Linus Torvalds has managed one of the most successful software development programmes ever witnessed and, with git, has sped development up by 2 times or more on the most complexly structured, most variable development team I've ever heard about.
If all you do is democode then Linus doesn't matter, for everybody else his opinion really counts.
What, XP can't see over 4GB on a 32 bit machine? Linux can see up to 36G (IIRC) on 32bit PC hardware.
Fact check time! 32-bit processors can directly address up to 4 GB of memory. This is a limitation of the processor architecture, and has nothing to do with the OS.
I'll see your fact check and raise you one. The 32GB limit is per page directory, that means that there is a 32GB linear *address space* but the 36GB of physical memory has been addressable by the page translation hardware since, I think, the pentium pro. This is a feature intel introduced for databases and other big iron applications which is called "Page Address Extensions". Check out the Intel Architecture Software Development Manuals, I think it's volume 2 that describes the addressing modes (as well as protection systems).
The upshot is that enterprise databases and large numerical systems that do their work across many processes can *really* make use of up to 36GB less OS and driver overheads, the OS might also use some of it for general purpose cache and will use it rather than swapping
I've snipped the most sudden mood swing I've ever seen, I do one little bit of Linux advocacy and get a lecture on the problems Linux has - I'm not a counsellor.
> and the 64-bitness (including finally seeing all 4 GB RAM).
What, XP can't see over 4GB on a 32 bit machine? Linux can see up to 36G (IIRC) on 32bit PC hardware. Perhaps you should have switched to Linux some time ago. Oh yeah, Linux has been running 64 bit on PC hardware for a good few years.
The Beeb is heavily dependent on its relationship with Virgin Media for the UKTV channels and will depend on them to kickstart demand for iPlayer-to-the-TV. Microsoft has a significant stake in Virgin Media, which is constantly working with Microsoft on various projects and depends on them to swell the capital for various developments and services.
Ultimately, the BBC thinks it needs Microsoft because it thinks it needs UKTV and it will do what Microsoft wants as long as that is true.
We are fortunate that Virgin Media is having its spat with BSkyB over Sky One because it makes the BBC slightly more important to it than it would otherwise have been since it needs the UKTV channels to make their basic service as valuable as it is just that little bit more. That in turn slightly decreases the BBCs dependence on Microsoft - probably not enough.
It is interesting that the BBC recently referred to "... the popular Linux..." and also, after months of responding to calls to produce a Linux version of the iPlayer with (paraphrasing) "the BBC is committed to releasing a version for Mac" finally said it would release a version for Linux - but that would depend on the actions of third parties, which I assumed meant that the BBC would release a Linux version if Microsoft agreed to permit Linux to be an option for the British license payer.
> Build it right and with few parts it does a lot. Don't cover things with more and more and more technology for features.
And when you need an extra room, don't convert the loft - just knock the house down and build a new one. Think continuous revenue stream^W^W^Wdifferent.
They do, you just don't know what they do, do for example, generic tuple types require metaprogramming, projects where you need to be sure of the quality of your program needs static typing.
Tuples require those things in C++. To swap two variables in Python, you just write a,b=b,a and you're done. To extract function parameters into a tuple in Perl, you just write my ($first, $second, $third) = @_; and you're done.
Oh, you can have a totally generic and slow tuple type in C++ too, its pretty easy to implement such. The missing thing is something I've been whining about for a while which you show in your example - a syntax for tuplish initialisation and return:
Not the best, but there's the ever-present back-compatibility problem that C++ will always have
I'm not even close to rising to the static typing claim.;-)
You should, it's a valuable discussion. Unit testing only goes so far in a cost-effective way, static typing can be very useful for checking a variety of things with simple declarative syntax (if a little repetitive sometimes) that doesn't then require equivalent unit tests, and also has a useful documentation effect. Unit tests are, of course, important and highly cost-effective for many things but even then, a language like epigram could relegate them to the most high-level of testing.
I can't use the latest eclipse on my 512M dev machine because of the garbage collection. The java allocator cycles through the mapped memory causing constant paging.
