My point is that as long as RedHat releases the source to any of their custom libraries we can easily run software that is "RedHat" specific on the distribution of our choice. It might require a little extra work, but not really that much extra work.
Why in the world should RedHat encourage ESRI to create packages for their competitors? RedHat has done their part by becoming LSB certified (they didn't have to do that, and if they hadn't become certified the LSB would be deader than a doornail).
RedHat has gained their lead by writing cool software and giving it away. Most of their erstwhile competition (ie Caldera, SuSE, TurboLinux) tried to lock their customers into proprietary software that they layered on top of Linux. RedHat, on the other hand, gave their tools away. RedHat's tools, and the RedHat distribution, became the most popular despite the fact that Caldera, SuSE, and others often had better tools. The difference was that RedHat's tools were Free.
Personally, I use Debian Linux, but I am tired of all of the bellyaching by the other commercial distributions.
I agree that this is a problem if these databases become universal. If police in Delaware can read about potential suspects in Arizona then there is a definite problem. If, on the other hand, this database is used in lieu of a bulletin board for small units of policemen then it could be very beneficial without being overly dangerous. As a rule of thumb I would say that as long as all of the policemen submitting and using the information know of each other (at least by reputation) then it is no more dangerous then policemen sitting at a donut shop comparing notes. This way it is easy to self-regulate against police officers using the system for their own ends. If an officer creates lots of potential suspects that are of poor quality it is easy to check into why he or she might be adding these people.
The question is simple. Would Caldera change their name if there Linux strategy was working?
Nope.
So instead of chasing the future, which is clearly Linux, Caldera err... SCO Group is going to focus squeezing the last few pennies out of their customers that are too dumb to have migrated to Linux. That's a clever strategy.
Yes, but you also have the right to an attorney present when you are questioned.
Which is why when the policeman asks for your license and registration, you comply. When he asks to take a look in your trunk, you tell him that you want to call your lawyer.
For the most part, however, the simplest thing to do is to simply not go out of your way to look suspicious. People that deliberately dress like gangsters shouldn't be surprised when the police mistake them for one. And don't give me any crap about that being racist. There are plenty of black people that don't look like criminals.
Don't get me wrong, I agree with you. I just believe that you can't blame the police for trying. In close knit groups such a database would not even be problematic. It would simply be a place for fellow policemen to share notes on cases. There is nothing diabolical there. The problem arises when the database is being used by policemen who don't personally know the guys submitting profiles. When a policeman can create "potential suspects" without any sort of check of balance that is a definite danger.
Personally I am of the belief that proper law enforcement starts with the community. I no longer live in a rural community, but I do live in a community that is close knit enough that "suspicious" people get reported to the police. On several occassions our local Neighborhood Watch has helped catch criminals. All it takes is a little bit of time to get to know your neighbors, and a willingness to report suspicious behavior.
In neighborhoods where this type of thing isn't possible it is no wonder that the police are trying to share notes. If they keep their databases localized, and don't share them between units I don't even see a problem with these databases.
This is nothing more than an extension of what is already being done in the law enforcement communities. In smaller communities, like the one where I used to live, police could simply memorize the names of the folks that they suspected of being "on a bad path."
As an example, I had a buddy that had a string of DUIs and got his license suspended. He was later pulled over by a policeman because his license plate light wasn't bright enough (or some other excuse) and given yet another hefty fine for driving without a license. He couldn't believe that the police would pull him over for such a piddly deal, especially considering the amount of traffic that has faulty lights on their vehicles (stand on a streetcorner and count sometime, you will be amazed). It goes without saying that years of alcohol abuse had severely effected my friend's thought processes. The policeman hadn't pulled him over because of a minor infraction. The policeman had pulled him over because he recognized the automobile!
In small towns police do this all of the time. They know who the criminals are, and they know that a quick sweep of everyone they are keeping tabs on (who isn't currently locked up) will generally net them their criminal. Of course, small town dwellers tend to understand that when the go out "in public" they are quite likely to be recognized by the people they encounter. We realize that none of us have a right to anonymity.
