First of all, I don't want to discuss the value of honorary degrees in general, but comment on whether or not Linus deserves a doctorate. I am holding a doctorate in computer science (which I got by writing a thesis and all that) and I am currently assistant professor at a computer science institute; so, in contrast to many of the earlier posters, I guess, I know what I am talking about.
What Linus did is, he conducted a large scale experiment in software engineering. He tried a bazaar style development on a scale that nobody before him did. As a result, we know more about this software development process today than we did before.
If you read some academic magazines, such as, Communications of the ACM, you will find that many respected researchers complain about the lack of experimentation in CS. Linus did it. Sure, he didn't really plan to do what he did and he didn't write his findings properly up; that's why he gets a honorary degree and not a "real" one. Or maybe ESR should also get a degree, because in CatB, he wrote much of what Linus discovered up (he even verified the experiment on a smaller scale).
And regarding the complaints that the degree comes from a mathematical faculty: it doesn't. The article says "mathematical and natural science faculty". In Europe, CS (including software engineering) is often more closely associated with mathematics than with the engineering disciplines.
Check out Novell's complaint about a similar study of Mindcraft concerning Novel NetWare 5 and NT. Maybe some of Novell's complaints about unprofessional methods also apply to the test Mindcraft did on Linux.
Do you mean, we should protect companies from open source competition? Why? Isn't competition improving the choice and value for the customers (= users of the software)? Furthermore, nobody prevents the company from evolving their business model, eg, going open source, too, and earning their money like other open source companies (eg, Cygnus) do. If a company doesn't manage this, it may be unfit in the free market sense and may die. Where is the problem?
I very much like the peer review character of the new moderator system, maybe because as a scientist I am very much used to rely on peer reviewed resources (conferences, journals etc). Looking at the scientific process of knowledge accumulation, I also find it hard to regard this system as censorship (in contrast to some of the other posters). It rather is a community-based information evaluation process (not filtering, as nothing is discarded). The method of selecting moderators - namely, on the basis of their previous "work" - is also akin to peer review in science.
However, I think, the system might still be improved by taking up some of the ideas of the well-tested peer review process used in science. So, here some suggestions (some of them already appeared in in other comments):
Moderators should attach a reason, at least when lowering scores. This has the benefit of giving the poster some feedback and forces the moderator to think about his/her action; thus, making moderation more objective.
A fixed range of possible scores (maybe -20 to 20), as open scores don't help much and can be abused.
Instead of allowing moderators to add and subtract points, a weighted absolute score combined with a upper limit on the number of moderators per article might be better.
Ok, here my humble attempt at a translation of the Japanese ZDNet article...
--- Tokyo, 2 March, at the Playstation Meeting 1999 organized by Sony Computer Entertainment (SCEI), the company provides first details of the much-spoken about next-generation Playstation.
Originally, the main news of the meeting should be the worldwide distribution of over 50 Million Playstations. However, as on the 16 February Toshiba presented in San Francisco a high-powered MPU for computer entertainment, the question arose as to "whether we should announce at this meeting the next-generation Playstation?"
In the announcement at 3:30pm, it was stated that the development OS for the next-generation Playstation is based on Linux; though, details of the hardware are still unclear.
The CPU was increased from the rumored 250MHz to 300MHz. Furthermore, according to SCE, sales might start in '99, which sounds as if it is difficult to meet. We try to get hold of more details about this machine. Stay tuned!
Slashdot Mirror
What Linus did is, he conducted a large scale experiment in software engineering. He tried a bazaar style development on a scale that nobody before him did. As a result, we know more about this software development process today than we did before.
If you read some academic magazines, such as, Communications of the ACM, you will find that many respected researchers complain about the lack of experimentation in CS. Linus did it. Sure, he didn't really plan to do what he did and he didn't write his findings properly up; that's why he gets a honorary degree and not a "real" one. Or maybe ESR should also get a degree, because in CatB, he wrote much of what Linus discovered up (he even verified the experiment on a smaller scale).
And regarding the complaints that the degree comes from a mathematical faculty: it doesn't. The article says "mathematical and natural science faculty". In Europe, CS (including software engineering) is often more closely associated with mathematics than with the engineering disciplines.
So, keep cool.
Chilli
Chilli
Chilli
However, I think, the system might still be improved by taking up some of the ideas of the well-tested peer review process used in science. So, here some suggestions (some of them already appeared in in other comments):
- Moderators should attach a reason, at least when lowering scores. This has the benefit of giving the poster some feedback and forces the moderator to think about his/her action; thus, making moderation more objective.
- A fixed range of possible scores (maybe -20 to 20), as open scores don't help much and can be abused.
- Instead of allowing moderators to add and subtract points, a weighted absolute score combined with a upper limit on the number of moderators per article might be better.
*Chilli bows gracefully*Ok, here my humble attempt at a translation of the Japanese ZDNet article...
---
Tokyo, 2 March, at the Playstation Meeting 1999 organized by Sony
Computer Entertainment (SCEI), the company provides first details of
the much-spoken about next-generation Playstation.
Originally, the main news of the meeting should be the worldwide
distribution of over 50 Million Playstations. However, as on the 16
February Toshiba presented in San Francisco a high-powered MPU for
computer entertainment, the question arose as to "whether we should
announce at this meeting the next-generation Playstation?"
In the announcement at 3:30pm, it was stated that the development
OS for the next-generation Playstation is based on Linux; though,
details of the hardware are still unclear.
The CPU was increased from the rumored 250MHz to 300MHz.
Furthermore, according to SCE, sales might start in '99, which sounds as if it is difficult to meet. We try to get hold of more details about this
machine. Stay tuned!