You choose a loaded example. For one thing, you're talking about Java, not GC in general. For another, you're talking about Eclipse, not Java programs in general. Go read a few of the papers that have been published over the past 2-3 years about improved approaches to garbage collection. Things can be multiple orders of magnitude faster than they are today in practice, and things like near-real-time performance are also possible. The fact that industrial Java VMs haven't implemented the improved techniques yet doesn't mean they don't exist.
I choose an important example. The problem is the garbage collector in the Sun VM which is the sort of thing C++ would have to have if it had GC all the while until one of these ideal collectors are implemented - if they can even be implemented such that it is easy for the developer (read as "cheap") to guarantee that where object lifetime can be determined statically it is. The problem is also not eclipse, it used only 75-80M of the heap at any time, but the heap was huge and constantly being swapped out. GC has another problem in that lifetime determination can't be used to free resources in anywhere near as neat a way as can be done in non-GC code. But I wouldn't be averse to GC versions of each type where requested by the dev (with the compiler generating wrappers as necessary and avoiding GC where it can determine the lifetime).
Programming in C++ is much more practical when you use funtional programming with function composition, type polymorphism, and its partial support for dependent types instead of OO and object polymorphism.
Or if that's your preference, you could just use a language designed to support those things, and doing so much better, from the start.
I often do, but beggers can't be choosers, sometimes - tools are all part of the equation and a language with poor tools is a poor language. Many of the new C++ features are designed to bring C++ closer to languages like ocaml and haskell with a modicum of type deduction (new auto uses) and concepts (Haskell counterpart -> type classes).
C++ is far too complex yes. But there is nothing that can really replace it. A language which supports functional, generic, procedural, object-oriented programming, with static typing, metaprogramming, and heavily geared towards native building?
The thing is, most development projects that use C++ today don't need all of that.
They do, you just don't know what they do, do for example, generic tuple types require metaprogramming, projects where you need to be sure of the quality of your program needs static typing. Generic is just polymorphic functional programming - ie, you don't need to keep writing reams of code introducing different bugs in each different place. Procedural is needed by most humans to describe the composition of many functions (eg Haskell "do" syntax). C++ supports non-native building just fine. You just need a VM to target, see LLVM as an example. I'll grant you the object-oriented problem - no language should focus as much on supporting OO as C++ does.
But C++ is not a good language for most application development, which is what a great deal of code written in it really is. I think there are several separate but somewhat related reasons for this.
One is the safety argument. Most people simply don't need the flexibility/footshootability [delete as applicable] of C++. You need only look at the much-hyped field of garbage collection to see that (a) many professional developers find this one feature useful that they regard Java as superior to C++ for this reason alone, (b) many more professional developers have no clue about the overheads involved (which are almost zero for typical applications using modern approaches to GC implementation), and most importantly (c) a great many developers using languages without GC make mistakes that developers using languages with GC wouldn't. Similar arguments apply to other routine problems, such as pointers/NULL.
I can't use the latest eclipse on my 512M dev machine because of the garbage collection. The java allocator cycles through the mapped memory causing constant paging. This is the behaviour caused by lazy destruction, which is a subset of leaky programming. GC is a leak that instead of causing you program to terminate causes persistent misbehaviour for the full lifecycle of a program instead of only until the first maintenance release after the leak is reported.
And finally, we have the tools argument. Working with header files sucks, and while just about everyone else is playing with their funky, auto-refactoring, navigation supporting editors, what do we have? VC++ (where refactoring still isn't available in native C++) and Eclipse (which is C++ forced into a Java-like IDE)?
Huh? Eclipse is like Java?
The really scary thing is that reality bites now. It's not like I'm the first person to identify these practical flaws with using C++ for application development. It's not like other people haven't developed languages and tools with all these improvements already. And yet C++ continues to be one of the most important application programming languages today. Why?
Momentum. That's why. Building a new programming language with a syntax that doesn't look like C is asking for trouble; just look at the arguments Python sees over whitespace. (Curiously, I've never seen such complaints made about Haskell. Perhaps this shows a difference between the insight of your average professional programmer vs. your average language geek academic?) More to the point, trying to advocate a new programming language for industrial application development that isn't some form of block-structured, OO-based clone is asking for trouble.
