Speaking of the benefits of Open Source for high-risk missions, perhaps that Mars probe would have fared better if only management had chosen to use Perl with the freely available Math::Units module!
I have no reason to doubt the numbers. But are they particularly exciting, when you consider the sheer size of the installed base of Windows compared to proprietary Unices? I don't know the numbers, but I suspect the rate of Un*x defectors may actually be higher than the rate of Windows defectors. And that, in the long run, is the focus of Redmond.
I will never forget how it felt to install Mr. Zubkoff's BusLogic drivers in a 2.0.8 kernel for the first time. Back then, the drivers hadn't yet made their way into Linus' tree. As a veteran of rolling my own kernel, having built X and gotten it up when that was still an accomplishment, and having bled on libc #defines, I settled in for major pain. But Mr. Zubkoff's driver dropped right in. Like butter. The nost seamless thing I'd ever seen. He will be missed, not only for great drivers but also for providing a model of how the Linux community could approach initially reluctant vendors for register-level APIs. Here's to you, sir!
Slashdot Mirror
So I assume there will a fitting observance in about a week's time? Landmarks like this must be celebrated!
Happy birthday, Smiley!
\end{ mode_:-) }
Speaking of the benefits of Open Source for high-risk missions, perhaps that Mars probe would have fared better if only management had chosen to use Perl with the freely available Math::Units module!
I have no reason to doubt the numbers. But are they particularly exciting, when you consider the sheer size of the installed base of Windows compared to proprietary Unices? I don't know the numbers, but I suspect the rate of Un*x defectors may actually be higher than the rate of Windows defectors. And that, in the long run, is the focus of Redmond.
I will never forget how it felt to install Mr. Zubkoff's BusLogic drivers in a 2.0.8 kernel for the first time. Back then, the drivers hadn't yet made their way into Linus' tree. As a veteran of rolling my own kernel, having built X and gotten it up when that was still an accomplishment, and having bled on libc #defines, I settled in for major pain. But Mr. Zubkoff's driver dropped right in. Like butter. The nost seamless thing I'd ever seen. He will be missed, not only for great drivers but also for providing a model of how the Linux community could approach initially reluctant vendors for register-level APIs. Here's to you, sir!