Do you believe that the sofware industry will have better luck with these tatics?
No, I think that the SPA is already as successful as it is ever going to be. And for much the same reasons as are stated in the other follow-up to my previous posting. Moreover I don't think that the music industry would be any more successful than the software industry in this regard.
But success isn't really my point. Sure they'll have declining market share and declining profits if they continue in the same vein; but I don't want to live in a country where that kind of business makes itself at home. It is too scarily close to the Corporate Police State. I can't understand why anyone who so clearly believes in freedom would want to bring about such a dystopian society.
This is a bigger picture problem than whether you or I listen to this kind of music. It isn't just music you know, it is any kind of digital product. Software has obviously been trying to live with the copyability of bits since its inception. You've seen the evolution that they've gone through in the last 20 years. First copy protection, through the current SPA actions, lawsuits, nasty shrinkwrap contracts, invasive searching for unlicensed software.
Do you want that to exist in the future for the music industry? Do you want your kid or your granduncle thrown in jail or fined thousands of dollars for copying copyrighted music? Do you want to have to produce on demand licenses showing that you are entitled to play the music on your hard drives? I know this is a bit of science
fiction, but it has already happened once.
Whether you listen to RIAA music or not (I don't) you have to see that such a world isn't the kind of world that you want to live in. It isn't that hard to change it, the cost really should be minimal. Like any regulated thing it becomes a mercantile fixed sized market. There isn't any growth, so there isn't anything to keep businesses there. They will expand into new markets, just as soon as they don't have to worry about losing their old one.
Fuck that. I, for one, rarely, if ever, download music. Why should I subsidize the leeches?
You should want to pay for the leeches so that the industry can collect reasonable profits. Reasonable and statutory pricing means that there is not any need to protect music or other media with encryption, which will ultimately either fail (destroying the entertainment industry in the process), or succeed (destroying the public domain.)
Since neither of those is a desirable outcome you should be willing to pony up a small amount to keep that from happening.
Why everyone instead of just music consumers? I admit the tax could be on 'music capable equipment' instead of on ISPs, but what computer isn't capable of rendering music these days? Any attempt to control who gets access to the data just results in the data being copied beyond those constraints, so that doesn't work too well. The only solution I can see that would work is a general levy.
While I doubt that the music industry has gotten to the point of admitting it yet, what they need to do is lobby congress to protect reasonable profits by establishing a flat license fee. Ultimately the fee would probably be paid by anyone who has an internet account, but the direct means would be through a levy on ISPs as a proportion of their customer base, with certain assumptions about how what percentage of people download music, and how much the download versus buy.
This is the same historically as radio play of music, cable playing of television, or library inclusion of books. There are other ways to do it besides canvassing the ISPs, but any way you look at it, it has to cover the vast majority of online users who could download music.
I really can't see what else they can do. From the technological standpoint, bits copy well.
One of the points that Larry made in the Apocalypse was that the new format allows regex syntax to be expanded in the future without breaking things. Embeddable code in whatever language you might find it written in can be inserted using the same constructs assuming they don't use the sequence ')>' as an operator. In Perl if you wanted to write a regex that calls out to Java I imagine you would write something like:
One option used by many companies is to compartmentalize the data that needs to be secured. HP sells VirtualVault, or you could brew your own by running a chroot jail, or more completely by running a seperate virtual machine like User Mode Linux.
Linux scales pretty well. Maybe not up to the level of Solaris, but it does a quite respectable job on 4 CPU systems. On the other end of the scale Linux scales much better than Solaris though (or perhaps you have Solaris running on your iPAQ and just haven't told anyone about it yet.)
In all honesty I'm not sure that Linux needs to get into data centers. At the E10k and Superdome scalability levels the OS is very much an afterthought. The only thing those companies care about are guarantees of less than 5 minutes downtime a year, and millions to Tpm. I think it is sufficient if Linux is on the network boundary between the backend systems and the Internet, doing what it does well (like giving killer network bandwidth.)