What these big city policeman are doing is simply this same principle on a larger scale. They want to be able to "remember" the people (as a group) that they thought were suspicious. Now, whether this is right or not is hard to say. All I can say is that profiling potential criminals in this matter has been working quite well in small communities since the beginning of time.
All things considered, I would strongly encourage folks that live in this part of the U.S. to take care to not dress like a stereotypical drug dealer.
Depending on how big and solvent Sigma is and how straightforward the case then it should be possible to get a lawyer to work on it for a share of the award. The XVID folks are almost certainly going to be able to get damages (if they win).
Lawyers are more than happy to work for a percentage on cases that have the potential for large damages.
The good news is that even stupid people want free stuff. Imagine the howls of derision when the rube realizes "Dude! I just bought a PC that can't play MP3s."
It's obvious to you and I that the DMCA is bad news, but that's because we happen to be part of the fringe that is currently effected by it (no, it's not your superior intellect that has clued you into this particular problem). Fortunately, Hollywood is likely to continue to push their draconian vision on America to the point where your home videos won't play on your television and where it is impossible to tape Oprah and watch it later. At that point the public will wake up and smack Hollywood in the head with a clue-by-four.
Hey, I like vi, but saying that a modal editor is more "intuitive" than Emacs is stretching it a bit. Besides, unless you happen to be editing text on a 386 in text mode you can close either one of them with a File->Exit, or by clicking on the "X" in the upper right hand corner (assuming your window manager doesn't use a different button style).
The difference is that Linux doesn't have a C# compiler. Of course, if the C++ course targetted Windows it would essentially be the same thing. I still would have to use Microsoft's tools. Oh well, I suppose if the students don't mind, why should I?
Now that's funny. Either you choose to target Mono or.NET, or you choose to only use the subsets of each language that overlap and that are completely identical in their behavior. Just like there are subtle differences between JVMs and sometimes even between the same JVM on different platforms there are also likely to be large differences in the Mono and the.NET CLR. This is especially true since the Mono CLR is still in heavy development. Heck, it's barely self-hosting. Not to mention the fact that useful parts of.NET like WinForms and ADO.Net are not part of the ECMA standard.
Now, don't take me for a Java bigot, I don't like Java either, but at least I can write, compile, and test Java programs on my Linux box. I can't do that currently with.NET, and I doubt that.NET and Mono will ever be 100% compatible.
To be honest, I think that what Miguel is doing with Mono sounds very interesting, and what I have seen of C# has also been impressive. Miguel has a long track record of very successful projects, and so I won't be one bit surprised to see Miguel succeed here as well.
In the meantime, I am not the least bit interested in programming in Java again (I currently work with Python with a bit C mixed in), but I would be interested in Mono once it's soup. Especially if they ever get around to creating a C# implementation of Python (like Jython). I think that such a beast would be a compelling development environment even if it wasn't completely compatible with.NET. I certainly don't consider myself a Mono basher.
The problem is that the classes are almost certainly going to require VS.Net, and the beginning classes will probably end up consisting of little more than learning to use VS. Personally, I think that is a shame, but then it's not my school or my tuition.
Yes, I know about Mono, and I know about SharpDevelop. And if the University of Waterloo allowed you to hand in assignments that targetted Mono then I would not complain.
For one thing a Java compiler and JVM exists for every platform that you can name. And you don't have to have a special development environment to hack Java. All you need is a Java compiler, a JVM and a text editor. I am not interested in paying money for Microsoft's tools, nor am I interested in booting into Windows to use them.
Re:Yes, but why does Microsoft need a stand...
on
Linuxworld Fun
·
· Score: 2
Good point. Now that I think about it I know plenty of developers that target Linux servers but write their software on Windows machines. Having the UNIX software on the development box is probably an advantage as well. I have been developing at a Linux box so long that I can't imagine using Windows, but plenty of people do.
Of course, I always figured that this was what Cygwin was for. Is Interix (or whatever Microsoft is calling it) so much better than Cygwin?