I hate OO, but I quite like C++ (still needs some work though). C++'s worst feature is that it supports near OO usage (and people taught programming between 1990 and 2001 seem to understand nothing *but* OO - yet it is such an ou
I suspect it is. Many people use it like that and have no knowledge of its sole use in the recent past, not even as an alternative.
It is quite possible that "which begs the question" was originally a special case of "which begs: 'what...' ", where a circular argument was being highlighted. It would certainly make sense like that. To qualify the following phrase as a question in the second possible usage causing the same wording as the circular demonstration would be quite reasonable in English, both written and spoken. Due to the popularity of the non-circular usage, the circular demonstration would probably now be more practical when stated as "which begs the original/previous question".
We basically now have a common phrasing to indicate *which* question we now would find useful, if answered.
Each object experiences a drag from the pressure applied to its bow and experiences an equal thrust from the pressure applied to its stern.
It doesn't matter what speed they're going at because they each experience a net "nothing" from the soup - except the casimir effect when another object is close by and, I suppose, a compression.
> If you're referring to the NX bit, 32-bit Linux supports it for K8+ as well.
No, I was referring to the fact that hacks often have to be written for a specific build and a specific arch. Hence the "current" qualified comment because as AMD64 becomes more popular it will be no better off.
> As for performance, I doubt she'd notice the difference, more registers or not.
The difference in generated heat, thus fan noise, can be significant for things like video decoding with the latest advanced codecs. AMD64 really beats out IA32 there. And people want the best regardless of how much computer experience they've had. I don't have to have been driving for 20 years to know that a Mazda 323 feels better to drive than a Peugeot 106 even though it only knocks 30 seconds off my commute.
And in a few years, when management sees the snowball and they say "Why the hell did we switch to Windows, we were already using Linux before?", and you say "Because somebody in upper management who'se now left the company to work for Microsoft told me to change all the computers over" then they say "You're fucking fired, you should have told him about the problems... You're the IT expert, upper management needs you to warn them when they make noises about a change, we can't be all-knowledgeable about the impacts of everything we want to do - we need the info. Get the hell out!"
And that is why you should always request a meeting, get your objections minuted, and keep a copy of the minutes. Because when they say "Who do we blame?" for Linux what they really mean is "If I go with Windows, I can blame YOU!"
> Why'd you give her 64 bit Linux? Does her computer have more than 3 gigs of RAM?
Since 32-bit Linux can use up to 36G of ram I don't think that's the point. Rather higher performance and (current) less susceptibility to hacking are probably the reasons.
The fact that they didn't say that when it would have been very easy to do so may be a useful defence - it could be argued to indicate that professionally advising legislators on intellectual property legislation is the sort of thing that is also supposed to be allowed.
The classic example is: Your buddy says, "You bastard, you slept with my wife!" If this was a lawsuit, you might respond
a. No I didn't! b. You said that I could! c. She wasn't your wife! d. I thought she was someone else! e. I was insane! I thought it was more like:
a. Please realise I've done nothing wrong because I didn't sleep with anyone b. Please realise I've done nothing wrong because, although I didn't sleep with her, you said I could so it would have been okay had I done c. Please realise I've done nothing wrong because if I slept with anyone it couldn't have been your wife d. Please realise I've done nothing wrong because, if I slept with anyone, it could only have happened during "lie about who you are" day so I would have thought it was someone else. e. Please realise I've done nothing wrong because I was insane on the only day I could have slept with anyone so no act of sleeping with someone would have been wrong had they happened at all.
But I thought that if you told the court A, but then you also say that you slept with somebody else (in order to exclude the possibility of it being the wife) to prove C then one of the two things you've told the court is a lie. And lying to the court is a no-no.
So while you can say my A-E since they don't contradict each other, the course of events that you assert to prove each of them must not contradict. Your A-E contradict each other right from the off since some of them tell the court that you *did* sleep with the wife yet some of them tell the court that you didn't - either way lying to the court is contempt and you have to be more careful to do it how I have done.