Since I personally saw some of the scenes referred to at the opening of Star Wars at the Old Mill Six theatres in Palo Alto, I know that they are not pure folk tale. I don't know what print they had, but in that theater the first scene of the movie started on Tatooine with Luke looking through electronic binoculars at space, and then speeding into Tashi station to tell his friends (who were unenthusiastic).
The NPR playhouse version adds quite a bit more backstory describing why the ships were there in the first place, and adding extra scenes with Biggs and a speeder canyon race.
None of those were present in the version I saw, but I did see them tease Luke about Biggs, and there were several other changes throughout the movie including changes to the holoprojector
images, and one scene that was cut from the first fight and added to the millenium falcon fight.
Re:CORBA is too heavy & EJB is too RMI/IIOP de
on
Web Services
·
· Score: 2
I've argued that for significant uses you don't even have to agree on the semantics of the data. The producer and consumer of the data can each form seperate semantic meaning for the same data as long as the meanings they form are useful to them.
You for example might produce a document that includes a field 'OrderID' which identifies the order number assoicated with the data in your database. Looking at the data I might have no use or interest in 'OrderID', but still be able to draw out (for example) the name of the customer without any prior agreement as to the meaning of the fields you provided.
The distinction becomes significant when you consider the number and types of communications that can exist. If each one must be standardized or agreed to in advance of use, then XML becomes a much weaker tool for data interchange. Instead someone looking at the data can make an educated guess about the uses that they would like to make of the data you provide, and even if those meanings don't overlap with your meanings, just so long as they get utility from their interpretation the result is still useful.
Possibly true. Certainly I wasn't aware of the
WMI or Windows Script features added to Windows
2000; but then I neither administrate Windows boxes, nor was this information presented under the Win2k 'What's New'. I stand corrected, and am glad to see that Microsoft has finally added some useful features rather than just the glitzy ones.
To answer your question, I've never really used Windows in any serious capacity at all. I have Win2k box on my desktop for getting at some of the corporate web sites that can't be accessed any other way, but mostly I avoid it. Personal preference.
Fast user switching in XP Pro, eh? Let me see ; I can think of at least three ways of doing that on Linux. From the command line the 'su' command has been around ever since Unix became multi-user. Likewise the SGID and SUID bits in the filesystem allow programs to switch your identity to that of the program owner. In either case the X Window system can just as easily open a window as root, as it can from your original user. And if none of those satisfy you then Linux also supports virtual terminals. Log in once as yourself, once as root, and run two desktops under different virtual terminals switching between them with a CTRL-Alt-F7 or CTRL-Alt-F8.
I guess it is my turn to describe you in one word now, but I'm not nearly so much of a boor. Thanks for the information regarding Windows though.
Unix started on expensive time sharing machines, where self protection and security were necessary, multiuser and multitasking from the start. It also ran on different machines. MSDOS started on dinky machines where there was no concept of sharing the machine, thus no security, no multitasking. The hardware grew up to match Unix, whereas MSDOS never grew up to match the hardware.
The proliferation of hardware MMUs was what allowed Unix to move to the desktop, which is surprising considering that when AT&T first split their research off of Multics, one of their main reasons for doing so was to avoid the high end hardware requirements of Multics such as MMUs.
Initially Unix and DOS were much the same. Unix began as a single user operating system with no address space seperation, and no security (and
all Unix flavors to this day can be booted in
single user mode.) But you are right that while
Unix grew to incorporate these features, DOS
never did. These days I think that Microsoft is
blind to the need for multiuser processing on a
single box. Their adoption of the GUI is so complete that they can't imagine someone needing
to adminstrate a headless box.
You read more into my post than I actually wrote. Nowhere did I intimate what Microsoft should do. While I do have opinions on
that subject, I haven't expressed them, and you have incorrectly assumed that they are shallow.
Secondly, what you call selfishness, I call loyalty to my community. You may not like to believe that there is a community of open source developers, but they exist none the less. My personal gain or loss has very little to do with it. The gain or loss of my friends, and how the community could or will be hurt by Microsoft's actions was the subject of my post.
