes the JVM, then its Well stuffed. Get a different one. Array out of bounds errors should throw an exception, and Java initialises *all* data to defaults (though for objects, this is nil, which will cause exceptions to be thrown).
Unless you use JNI, or some other kind of native code, a correctly written VM should never crash (though of course, it might *stop*).
This seems to be heresy at the moment, but: why does wireless matter ?
At the moment, the wireless technology in Europe consists of cellphone access to email and the web. Email I can see, but the web ? why ? Who wants to book railway tickets from the bus, or order books from the middle lane of the M25 ? Let alone read some random person's ramblings on their home page on a screen smaller than my fingernails.
I'm not saying wireless is useless. Wireless web access on devices with large enoungh screens to be usable, or access from cellphones to services that would be useful while on the move would both be handy. Cellphone applications especially, but it would rely on having phones with positioning technology.
Applications range from a map showing my current position, to the timetable for the railway station I'm currently standing in, to the location of the nearest McDonalds, or a list of movies showing in town tonight.
None of these things can be done yet. The usefulness of a WML browser on a unadorned cellphone is about nil. You Americans aren't missing anything.
You're quite right that most of an interview depends on getting on with the interviewer. In fact, the importance of every second decreases exponentially as the interview goes on. Many interviewers will have decided whether they'll recommend hiring you or not before you even open your mouth. Indeed, quite a lot of training *to* interview (rather than be interviewed) involves learning to keep an open mind for as long as possible.
However, I do think technical questions at interviews matter. Not necessarily the "How does this work in Java ?" type of questions (which are really just checks that you didn't exaggerate your knowledge on your CV), but problem solving and puzzles. Interviewers for really serious technical posts (not database grinding or HTML hacking) want to see your though processes and problem solving style, both to check that you really can do it, and to check that your "style" is compatible with "how we do it here". This can only be done at interview. On CVs, or over the phone, these things don't come through.
Its definitely true that you get more out of UML the more diagrams you use, and I think we agree that its a reasonable way to communicate a view of a project.
There's an AC post attached to mine that should really have been moderated up, pointing out that Christopher Alexander (on whose architecture work Design Patterns are based) made a distinction between formal and functional description, where the formal description describes what was done, and the functional description says why. I think this is important here: UML is a formal description. It needs to be accompanied by functional description to make useful sense.
My experience with UML is that the diagrams do not serve much useful purpose on their own. They almost always need to be accompanied by a verbal explanation of the problem being solved and the type of solution employed before they start to make sense. Possibly this is why this book does not go across very well. I do not see the addition of colour helping very much.
At best these kinds of diagrams and things are an aid to communication, but they are not a substitute for it, and they certainly are not a substitute for design. There seems to be quite a common belief that the existence of UML diagrams somehow implies that the design they represent is well thought through, or even coherent. It does not.
Copyright grants to the creator of a work the right to control the distribution of his work for profit (or otherwise, as he chooses). Its relevant that this is (in most people's opinion) not a "natural" right, because that makes its scope and form questionable, and up to society (through the state) to control.
Your copyright allows you to make demands of the licensee in exchange for your work. So far, so good. This allows you to impose certain conditions - in the case of the Microsoft EULA, payment, in the case of the GPL/BSDL certain conditions relating to copying and modification.
However, there are limits to what you can do with such a license. In short, you can't deprive the licensee or any other individual of their other rights through it. You can't (to pick an absurd example) require that someone go out and commit a murder in order to use you software.
To get back to the original issue: The question is not whether copyright is good or bad, but what the legitimate scope of a copyright license is. Can you require that only approved devices be used to access the work ? can you require that no copies be made ? Can you override somebody's right to free expression on the grounds that the knowledge they've expressed might be used to undermine your monopoly on players, or copy your work ?
According the doctrine of fair use, which is fundamental to the law in this matter, the answer to the first question is "no", you can't create a monopoly on players in this way, as it has nothing to do with your profiting from your work. To the second "sometimes", you can only control the making of copies where you can reasonably argue that you're being deprived of a sale you'd otherwise have made.
The answer to the third question is clearly the critical matter in this case. To us, the answer is pretty clearly no. The MPAA has no right to maintain its monopoly on players, and no right to restrict information about copying (thats like outlawing the instructions for making a photocopier).
Frankly, I find it disturbing that so many people have such a naive idea of copyright as a form of "ownership" that allows the owners copyright to override everything else.
Its true they have too many people processing for them - or at least that they don't particularly want more, or want to speed up the client. You can check the FAQ on their website There's only a limited amount of data coming from the radio observatories, and they have enough clients to process it.
I don't know of them sending out false data, but they've certainly sent out duplicate data. This, however, is partly a security precaution to catch people who modify the client (either for silly reasons (to get more points), or to speed it up) and break the core algorithms.
Also features in the British mini-series "Eureka Street". I think its been around longer than either.
Simon
I don't think Robin was arguing that employees are not responsible for the actions of their employers. I think he was saying that they *determine* the actions of their employers.
In most large companies, there is a distinct lack of overall strategy coming from the top that actually succeeds in driving the whole organisation. That the point - there are no orders. The actions of Microsoft, or ZDNet, or Slashdot are just the actions of individuals acting more-or-less independently.
Nope. Its only a copyright violation if you copy text verbatim and try to pass it off as your own, or sell it. Summaries, reviews, reports, even properly attributed quotes, are all OK.
/. readers can't get it straight.
Its no wonder, really, that the public is so easily confused about copyright "theft" when even