In my A-E, if you can prove that it was "lie about who you are" day on the day of the alleged act then you've defended yourself and you don't have to prove any of the others. If you can prove that you didn't sleep with anyone then you don't have to prove that it couldn't have been their wife. But nothing stated to the court can be shown to be untrue because all the things I've said could all be true in this world at the same time.
Microsoft has sounded the death knell for Novell/SuSE and Linspire. As the software moves on, they will have to maintain their own clean-room tree's under GPLv2 in order to both continue doing business and to satisfy their contracts with Microsoft. The problem is Novell and SuSE have been extremely good for the Linux using/developing community and I think Microsoft did it deliberately. With a few simple contract deals, Microsoft has significantly dented the appeal of Linux. While that hasn't really been popularly accepted yet due to the euphoria of the GPLv3 attack against Microsoft the economic after-effects will start to trickle through in the next couple of years.
because management hope to "eventually" get rid of the system entirely, they refuse to update it to work on any other browser. That would be on the advice of the IT department. If it is internally updatable IT will want rid of it (IT does not trust software if the company has access to the code and will sell control of the business to avoid flexibility every time), and making the user-crap continue longer and longer is the best way to make management finally get rid of it. The IT department wants buy an annual subscription (cost shared among all departments) to some off the shelf solution that cuts one person a year from their department even though it costs the company 3 people a year - because it makes the head of IT look like he's saved money - and look like everybody else is losing money, giving him/her a bigger bonus and a better chance of joining the board.
After two years of undergraduate linguistics, though it was many years ago, I'd like to point out that if it is one of the most misused it is probably now correct (not that correct means much when talking about a natural language).
"Correct" makes sense when talking about latin because it is not a natural language - latin is a logic game; english, however, is used to satisfy the fashions and fads of any given week and changes at the demand of its users.
Slashdot Mirror
Linus Torvalds has managed one of the most successful software development programmes ever witnessed and, with git, has sped development up by 2 times or more on the most complexly structured, most variable development team I've ever heard about.
If all you do is democode then Linus doesn't matter, for everybody else his opinion really counts.
"Aids for communicating with hospitalized people"
What a cruel and unusual punishment, and I thought communicating with hospitalized people was regarded as something desirable.
The other guy said it didn't, hence moving to x64 windows, so I told him that Linux did... Just advocacy, nothing to write home about.
Except I was waffling on about a 32GB vs 36GB limit instead of 4GB vs 36GB :) That kinda spoiled the moment.
Fact check time! 32-bit processors can directly address up to 4 GB of memory. This is a limitation of the processor architecture, and has nothing to do with the OS.
I'll see your fact check and raise you one. The 32GB limit is per page directory, that means that there is a 32GB linear *address space* but the 36GB of physical memory has been addressable by the page translation hardware since, I think, the pentium pro. This is a feature intel introduced for databases and other big iron applications which is called "Page Address Extensions". Check out the Intel Architecture Software Development Manuals, I think it's volume 2 that describes the addressing modes (as well as protection systems).
The upshot is that enterprise databases and large numerical systems that do their work across many processes can *really* make use of up to 36GB less OS and driver overheads, the OS might also use some of it for general purpose cache and will use it rather than swapping
I've snipped the most sudden mood swing I've ever seen, I do one little bit of Linux advocacy and get a lecture on the problems Linux has - I'm not a counsellor.
> and the 64-bitness (including finally seeing all 4 GB RAM).
What, XP can't see over 4GB on a 32 bit machine? Linux can see up to 36G (IIRC) on 32bit PC hardware. Perhaps you should have switched to Linux some time ago. Oh yeah, Linux has been running 64 bit on PC hardware for a good few years.
The Beeb is heavily dependent on its relationship with Virgin Media for the UKTV channels and will depend on them to kickstart demand for iPlayer-to-the-TV. Microsoft has a significant stake in Virgin Media, which is constantly working with Microsoft on various projects and depends on them to swell the capital for various developments and services.