My guess is that this problem will get solved as more and more Chinese folks use Linux. As the Chinese government begins to actually roll out Linux they will undoubtedly notice that input could be slicker, and someone will fix it. Either it will be Sun trying to sell the Chinese StarOffice, or it will be hackers at Red Flag, or some Chinese college student trying to make a name for himself, but it is sure to happen. One thing is certain about Free software and that is the more use it gets the more polished it becomes. Right now Chinese input is not a big deal in Linux. Few Chinese people actually use Linux, and those that do have gotten used to the various hacks Chinese input requires. My guess is that it will become a big deal soon.
I am not Chinese, but it would seem the answer is fairly obvious. The primary reason that China is looking at Free Software is that Free Software is less expensive than the alternatives. This sort of thing didn't matter before the WTO started pressing China to stamp out software piracy, but it does now.
The second reason that Free Software is advantageous to the Chinese is that it allows them to bootstrap their own software economy instead being second or third class citizens in an American-led software economy. Their are plenty of bright folks in China who can write software. China would much rather put them to work than to pay software developers from overseas to do the work for them. The fact that Chinese developers are far less expensive than American ones doesn't hurt either.
The third reason has to do with Chinese national security. China has no idea what is in most U.S. written commercial software, but they do know that versions of Excell shipped with a flight simulator, and that before it was GPLed Interbase had a backdoor password for years. It's hard enough trusting commercial software on the very best of days, but trusting commercial software written by foreign nationals is a very sticky subject if you happen to be the Chinese government.
One thing is certain, China is not afraid of Microsoft. Microsoft and the BSA might seem scary to companies in the United States, but China is a sovereign nation (and a powerful one at that). If the BSA got too pushy the Chinese government could run their representatives over with tanks and there would be nothing that the BSA could do about it. China is cleaning up its act as regards to software piracy only because the U.S. has threatened to put sanctions on Chinese trade if they didn't. The U.S. market is important to the Chinese, and so they are trying to comply.
That's the question I would like to see answered as well. How many machines running Websphere and DB2 will it take to replace their existing PHP + PostgreSQL setup?
First of all Sourceforge is shifting to DB2, not Oracle, so you are still SOL. I certainly understand your point. If you have a large investment in Oracle, it makes sense to use Oracle software. I agree that sometimes Oracle on big iron is the correct solution to the problem. Sometimes you simply pay your money and get the best. You just assumed that I was slamming commercial software because this is/.
That being said, sourceforge has been remarkably stable, certainly stable enough for the development efforts of any enterprise I can think of. After all, if you can't get to your corporate sourceforge portal because its down you probably don't lose "millions of dollars." And even if you did lose millions of dollars everytime your sourceforge portal went offline there still isn't any evidence that the as yet to be written version that uses Websphere and DB2 is going to be more stable than the version that sourceforge is using to support thousands of developers on sourceforge.net. Websphere and DB2 are both good products, but the new sourceforge could be awful.
Both of us know that commercial software does not mean software that is completely bug free. In fact, I would rather trust something that comes with source and has a large user base over a brand new commercial product any day of the week. Time tested mature products like Oracle, however, are a different beast altogether. Clearly Oracle has a well-deserved reputation for stability.
Re:Yes, but why does Microsoft need a stand...
on
Linuxworld Fun
·
· Score: 3, Informative
Microsoft's Services for UNIX allows you to recompile your Linux programs for Windows. Of course, you get to pay extra for this privilege, and your UNIX services will feel "bolted on." Besides, Linux and Windows already play pretty well (no thanks to Microsoft).
What Microsoft is trying to do with UNIX services for Windows is give UNIX users a way to migrate to Windows, pure and simple.
I agree with you that this is a perfectly fair way to play, and I also agree that it isn't likely to work. After all, who is going to take a working Linux application and move it to a Windows box? That makes no sense at all.
