This is not going to work because the real problem is about time. It takes significantly longer to be nice and show patience than it does to be mean and to the point. Someone with Linus' responsibilities just isn't ever going to feel like they have the time to be nice to everybody. If you want to improve peoples' attitudes in something as huge as the kernel community, than you need to change the heirachical structure to something with a lot more depth. Linus shouldn't be dealing with 2/3 of the people he is currently dealing with. If he isn't directly interacting with them, then they won't be getting offended by him either. Those people should be interacting solely with developers with less responsibility and therefore more time to be nice.
I work for a medium size engineering firm, Smiths Industries Aerospace in Grand Rapids, MI.
We develop Flight Management Systems for various types of aircraft (if it's bigger than a Cessna, odds are the pilot isn't really flying the plane, after all). We've been a VAX/VMS shop for a long time, and we still do a fair amount of devel on that platform, but with the decline in PC prices and the increase in PC power, we've been searching for a new approach.
Of course, the powers that be decided the future was in a Microsoft solution, but a number of us lowly engineer types were too revolted not to try and find something else. Solaris was the first choice, but since we write the majority of our code in Ada (no comments from the peanut gallery), we were limited in our choice of compilers between one by Rational and one by VADS. Well, neither of these produce very good code on the PowerPC considering how expensive they are.
Too hasten to the point, someone had the bright idea of trying a GCC cross-compiler with NYU's Ada front end, GNAT. You wouldn't believe how much better code it produces (30% - 50% better) on our benchmarks. Now we're looking at a linux based development environment to go with it.
I just goes to show how sneaky you have to be to get management to even try open source stuff sometimes. First you let them spend a pile of money on a poor solution, then you step in and replace it with a vastly better solution for -- what was that price again? -- oh, yes, for FREE. The suits are usually in such hot water at the time, they can't do anything about it anyway.
All right so I said more than I had too, but it's a kind of interesting story, isn't it?
Slashdot Mirror
This is not going to work because the real problem is about time. It takes significantly longer to be nice and show patience than it does to be mean and to the point. Someone with Linus' responsibilities just isn't ever going to feel like they have the time to be nice to everybody. If you want to improve peoples' attitudes in something as huge as the kernel community, than you need to change the heirachical structure to something with a lot more depth. Linus shouldn't be dealing with 2/3 of the people he is currently dealing with. If he isn't directly interacting with them, then they won't be getting offended by him either. Those people should be interacting solely with developers with less responsibility and therefore more time to be nice.
I work for a medium size engineering firm, Smiths Industries Aerospace in Grand Rapids, MI.
We develop Flight Management Systems for various types of aircraft (if it's bigger than a Cessna, odds are the pilot isn't really flying the plane, after all). We've been a VAX/VMS shop for a long time, and we still do a fair amount of devel on that platform, but with the decline in PC prices and the increase in PC power, we've been searching for a new approach.
Of course, the powers that be decided the future was in a Microsoft solution, but a number of us lowly engineer types were too revolted not to try and find something else. Solaris was the first choice, but since we write the majority of our code in Ada (no comments from the peanut gallery), we were limited in our choice of compilers between one by Rational and one by VADS. Well, neither of these produce very good code on the PowerPC considering how expensive they are.
Too hasten to the point, someone had the bright idea of trying a GCC cross-compiler with NYU's Ada front end, GNAT. You wouldn't believe how much better code it produces (30% - 50% better) on our benchmarks. Now we're looking at a linux based development environment to go with it.
I just goes to show how sneaky you have to be to get management to even try open source stuff sometimes. First you let them spend a pile of money on a poor solution, then you step in and replace it with a vastly better solution for -- what was that price again? -- oh, yes, for FREE. The suits are usually in such hot water at the time, they can't do anything about it anyway.
All right so I said more than I had too, but it's a kind of interesting story, isn't it?
wsh
Looks like a good reason not to buy Toshibas.