Slashdot Mirror
WSJ Interview with Linus
Thanks to several of our readers who noted the public version of the Wall Street Journal's interview with Linus, conducted by Lee Gomes. Some of the standard issues - 2.4, Linux on the desktop are talked about but the statements concerning ye olde Mindcraft tests are particularly interesting.
There are a couple of tidbits from kernel traffic which may be helpful. One is a discussion on Joe Pranevich's DRAFT of his The Wonderful World Of Linux 2.4. The other is the DISCUSSION of the DRAFT of a PROPOSED press release which highlights the big features for when 2.4 comes out.
If you're into finding out things before they are final and are an early adopter, these may be of use to you. They are certainly not finished documents and should not be treated as such. They may contain misleading statements, misunderstandable statements, misunderstood points, mention of features that don't make it, and/or outright lies with the intention of deceit.
It is fundamentally untrue that "Linux is not ready for the desktop."
I use Linux as my primary desktop OS.
I am not a programmer.
I understand that it is not appropriate for most (l)users, but that does not make it "not ready."
My grandmother does not drive a Porsche, it would be too much for her, and wouldn't carry groceries well, but that doesn't make it "not ready for the highway."
In the same way, Linux does not make a good platform for Word and AOL, but that does not make it not ready.
This whole "not ready for the desktop" idea revolves around the assumption that there is some ideal desktop for everyone, and that Windows (or MacOS) is closer to this that "Linux" (really, KDE or GNOME or Windowmaker, or something) and that "Linux" has further to go towards this ideal.
I disagree. For me, customizability is king, and "Linux" takes the prize in this field. (note: BeOS might, but HW support is too weak -for me-)
Windows is not ready for MY desktop.
-Peter
I think that last statement is untrue. About a month ago, many of the audio engineers you spoke of released an "open letter" to the linux-kernel mailing list. They requested that the average and max latencies be reduced (for exactly the same reasons you mention). They suggested a set of patches created by Ingo Molnar, but not integrated into the main tree. This started a flame war on linux-kernel over the validity of the patches, but the views of Linus were well-stated: he believed that reducing latency was a "serious issue" and considered it very important.
As happens with many such requests/patches, the implementation details were shouted over at length, but most of the kernel hackers seemed to be in agreement that reducing latency was a big concern.
Kernel traffic (what a great resource!) has the thread here.
Have you been following the news on linux-audio-dev? For the past year or so, there has been a kernel patch (by Ingo Molnar) that guarantees latency times below 4 msec. This is on par with BeOS, previously the market leader. (Windows, for reference, gets about 20 msec; I'm not sure about the old Mac OS, but I know it's not that great.) Unfortunately, this patch was kind of kludgy (by Ingo's own admission) and Linus wouldn't include it. But in the past few weeks, Andrew Morton has written a clean and successful low-latency patch that is likely to be accepted. During testing, it has been found to be stable and get sub-2 msec latencies. We're already beating BeOS, and we've been beating Mac OS and Windows for quite some time now.
As for cron, it's a DDT (Don't Do That!). There's no need to run cron on a DAW, so disable it.
On the topic of MOSX, I think it should do quite well. It's heavily based on NeXTSTEP, which was quite popular in serious computer music fields (CCRMA and so forth). The MusicKit is also being actively updated for MOSX (and somewhat less actively ported to Linux/GNUstep). I have to say that I'm increasingly excited about the developments in computer music lately--it's an exciting time to be doing this.
A good interview. I'm consistantly impressed with the candor and openness that Linus exhibits when he talks about a project that is, in a big way, his baby. He has no trouble admitting the shortcomings of Linux, nor any trouble touching on its strengths. His comments prove that it is possible to be a very strong advocate without resorting to extremism or flame warring. How many people who aren't nearly as involved as Linus can wade into the OS debate while maintaining some perspective? Linus Torvalds will convince more people of the virtues of Linux with his balanced take on the issue than any given hundred /. "M$ is the evil empire, windoze sucks, Linux roolz" posters.
"Sweet creeping zombie Jesus!"