It is a part of the nature of a competition that I would prefer that we win, and they lose. I haven't claimed any impartiality, nor do I particularly want to. And you are incorrect in characterizing me (and by inference my friends) as wanting 'something for nothing'. I want something for something, and I have and will continue to put my own sweat into getting it.
Actually, since the (new) BSD license is compatible with GPL, and compatible with Microsoft, only the
code that interfaces between GPL code and Microsoft
code would have to be licensed under BSD. (As long
as the GPL code never touches the Microsoft APIs and
vice versa, then the two are not linked together, but each is instead linked to an intermediary which is itself compatible with both licenses.
I don't anticipate the Samba developers relicensing their software though. Although
large projects have shown that it is feasible,
I think the Samba developers have a healthy
respect for the protections that they get from
GPL and will not happily give them up. Otherwise
Microsoft could very easily add a new extension
to SMB, and release a similarly configured
extension for Samba, available in binary form
only, and quickly take over control of the
project.
Windows also incorporates the BSD TCP/IP stack, but I for one haven't noticed any
contributions from Redmond being returned to
BSD in exchange for their work. Nor can you take the BSD stack and compile it for windows without
Microsoft's changes. So open source is completely useless when licensed BSD-style for
users of Windows. Want to play with IPv6?
Want to implement IP Masquerading? Want implement filters and firewalls? Want to
implement IPIP encapsulation? Sorry, you'll
have to use another OS even though the
Windows stack came from Open Source.
It is a question of definitions. Microsoft defines Windows as WOSA; the Windows Operating System Architecture. It is their strategy for binding developers to their product.
In a nutshell WOSA is the layer between the device drivers and the application code that allows app developers to ignore the specifics of different hardware, and allows hardware developers to ignore the specifics of different applications.
Microsoft defines WOSA so that they benefit from the mix and match value of many different types of hardware and software, but always just thick enough so that Windows can't be done without.
So when I say 'port Windows to Linux', what I mean is port the device driver interfaces, and the application APIs. Those together are really all that Windows is.
But getting back to the replacing X with Windows idea, while I like it (God knows I've spent more time that I should have had to to get X working right), if you want to use Windows, why bother with Linux at all?
But that is exactly the point. Right now people have a reason to use Linux. I think that well executed, a port of Windows to Linux would have confused the Linux market sufficiently so as to take away any real reason for running it; again allowing a transition to XP without much complaint. Drop Win/Lin after five or six years of support and everyone hooked on the apps would move to XP out of need.
The goal of the strategy I outlined wasn't to make Linux more successful, (why would MS want to do that?), it was to kill it.
I've always thought that Microsoft should port
Windows to Linux. There are a significant
fraction of users who are unhappy with X (for silly reasons IMHO), and Microsoft could easily have killed Wine in its infancy, and have been the major desktop by now if they had tried to. Ultimately that would have been sufficient I think to maintain their control point, no less
control than using Intel hardware.
Even those of us who would prefer to remain on X would probably install Windows for major applications like Office and the result would eventually be a practical dependence on Windows for all of the commercial application base. Owning the API they could then do whatever they
like with the underlying platform; it becomes almost unimportant.
Perhaps I overlook some strategic barrier to such an approach, but luckily even if I'm right Microsoft is too staid to make the leap anyway. I don't expect any Microsoft Linux applications in the near future.
What a fun neologism. I looked up Nerfed on E2, but no luck. For those who don't play EverQuest (or similar games), Nerfed seems to mean that some skill associated with your character class has been rebalanced by the game designers to keep the game fair. I would guess that its etymology lies in Nerf toys; swords, guns and other weapons among them, that are made of foam rubber so that kids can play with them without getting hurt.
The source I found was on Gamespy which includes this delightful description:
The definition of balance tweaking is when the developer adjusts another player's class and makes it a bit less powerful. The definition of a nerf is when it's your class that is tweaked and made less powerful. To give you an example, non-archers in Dark Age shrug and say things like, "Archery was insanely overpowered anyway. It's about time Mythic did something." Archers say, "I CAN'T BELIEVE YOU DID THIS! NOW WE ARE USELESS AS A CLASS. YOU SUCK, MYTHIC!!!!!