Re:Slashdot to change?
on
Linuxworld Fun
·
· Score: 3, Insightful
No, they want to have the corporate version of sourceforge run on DB2 and WebSphere. My guess is that VA Software won't be migrating sourceforge.net. Like you said early sourceforge.net handles a ton of activity, and it does it with PostgreSQL and PHP. Migrating to a totally new technology would almost certainly cause problems, and the last thing that IBM wants is for VA Software to switch sourceforge.net from Free Software to their expensive commercial software and have the new setup be buggy or have performance problems.
The corporate accounts paying for the commercial version of sourceforge are undoubtedly dumber than a box of rocks. They would rather purchase a reimplementation of sourceforge.net on a different technology base than download the software that is good enough for sourceforge.net for free and learn to set it up themselves. Heck, they could even pay someone else to set it up for them (the folks at savannah.gnu.org have some experience in setting up sourceforge). In my opinion these are precisely the type of customers to target. It is almost trivially easy to make money from people who have more money than sense.
If you think that the man cutting your hair is gay then you are almost certainly not in a barbershop. There is nothing remotely feminine about a barbershop. Heck, the one I frequent has a "No Spitting" sign on the door, a baseball game on the old television. Also, my barber hasn't used scissors since 1947. Apparently scissors are for girls.
The reason that you would "make it visible" in this particular case is because you want to be able to pay the tolls on the bridge without stopping and handing over coins. In other words, your new visibility is a feature. You have placed a transponder on your car for the express purpose of being recognized by the authorities. You could leave the house with a bag over your head so people wouldn't recognize you, but being recognized is generally thought of as a good thing.
Unless of course, it means being caught doing something illegal or unethical.
Slashdot Mirror
I pity you fools that spent months encoding all of your music as MP3s.
My point is that as long as RedHat releases the source to any of their custom libraries we can easily run software that is "RedHat" specific on the distribution of our choice. It might require a little extra work, but not really that much extra work.
Why in the world should RedHat encourage ESRI to create packages for their competitors? RedHat has done their part by becoming LSB certified (they didn't have to do that, and if they hadn't become certified the LSB would be deader than a doornail).
RedHat has gained their lead by writing cool software and giving it away. Most of their erstwhile competition (ie Caldera, SuSE, TurboLinux) tried to lock their customers into proprietary software that they layered on top of Linux. RedHat, on the other hand, gave their tools away. RedHat's tools, and the RedHat distribution, became the most popular despite the fact that Caldera, SuSE, and others often had better tools. The difference was that RedHat's tools were Free.
Personally, I use Debian Linux, but I am tired of all of the bellyaching by the other commercial distributions.
I agree that this is a problem if these databases become universal. If police in Delaware can read about potential suspects in Arizona then there is a definite problem. If, on the other hand, this database is used in lieu of a bulletin board for small units of policemen then it could be very beneficial without being overly dangerous. As a rule of thumb I would say that as long as all of the policemen submitting and using the information know of each other (at least by reputation) then it is no more dangerous then policemen sitting at a donut shop comparing notes. This way it is easy to self-regulate against police officers using the system for their own ends. If an officer creates lots of potential suspects that are of poor quality it is easy to check into why he or she might be adding these people.
The question is simple. Would Caldera change their name if there Linux strategy was working?
Nope.
So instead of chasing the future, which is clearly Linux, Caldera err... SCO Group is going to focus squeezing the last few pennies out of their customers that are too dumb to have migrated to Linux. That's a clever strategy.
Yes, but you also have the right to an attorney present when you are questioned.
Which is why when the policeman asks for your license and registration, you comply. When he asks to take a look in your trunk, you tell him that you want to call your lawyer.
For the most part, however, the simplest thing to do is to simply not go out of your way to look suspicious. People that deliberately dress like gangsters shouldn't be surprised when the police mistake them for one. And don't give me any crap about that being racist. There are plenty of black people that don't look like criminals.
Don't get me wrong, I agree with you. I just believe that you can't blame the police for trying. In close knit groups such a database would not even be problematic. It would simply be a place for fellow policemen to share notes on cases. There is nothing diabolical there. The problem arises when the database is being used by policemen who don't personally know the guys submitting profiles. When a policeman can create "potential suspects" without any sort of check of balance that is a definite danger.