..." and also, after months of responding to calls to produce a Linux version of the iPlayer with (paraphrasing) "the BBC is committed to releasing a version for Mac" finally said it would release a version for Linux - but that would depend on the actions of third parties, which I assumed meant that the BBC would release a Linux version if Microsoft agreed to permit Linux to be an option for the British license payer.
Ultimately, the BBC thinks it needs Microsoft because it thinks it needs UKTV and it will do what Microsoft wants as long as that is true.
We are fortunate that Virgin Media is having its spat with BSkyB over Sky One because it makes the BBC slightly more important to it than it would otherwise have been since it needs the UKTV channels to make their basic service as valuable as it is just that little bit more. That in turn slightly decreases the BBCs dependence on Microsoft - probably not enough.
It is interesting that the BBC recently referred to "... the popular Linux
> Build it right and with few parts it does a lot. Don't cover things with more and more and more technology for features.
And when you need an extra room, don't convert the loft - just knock the house down and build a new one. Think continuous revenue stream^W^W^Wdifferent.
They do, you just don't know what they do, do for example, generic tuple types require metaprogramming, projects where you need to be sure of the quality of your program needs static typing.
Tuples require those things in C++. To swap two variables in Python, you just write a,b=b,a and you're done. To extract function parameters into a tuple in Perl, you just write my ($first, $second, $third) = @_; and you're done.
Oh, you can have a totally generic and slow tuple type in C++ too, its pretty easy to implement such. The missing thing is something I've been whining about for a while which you show in your example - a syntax for tuplish initialisation and return:
~{int x; y.front()} = tuple_returner(params);
~{int z; a.front()} = {x, y.front()};
Not the best, but there's the ever-present back-compatibility problem that C++ will always have
I'm not even close to rising to the static typing claim. ;-)
You should, it's a valuable discussion. Unit testing only goes so far in a cost-effective way, static typing can be very useful for checking a variety of things with simple declarative syntax (if a little repetitive sometimes) that doesn't then require equivalent unit tests, and also has a useful documentation effect. Unit tests are, of course, important and highly cost-effective for many things but even then, a language like epigram could relegate them to the most high-level of testing.
I can't use the latest eclipse on my 512M dev machine because of the garbage collection. The java allocator cycles through the mapped memory causing constant paging.
You choose a loaded example. For one thing, you're talking about Java, not GC in general. For another, you're talking about Eclipse, not Java programs in general. Go read a few of the papers that have been published over the past 2-3 years about improved approaches to garbage collection. Things can be multiple orders of magnitude faster than they are today in practice, and things like near-real-time performance are also possible. The fact that industrial Java VMs haven't implemented the improved techniques yet doesn't mean they don't exist.
I choose an important example. The problem is the garbage collector in the Sun VM which is the sort of thing C++ would have to have if it had GC all the while until one of these ideal collectors are implemented - if they can even be implemented such that it is easy for the developer (read as "cheap") to guarantee that where object lifetime can be determined statically it is. The problem is also not eclipse, it used only 75-80M of the heap at any time, but the heap was huge and constantly being swapped out. GC has another problem in that lifetime determination can't be used to free resources in anywhere near as neat a way as can be done in non-GC code. But I wouldn't be averse to GC versions of each type where requested by the dev (with the compiler generating wrappers as necessary and avoiding GC where it can determine the lifetime).
Programming in C++ is much more practical when you use funtional programming with function composition, type polymorphism, and its partial support for dependent types instead of OO and object polymorphism.
Or if that's your preference, you could just use a language designed to support those things, and doing so much better, from the start.
I often do, but beggers can't be choosers, sometimes - tools are all part of the equation and a language with poor tools is a poor language. Many of the new C++ features are designed to bring C++ closer to languages like ocaml and haskell with a modicum of type deduction (new auto uses) and concepts (Haskell counterpart -> type classes).
Don'
C++ is far too complex yes. But there is nothing that can really replace it. A language which supports functional, generic, procedural, object-oriented programming, with static typing, metaprogramming, and heavily geared towards native building?