FWIW, the US legal term for a youth who is legally recognized as an adult is "Emancipation". There is a pretty good write-up on the legal meaning here.
Emancipation is a legal process in which a minor (a person under 18) petitions the court to have herself declared a legal adult. Emancipation laws vary from state to state, but generally emancipation ends the parents= legal duty to support the minor, and also ends the parents= right to make decisions about the minor=s residence, education, health care, and to control the minor=s conduct. However, it does not mean that the minor is the same as an adult for all purposes (for example, voting and alcohol-purchase age laws are not affected by emancipation). The extent to which an emancipated minor is treated as an adult varies from state to state, but emancipated minors generally can enter into binding contracts, sue and be sued, establish a residence, and consent to medical treatment on the same basis as adults.
In the United States at least this is something that comes up all the time. Child actors for example often have to enter into contracts with the studios producing their movies. I don't know how you would go about doing it in the UK, but here there is a legal process for becoming recognized as a legal adult. (Though it doesn't confer the right to vote, drive or drink, just the ability to enter into a binding contract AFAIK.)
Re:What about inventing a new, non-gnu "language"?
on
Abusing the GPL?
·
· Score: 2
What if I call my obfuscation a new language, 'C--' and I offer to sell a compiler that compiles my C-- language to C?
If the language is obfuscated then you probably
don't use it to maintain your code. If you don't use it to maintain the code yourself, then it obviously isn't the preferred form for modification, thus you would be in violation of the GPL.
On the other hand, if you invent a new language
that links easily to C but isn't C, and distribute the source for your program in that language then indeed the binaries would conform to the GPL even if there is no freely available compiler for that language. Most people don't go to the trouble of writing a new compiler though just to avoid GPL violations.
As far as libraries go, GPL'd code can be linked to proprietary system libraries without violation of the GPL, but that doesn't mean that they can be linked to language specific libraries which are not a part of the operating system.
Interfaces are a subset of multiple inheritance, so they cannot be more powerfull.
Sometimes less is more. Having spent about 5 years writing and designing in C++, and about 3 in Java I tend to agree with the previous poster -- Interfaces are a more useful concept than MI. MI more often gets you into problem situations, and C++ MI doesn't express the Java concept of Interfaces nearly as cleanly as Java does. Java strongly encourages the programmer to seperate interface from implementation; C++ programmers think that they've already done that by writing a header file, but headers in C++ aren't nearly as easily extended as a Java Interface.
Not that I think Java is a perfect language by any means. Personally I think that whoever put garbage collection in the language should be roundly chastised by the thousands of programmers who've had to go hunting for memory leaks, but that is another story.
In all I'd guess that the Manager in question was asking the engineer to pick 'any language so long at it is C++'.
The main slowdown for X doesn't come from Marshalling data. The marshallers are quite fast, and run at CPU speed. The latency mostly grows in the interaction between the X server and the display hardware. Since the X server is running in user mode it doesn't have access to Hardware interrupts. As a result the server ends up polling the hardware registers until the operation is complete, which wastes huge amounts of time. Still a lot faster than blitting video memory in the CPU, but not nearly as fast as Windows can manage.
Not that I'm an expert on Video hardware or anything, but that is paraphrased from an explanation given by X-inside for their X server.
Oh, I don't agree. KDE developers might flock to a
better C++ compiler. And users of AMD chips might be attracted to better optimization code since gcc basically sucks there (instruction schedulers in recent gcc builds are badly dis-optimized for AMD.)
Precompiled headers might be nice too, and the Borland compilers have long had a reputation for
compilation speed which itself is useful.
Personally I remember liking the built in debugger and editor simplifying the compile/edit/debug cycle, so that would tend to
attract me as well; in fact I was considering getting a box myself, and I rarely even have time
to code in C++ any more.