Personally I am of the belief that proper law enforcement starts with the community. I no longer live in a rural community, but I do live in a community that is close knit enough that "suspicious" people get reported to the police. On several occassions our local Neighborhood Watch has helped catch criminals. All it takes is a little bit of time to get to know your neighbors, and a willingness to report suspicious behavior.
In neighborhoods where this type of thing isn't possible it is no wonder that the police are trying to share notes. If they keep their databases localized, and don't share them between units I don't even see a problem with these databases.
This is nothing more than an extension of what is already being done in the law enforcement communities. In smaller communities, like the one where I used to live, police could simply memorize the names of the folks that they suspected of being "on a bad path."
As an example, I had a buddy that had a string of DUIs and got his license suspended. He was later pulled over by a policeman because his license plate light wasn't bright enough (or some other excuse) and given yet another hefty fine for driving without a license. He couldn't believe that the police would pull him over for such a piddly deal, especially considering the amount of traffic that has faulty lights on their vehicles (stand on a streetcorner and count sometime, you will be amazed). It goes without saying that years of alcohol abuse had severely effected my friend's thought processes. The policeman hadn't pulled him over because of a minor infraction. The policeman had pulled him over because he recognized the automobile!
In small towns police do this all of the time. They know who the criminals are, and they know that a quick sweep of everyone they are keeping tabs on (who isn't currently locked up) will generally net them their criminal. Of course, small town dwellers tend to understand that when the go out "in public" they are quite likely to be recognized by the people they encounter. We realize that none of us have a right to anonymity. What these big city policeman are doing is simply this same principle on a larger scale. They want to be able to "remember" the people (as a group) that they thought were suspicious. Now, whether this is right or not is hard to say. All I can say is that profiling potential criminals in this matter has been working quite well in small communities since the beginning of time. All things considered, I would strongly encourage folks that live in this part of the U.S. to take care to not dress like a stereotypical drug dealer.
Depending on how big and solvent Sigma is and how straightforward the case then it should be possible to get a lawyer to work on it for a share of the award. The XVID folks are almost certainly going to be able to get damages (if they win).
Lawyers are more than happy to work for a percentage on cases that have the potential for large damages.
The good news is that even stupid people want free stuff. Imagine the howls of derision when the rube realizes "Dude! I just bought a PC that can't play MP3s."
It's obvious to you and I that the DMCA is bad news, but that's because we happen to be part of the fringe that is currently effected by it (no, it's not your superior intellect that has clued you into this particular problem). Fortunately, Hollywood is likely to continue to push their draconian vision on America to the point where your home videos won't play on your television and where it is impossible to tape Oprah and watch it later. At that point the public will wake up and smack Hollywood in the head with a clue-by-four.
Hey, I like vi, but saying that a modal editor is more "intuitive" than Emacs is stretching it a bit. Besides, unless you happen to be editing text on a 386 in text mode you can close either one of them with a File->Exit, or by clicking on the "X" in the upper right hand corner (assuming your window manager doesn't use a different button style).
The difference is that Linux doesn't have a C# compiler. Of course, if the C++ course targetted Windows it would essentially be the same thing. I still would have to use Microsoft's tools. Oh well, I suppose if the students don't mind, why should I?
jearl@porter:~$ csc
bash: csc: command not found
jearl@porter:~$ apt-cache search csc
cbrowser - a C/C++ source code indexing, querying and browsing tool
cscope - Interactively examine a C program source
libcteco50000 - Orga Eco 5000 smartcard reader PCSC and CT-API driver
libgempc410 - PC/SC driver for the GemPC 410 smart card reader
libgempc430 - PC/SC driver for the GemPC 430 smart card reader
libpcsc-perl - Perl interface to the PC/SC smart card library
libpcsclite-dev - PCSC Lite client development files
libpcsclite0 - PCSC Lite client library
libslbreflex2 - Reflex 62/64 smartcard reader PCSC and CT-API driver
libstring-approx-perl - Perl extension for approximate matching (fuzzy matching)
libtowitoko2 - Towitoko smartcard reader PCSC and CT-API driver
pcsc-tools - Some tools to be used with smart cards and PC/SC
pcscd - PCSC Lite resource manager daemon
slib - Portable Scheme library.