The thing is, most development projects that use C++ today don't need all of that.
They do, you just don't know what they do, do for example, generic tuple types require metaprogramming, projects where you need to be sure of the quality of your program needs static typing. Generic is just polymorphic functional programming - ie, you don't need to keep writing reams of code introducing different bugs in each different place. Procedural is needed by most humans to describe the composition of many functions (eg Haskell "do" syntax). C++ supports non-native building just fine. You just need a VM to target, see LLVM as an example. I'll grant you the object-oriented problem - no language should focus as much on supporting OO as C++ does.
But C++ is not a good language for most application development, which is what a great deal of code written in it really is. I think there are several separate but somewhat related reasons for this.
One is the safety argument. Most people simply don't need the flexibility/footshootability [delete as applicable] of C++. You need only look at the much-hyped field of garbage collection to see that (a) many professional developers find this one feature useful that they regard Java as superior to C++ for this reason alone, (b) many more professional developers have no clue about the overheads involved (which are almost zero for typical applications using modern approaches to GC implementation), and most importantly (c) a great many developers using languages without GC make mistakes that developers using languages with GC wouldn't. Similar arguments apply to other routine problems, such as pointers/NULL.
I can't use the latest eclipse on my 512M dev machine because of the garbage collection. The java allocator cycles through the mapped memory causing constant paging. This is the behaviour caused by lazy destruction, which is a subset of leaky programming. GC is a leak that instead of causing you program to terminate causes persistent misbehaviour for the full lifecycle of a program instead of only until the first maintenance release after the leak is reported.
And finally, we have the tools argument. Working with header files sucks, and while just about everyone else is playing with their funky, auto-refactoring, navigation supporting editors, what do we have? VC++ (where refactoring still isn't available in native C++) and Eclipse (which is C++ forced into a Java-like IDE)?
Huh? Eclipse is like Java?
The really scary thing is that reality bites now. It's not like I'm the first person to identify these practical flaws with using C++ for application development. It's not like other people haven't developed languages and tools with all these improvements already. And yet C++ continues to be one of the most important application programming languages today. Why?
Momentum. That's why. Building a new programming language with a syntax that doesn't look like C is asking for trouble; just look at the arguments Python sees over whitespace. (Curiously, I've never seen such complaints made about Haskell. Perhaps this shows a difference between the insight of your average professional programmer vs. your average language geek academic?) More to the point, trying to advocate a new programming language for industrial application development that isn't some form of block-structured, OO-based clone is asking for trouble.
I hate OO, but I quite like C++ (still needs some work though). C++'s worst feature is that it supports near OO usage (and people taught programming between 1990 and 2001 seem to understand nothing *but* OO - yet it is such an ou
I suspect it is. Many people use it like that and have no knowledge of its sole use in the recent past, not even as an alternative.
It is quite possible that "which begs the question" was originally a special case of "which begs: 'what...' ", where a circular argument was being highlighted. It would certainly make sense like that. To qualify the following phrase as a question in the second possible usage causing the same wording as the circular demonstration would be quite reasonable in English, both written and spoken. Due to the popularity of the non-circular usage, the circular demonstration would probably now be more practical when stated as "which begs the original/previous question".
We basically now have a common phrasing to indicate *which* question we now would find useful, if answered.
They've been watching too much powerpuff girls.
Each object experiences a drag from the pressure applied to its bow and experiences an equal thrust from the pressure applied to its stern.
It doesn't matter what speed they're going at because they each experience a net "nothing" from the soup - except the casimir effect when another object is close by and, I suppose, a compression.
> If you're referring to the NX bit, 32-bit Linux supports it for K8+ as well.
No, I was referring to the fact that hacks often have to be written for a specific build and a specific arch. Hence the "current" qualified comment because as AMD64 becomes more popular it will be no better off.
> As for performance, I doubt she'd notice the difference, more registers or not.