As Richard Nixon learned first hand, a couple of months backlog can be a...um... excruciating source of embarrassment.
I'm surprised at the question though. Are companies really so worried about their business practices that they must destroy evidence in order
to remove liability? I should imagine that internal
auditors would be more effective at keeping a company out of trouble than any policy of document destruction.
Has anyone else noticed that the latest Redhat releases all come with the standard installer 'You must agree to this license to install this software' dialogue? The funny thing is that the license it displays is the GPL, which in turn says 'You don't have to agree to this license', but if you click 'Decline' the installer refuses to install.
So I guess that even Redhat hasn't read their own EULA very closely.
Slashdot Mirror
No, I think that the SPA is already as successful as it is ever going to be. And for much the same reasons as are stated in the other follow-up to my previous posting. Moreover I don't think that the music industry would be any more successful than the software industry in this regard.
But success isn't really my point. Sure they'll have declining market share and declining profits if they continue in the same vein; but I don't want to live in a country where that kind of business makes itself at home. It is too scarily close to the Corporate Police State. I can't understand why anyone who so clearly believes in freedom would want to bring about such a dystopian society.
Do you want that to exist in the future for the music industry? Do you want your kid or your granduncle thrown in jail or fined thousands of dollars for copying copyrighted music? Do you want to have to produce on demand licenses showing that you are entitled to play the music on your hard drives? I know this is a bit of science fiction, but it has already happened once.
Whether you listen to RIAA music or not (I don't) you have to see that such a world isn't the kind of world that you want to live in. It isn't that hard to change it, the cost really should be minimal. Like any regulated thing it becomes a mercantile fixed sized market. There isn't any growth, so there isn't anything to keep businesses there. They will expand into new markets, just as soon as they don't have to worry about losing their old one.
You should want to pay for the leeches so that the industry can collect reasonable profits. Reasonable and statutory pricing means that there is not any need to protect music or other media with encryption, which will ultimately either fail (destroying the entertainment industry in the process), or succeed (destroying the public domain.) Since neither of those is a desirable outcome you should be willing to pony up a small amount to keep that from happening.
Why everyone instead of just music consumers? I admit the tax could be on 'music capable equipment' instead of on ISPs, but what computer isn't capable of rendering music these days? Any attempt to control who gets access to the data just results in the data being copied beyond those constraints, so that doesn't work too well. The only solution I can see that would work is a general levy.
This is the same historically as radio play of music, cable playing of television, or library inclusion of books. There are other ways to do it besides canvassing the ISPs, but any way you look at it, it has to cover the vast majority of online users who could download music.
I really can't see what else they can do. From the technological standpoint, bits copy well.
One option used by many companies is to compartmentalize the data that needs to be secured. HP sells VirtualVault, or you could brew your own by running a chroot jail, or more completely by running a seperate virtual machine like User Mode Linux.
In all honesty I'm not sure that Linux needs to get into data centers. At the E10k and Superdome scalability levels the OS is very much an afterthought. The only thing those companies care about are guarantees of less than 5 minutes downtime a year, and millions to Tpm. I think it is sufficient if Linux is on the network boundary between the backend systems and the Internet, doing what it does well (like giving killer network bandwidth.)
Since I personally saw some of the scenes referred to at the opening of Star Wars at the Old Mill Six theatres in Palo Alto, I know that they are not pure folk tale. I don't know what print they had, but in that theater the first scene of the movie started on Tatooine with Luke looking through electronic binoculars at space, and then speeding into Tashi station to tell his friends (who were unenthusiastic).
The NPR playhouse version adds quite a bit more backstory describing why the ships were there in the first place, and adding extra scenes with Biggs and a speeder canyon race.
None of those were present in the version I saw, but I did see them tease Luke about Biggs, and there were several other changes throughout the movie including changes to the holoprojector images, and one scene that was cut from the first fight and added to the millenium falcon fight.
I've argued that for significant uses you don't even have to agree on the semantics of the data. The producer and consumer of the data can each form seperate semantic meaning for the same data as long as the meanings they form are useful to them.