I apparently don't have a csc compiler available. Perhaps it's in non-free?
Now that's funny. Either you choose to target Mono or .NET, or you choose to only use the subsets of each language that overlap and that are completely identical in their behavior. Just like there are subtle differences between JVMs and sometimes even between the same JVM on different platforms there are also likely to be large differences in the Mono and the .NET CLR. This is especially true since the Mono CLR is still in heavy development. Heck, it's barely self-hosting. Not to mention the fact that useful parts of .NET like WinForms and ADO.Net are not part of the ECMA standard.
Now, don't take me for a Java bigot, I don't like Java either, but at least I can write, compile, and test Java programs on my Linux box. I can't do that currently with .NET, and I doubt that .NET and Mono will ever be 100% compatible.
To be honest, I think that what Miguel is doing with Mono sounds very interesting, and what I have seen of C# has also been impressive. Miguel has a long track record of very successful projects, and so I won't be one bit surprised to see Miguel succeed here as well.
In the meantime, I am not the least bit interested in programming in Java again (I currently work with Python with a bit C mixed in), but I would be interested in Mono once it's soup. Especially if they ever get around to creating a C# implementation of Python (like Jython). I think that such a beast would be a compelling development environment even if it wasn't completely compatible with .NET. I certainly don't consider myself a Mono basher.
The problem is that the classes are almost certainly going to require VS.Net, and the beginning classes will probably end up consisting of little more than learning to use VS. Personally, I think that is a shame, but then it's not my school or my tuition.
Yes, I know about Mono, and I know about SharpDevelop. And if the University of Waterloo allowed you to hand in assignments that targetted Mono then I would not complain.
What are the chances of that happening?
For one thing a Java compiler and JVM exists for every platform that you can name. And you don't have to have a special development environment to hack Java. All you need is a Java compiler, a JVM and a text editor. I am not interested in paying money for Microsoft's tools, nor am I interested in booting into Windows to use them.
Good point. Now that I think about it I know plenty of developers that target Linux servers but write their software on Windows machines. Having the UNIX software on the development box is probably an advantage as well. I have been developing at a Linux box so long that I can't imagine using Windows, but plenty of people do.
Of course, I always figured that this was what Cygwin was for. Is Interix (or whatever Microsoft is calling it) so much better than Cygwin?
My guess is that this problem will get solved as more and more Chinese folks use Linux. As the Chinese government begins to actually roll out Linux they will undoubtedly notice that input could be slicker, and someone will fix it. Either it will be Sun trying to sell the Chinese StarOffice, or it will be hackers at Red Flag, or some Chinese college student trying to make a name for himself, but it is sure to happen. One thing is certain about Free software and that is the more use it gets the more polished it becomes. Right now Chinese input is not a big deal in Linux. Few Chinese people actually use Linux, and those that do have gotten used to the various hacks Chinese input requires. My guess is that it will become a big deal soon.
I am not Chinese, but it would seem the answer is fairly obvious. The primary reason that China is looking at Free Software is that Free Software is less expensive than the alternatives. This sort of thing didn't matter before the WTO started pressing China to stamp out software piracy, but it does now.
The second reason that Free Software is advantageous to the Chinese is that it allows them to bootstrap their own software economy instead being second or third class citizens in an American-led software economy. Their are plenty of bright folks in China who can write software. China would much rather put them to work than to pay software developers from overseas to do the work for them. The fact that Chinese developers are far less expensive than American ones doesn't hurt either.