The difference in generated heat, thus fan noise, can be significant for things like video decoding with the latest advanced codecs. AMD64 really beats out IA32 there. And people want the best regardless of how much computer experience they've had. I don't have to have been driving for 20 years to know that a Mazda 323 feels better to drive than a Peugeot 106 even though it only knocks 30 seconds off my commute.
Great! Now I can keep the damn geckos of my windows, I just need to coat them with negative index of refraction materials.
I happened to have a bag of sweet chilli coated nuts in the cupboard and your theory appears to be true because I didn't stay on the ground for long.
Because of the equal and opposite thrust from the virtual particles on the other side of the object.
And in a few years, when management sees the snowball and they say "Why the hell did we switch to Windows, we were already using Linux before?", and you say "Because somebody in upper management who'se now left the company to work for Microsoft told me to change all the computers over" then they say "You're fucking fired, you should have told him about the problems... You're the IT expert, upper management needs you to warn them when they make noises about a change, we can't be all-knowledgeable about the impacts of everything we want to do - we need the info. Get the hell out!"
And that is why you should always request a meeting, get your objections minuted, and keep a copy of the minutes. Because when they say "Who do we blame?" for Linux what they really mean is "If I go with Windows, I can blame YOU!"
> Why'd you give her 64 bit Linux? Does her computer have more than 3 gigs of RAM?
Since 32-bit Linux can use up to 36G of ram I don't think that's the point. Rather higher performance and (current) less susceptibility to hacking are probably the reasons.
The fact that they didn't say that when it would have been very easy to do so may be a useful defence - it could be argued to indicate that professionally advising legislators on intellectual property legislation is the sort of thing that is also supposed to be allowed.
a. No I didn't!
b. You said that I could!
c. She wasn't your wife!
d. I thought she was someone else!
e. I was insane! I thought it was more like:
a. Please realise I've done nothing wrong because I didn't sleep with anyone
b. Please realise I've done nothing wrong because, although I didn't sleep with her, you said I could so it would have been okay had I done
c. Please realise I've done nothing wrong because if I slept with anyone it couldn't have been your wife
d. Please realise I've done nothing wrong because, if I slept with anyone, it could only have happened during "lie about who you are" day so I would have thought it was someone else.
e. Please realise I've done nothing wrong because I was insane on the only day I could have slept with anyone so no act of sleeping with someone would have been wrong had they happened at all.
But I thought that if you told the court A, but then you also say that you slept with somebody else (in order to exclude the possibility of it being the wife) to prove C then one of the two things you've told the court is a lie. And lying to the court is a no-no.
So while you can say my A-E since they don't contradict each other, the course of events that you assert to prove each of them must not contradict. Your A-E contradict each other right from the off since some of them tell the court that you *did* sleep with the wife yet some of them tell the court that you didn't - either way lying to the court is contempt and you have to be more careful to do it how I have done.
In my A-E, if you can prove that it was "lie about who you are" day on the day of the alleged act then you've defended yourself and you don't have to prove any of the others. If you can prove that you didn't sleep with anyone then you don't have to prove that it couldn't have been their wife. But nothing stated to the court can be shown to be untrue because all the things I've said could all be true in this world at the same time.
Microsoft has sounded the death knell for Novell/SuSE and Linspire. As the software moves on, they will have to maintain their own clean-room tree's under GPLv2 in order to both continue doing business and to satisfy their contracts with Microsoft. The problem is Novell and SuSE have been extremely good for the Linux using/developing community and I think Microsoft did it deliberately. With a few simple contract deals, Microsoft has significantly dented the appeal of Linux. While that hasn't really been popularly accepted yet due to the euphoria of the GPLv3 attack against Microsoft the economic after-effects will start to trickle through in the next couple of years.
After two years of undergraduate linguistics, though it was many years ago, I'd like to point out that if it is one of the most misused it is probably now correct (not that correct means much when talking about a natural language).
"Correct" makes sense when talking about latin because it is not a natural language - latin is a logic game; english, however, is used to satisfy the fashions and fads of any given week and changes at the demand of its users.
Stallman's contracts *specifically* grant you the right to his software for any purpose whatever your background/opinions/culture.