Slashdot Mirror
Tanenbaum Rebuts Ken Brown
Stephan Schulz writes "Andrew Tanenbaum has rebutted Ken Brown's reply to his original comments on the (in)famous AdTI report on Linux's origin. It's quite entertaining, and leaves little doubt (well, even less than before) that Brown is conciously twisting the truth. Choice excerpt: 'I'm pretty animated all the time. But I only get tense when people try to put words in my mouth. After half an hour of repeatedly answering the question "Could Linus have written the Linux kernel by himself?" in the affirmative, I was getting a bit irritated. ... People who know me would probably confirm that I do not suffer fools gladly.' I'd add that being called 'the good Professor' repeatedly would have me exploding in no time..."
Critique of Ken Brown's response
For an even funnier laugh, I recommend reading this one Is Brown Really the Father of Samizdat? - A Parody by Justin Moore to counter the Fake Research, hmm did I mention about their Fake Research?
Not true - as mentioned elswhere here and on Groklaw, it's being published by a "vanity" press, which means no huge advances, royalties, or anything. He'll be lucky to break even on the publishing costs.
C|N>K
Every year at the University of Waterloo the Computer Engineering and Computer Science students personally build their own operating systems (including documentation) in less than four months. This is done without any prior knowledge of how OSes work and without being taught C.
I'm sure many universities and colleges around the world do the same. Perhaps Ken Brown should investigate them as well.
http://www.ece.uwaterloo.ca/~ece354/
http://www.student.cs.uwaterloo.ca/~cs452/
Many of the recent Slashdot comments regarding the ADTI President Ken Brown's defense of his controversial tome noted that his principle audience was not the Linux community, or even the IT industry. His target audience is the policy-makers in Washington D.C. How is that group informed about issues surrounding open source in general and the Linux kernel specifically? One 'trade' publication, FCW Media Group, "produces information resources that help government IT buyers... form an integrated information system to help them purchase, build and manage technology in government." They are 'our' target audience in defending the concept of software libre, in advancing open protocols and other standards, and in correcting FUD. The May 3rd online issue provides one such opportunity to advance Linux in government research.
Nothing stops the flow of FUD like well-positioned information.
"Rocky Rococo, at your cervix!"
The fishs translation (which is pretty hillarious in itself) can be found here.
ich bin der musikant
mit taschenrechner in der hand
kraftwerk
We can conspire about why he's so driven to his (repeatedly refuted) belief that Linus couldn't have written Linux without ripping someone else's code off all day, but the fact remains that KB's own consultants have contradicted him!
I used to be a TA for CS452 Real-Time Programming course at the University of Waterloo. The assignments for that course came in two parts: (1) design and implement your own real-time multitasking kernel, and (2) use it to design and implement a real-time control system for either a robot arm or a model train.
The students had about a month and a half to complete the first part, which was broken into four assignments. The kernels had a microkernel architecture, but I don't think that really alters the development time that much. (The message passing was highly synchronous, which helps to limit the mind-boggling complexity of debugging a distributed program, which Dr. Tanenbaum doesn't seem to discuss.)
They worked in teams of two, but when I took the course, my lab partner conked out on me, so I ended writing the kernel myself, but that was okay since I had written multitasking kernels twice before, one in MACHINE LANGUAGE (no, not that wimpy symbolic-assembler stuff!) for a Commodore-128.
So, it's quite do-able for a motivated student to write a relatively simple kernel in the amount of time that Linus took. Just ask the CS452 students--they had to build their kernels in just six weeks, plus they had other courses and limited resources in the lab.