You for example might produce a document that includes a field 'OrderID' which identifies the order number assoicated with the data in your database. Looking at the data I might have no use or interest in 'OrderID', but still be able to draw out (for example) the name of the customer without any prior agreement as to the meaning of the fields you provided.
The distinction becomes significant when you consider the number and types of communications that can exist. If each one must be standardized or agreed to in advance of use, then XML becomes a much weaker tool for data interchange. Instead someone looking at the data can make an educated guess about the uses that they would like to make of the data you provide, and even if those meanings don't overlap with your meanings, just so long as they get utility from their interpretation the result is still useful.
Possibly true. Certainly I wasn't aware of the WMI or Windows Script features added to Windows 2000; but then I neither administrate Windows boxes, nor was this information presented under the Win2k 'What's New'. I stand corrected, and am glad to see that Microsoft has finally added some useful features rather than just the glitzy ones.
To answer your question, I've never really used Windows in any serious capacity at all. I have Win2k box on my desktop for getting at some of the corporate web sites that can't be accessed any other way, but mostly I avoid it. Personal preference.
Fast user switching in XP Pro, eh? Let me see ; I can think of at least three ways of doing that on Linux. From the command line the 'su' command has been around ever since Unix became multi-user. Likewise the SGID and SUID bits in the filesystem allow programs to switch your identity to that of the program owner. In either case the X Window system can just as easily open a window as root, as it can from your original user. And if none of those satisfy you then Linux also supports virtual terminals. Log in once as yourself, once as root, and run two desktops under different virtual terminals switching between them with a CTRL-Alt-F7 or CTRL-Alt-F8.
I guess it is my turn to describe you in one word now, but I'm not nearly so much of a boor. Thanks for the information regarding Windows though.
The proliferation of hardware MMUs was what allowed Unix to move to the desktop, which is surprising considering that when AT&T first split their research off of Multics, one of their main reasons for doing so was to avoid the high end hardware requirements of Multics such as MMUs.
Initially Unix and DOS were much the same. Unix began as a single user operating system with no address space seperation, and no security (and all Unix flavors to this day can be booted in single user mode.) But you are right that while Unix grew to incorporate these features, DOS never did. These days I think that Microsoft is blind to the need for multiuser processing on a single box. Their adoption of the GUI is so complete that they can't imagine someone needing to adminstrate a headless box.
Secondly, what you call selfishness, I call loyalty to my community. You may not like to believe that there is a community of open source developers, but they exist none the less. My personal gain or loss has very little to do with it. The gain or loss of my friends, and how the community could or will be hurt by Microsoft's actions was the subject of my post.
It is a part of the nature of a competition that I would prefer that we win, and they lose. I haven't claimed any impartiality, nor do I particularly want to. And you are incorrect in characterizing me (and by inference my friends) as wanting 'something for nothing'. I want something for something, and I have and will continue to put my own sweat into getting it.
I don't anticipate the Samba developers relicensing their software though. Although large projects have shown that it is feasible, I think the Samba developers have a healthy respect for the protections that they get from GPL and will not happily give them up. Otherwise Microsoft could very easily add a new extension to SMB, and release a similarly configured extension for Samba, available in binary form only, and quickly take over control of the project.
Windows also incorporates the BSD TCP/IP stack, but I for one haven't noticed any contributions from Redmond being returned to BSD in exchange for their work. Nor can you take the BSD stack and compile it for windows without Microsoft's changes. So open source is completely useless when licensed BSD-style for users of Windows. Want to play with IPv6? Want to implement IP Masquerading? Want implement filters and firewalls? Want to implement IPIP encapsulation? Sorry, you'll have to use another OS even though the Windows stack came from Open Source.
It is a question of definitions. Microsoft defines Windows as WOSA; the Windows Operating System Architecture. It is their strategy for binding developers to their product.
In a nutshell WOSA is the layer between the device drivers and the application code that allows app developers to ignore the specifics of different hardware, and allows hardware developers to ignore the specifics of different applications.