The third reason has to do with Chinese national security. China has no idea what is in most U.S. written commercial software, but they do know that versions of Excell shipped with a flight simulator, and that before it was GPLed Interbase had a backdoor password for years. It's hard enough trusting commercial software on the very best of days, but trusting commercial software written by foreign nationals is a very sticky subject if you happen to be the Chinese government.
One thing is certain, China is not afraid of Microsoft. Microsoft and the BSA might seem scary to companies in the United States, but China is a sovereign nation (and a powerful one at that). If the BSA got too pushy the Chinese government could run their representatives over with tanks and there would be nothing that the BSA could do about it. China is cleaning up its act as regards to software piracy only because the U.S. has threatened to put sanctions on Chinese trade if they didn't. The U.S. market is important to the Chinese, and so they are trying to comply.
That's the question I would like to see answered as well. How many machines running Websphere and DB2 will it take to replace their existing PHP + PostgreSQL setup?
First of all Sourceforge is shifting to DB2, not Oracle, so you are still SOL. I certainly understand your point. If you have a large investment in Oracle, it makes sense to use Oracle software. I agree that sometimes Oracle on big iron is the correct solution to the problem. Sometimes you simply pay your money and get the best. You just assumed that I was slamming commercial software because this is /.
That being said, sourceforge has been remarkably stable, certainly stable enough for the development efforts of any enterprise I can think of. After all, if you can't get to your corporate sourceforge portal because its down you probably don't lose "millions of dollars." And even if you did lose millions of dollars everytime your sourceforge portal went offline there still isn't any evidence that the as yet to be written version that uses Websphere and DB2 is going to be more stable than the version that sourceforge is using to support thousands of developers on sourceforge.net. Websphere and DB2 are both good products, but the new sourceforge could be awful.
Both of us know that commercial software does not mean software that is completely bug free. In fact, I would rather trust something that comes with source and has a large user base over a brand new commercial product any day of the week. Time tested mature products like Oracle, however, are a different beast altogether. Clearly Oracle has a well-deserved reputation for stability.
Microsoft's Services for UNIX allows you to recompile your Linux programs for Windows. Of course, you get to pay extra for this privilege, and your UNIX services will feel "bolted on." Besides, Linux and Windows already play pretty well (no thanks to Microsoft).
What Microsoft is trying to do with UNIX services for Windows is give UNIX users a way to migrate to Windows, pure and simple.
I agree with you that this is a perfectly fair way to play, and I also agree that it isn't likely to work. After all, who is going to take a working Linux application and move it to a Windows box? That makes no sense at all.
No, they want to have the corporate version of sourceforge run on DB2 and WebSphere. My guess is that VA Software won't be migrating sourceforge.net. Like you said early sourceforge.net handles a ton of activity, and it does it with PostgreSQL and PHP. Migrating to a totally new technology would almost certainly cause problems, and the last thing that IBM wants is for VA Software to switch sourceforge.net from Free Software to their expensive commercial software and have the new setup be buggy or have performance problems.
The corporate accounts paying for the commercial version of sourceforge are undoubtedly dumber than a box of rocks. They would rather purchase a reimplementation of sourceforge.net on a different technology base than download the software that is good enough for sourceforge.net for free and learn to set it up themselves. Heck, they could even pay someone else to set it up for them (the folks at savannah.gnu.org have some experience in setting up sourceforge). In my opinion these are precisely the type of customers to target. It is almost trivially easy to make money from people who have more money than sense.
If you think that the man cutting your hair is gay then you are almost certainly not in a barbershop. There is nothing remotely feminine about a barbershop. Heck, the one I frequent has a "No Spitting" sign on the door, a baseball game on the old television. Also, my barber hasn't used scissors since 1947. Apparently scissors are for girls.
The reason that you would "make it visible" in this particular case is because you want to be able to pay the tolls on the bridge without stopping and handing over coins. In other words, your new visibility is a feature. You have placed a transponder on your car for the express purpose of being recognized by the authorities. You could leave the house with a bag over your head so people wouldn't recognize you, but being recognized is generally thought of as a good thing.
Unless of course, it means being caught doing something illegal or unethical.