Microsoft defines WOSA so that they benefit from the mix and match value of many different types of hardware and software, but always just thick enough so that Windows can't be done without.
So when I say 'port Windows to Linux', what I mean is port the device driver interfaces, and the application APIs. Those together are really all that Windows is.
But getting back to the replacing X with Windows idea, while I like it (God knows I've spent more time that I should have had to to get X working right), if you want to use Windows, why bother with Linux at all?
But that is exactly the point. Right now people have a reason to use Linux. I think that well executed, a port of Windows to Linux would have confused the Linux market sufficiently so as to take away any real reason for running it; again allowing a transition to XP without much complaint. Drop Win/Lin after five or six years of support and everyone hooked on the apps would move to XP out of need.
The goal of the strategy I outlined wasn't to make Linux more successful, (why would MS want to do that?), it was to kill it.
Even those of us who would prefer to remain on X would probably install Windows for major applications like Office and the result would eventually be a practical dependence on Windows for all of the commercial application base. Owning the API they could then do whatever they like with the underlying platform; it becomes almost unimportant.
Perhaps I overlook some strategic barrier to such an approach, but luckily even if I'm right Microsoft is too staid to make the leap anyway. I don't expect any Microsoft Linux applications in the near future.
What a fun neologism. I looked up Nerfed on E2, but no luck. For those who don't play EverQuest (or similar games), Nerfed seems to mean that some skill associated with your character class has been rebalanced by the game designers to keep the game fair. I would guess that its etymology lies in Nerf toys; swords, guns and other weapons among them, that are made of foam rubber so that kids can play with them without getting hurt.
The source I found was on Gamespy which includes this delightful description:
In the United States at least this is something that comes up all the time. Child actors for example often have to enter into contracts with the studios producing their movies. I don't know how you would go about doing it in the UK, but here there is a legal process for becoming recognized as a legal adult. (Though it doesn't confer the right to vote, drive or drink, just the ability to enter into a binding contract AFAIK.)
What if I call my obfuscation a new language, 'C--' and I offer to sell a compiler that compiles my C-- language to C?
If the language is obfuscated then you probably don't use it to maintain your code. If you don't use it to maintain the code yourself, then it obviously isn't the preferred form for modification, thus you would be in violation of the GPL.
On the other hand, if you invent a new language that links easily to C but isn't C, and distribute the source for your program in that language then indeed the binaries would conform to the GPL even if there is no freely available compiler for that language. Most people don't go to the trouble of writing a new compiler though just to avoid GPL violations.
As far as libraries go, GPL'd code can be linked to proprietary system libraries without violation of the GPL, but that doesn't mean that they can be linked to language specific libraries which are not a part of the operating system.
Sometimes less is more. Having spent about 5 years writing and designing in C++, and about 3 in Java I tend to agree with the previous poster -- Interfaces are a more useful concept than MI. MI more often gets you into problem situations, and C++ MI doesn't express the Java concept of Interfaces nearly as cleanly as Java does. Java strongly encourages the programmer to seperate interface from implementation; C++ programmers think that they've already done that by writing a header file, but headers in C++ aren't nearly as easily extended as a Java Interface.
Not that I think Java is a perfect language by any means. Personally I think that whoever put garbage collection in the language should be roundly chastised by the thousands of programmers who've had to go hunting for memory leaks, but that is another story.
In all I'd guess that the Manager in question was asking the engineer to pick 'any language so long at it is C++'.
Not that I'm an expert on Video hardware or anything, but that is paraphrased from an explanation given by X-inside for their X server.
Personally I remember liking the built in debugger and editor simplifying the compile/edit/debug cycle, so that would tend to attract me as well; in fact I was considering getting a box myself, and I rarely even have time to code in C++ any more.
I'm surprised at the question though. Are companies really so worried about their business practices that they must destroy evidence in order to remove liability? I should imagine that internal auditors would be more effective at keeping a company out of trouble than any policy of document destruction.
So I guess that even Redhat hasn't read their own EULA very closely.