Letters from our readers.
April 2 1999 3:30 AM

Address your e-mail to the editors to letters@slate.com. You must include your address and daytime phone number (for confirmation only).

Apples and Oranges


In "A Labor of Linux," Eliza Truitt discusses installing Linux on a PC with Microsoft Windows already installed on it. She doesn't compare the reverse; that is, installing Windows from scratch on a machine with Linux pre-installed on it. It is no piece of cake, if not downright impossible, for the average user to perform this sort of install. Installing Linux on a PC with Windows already loaded is far easier.

In fact, people like Truitt do not generally install their own operating systems of any type and make heavy use of help desks on a day-to-day basis to keep their systems running. She talks about configuring printers, connecting to networks, and so forth on Linux--I wonder if she has ever done the same on a Windows machine. Very few computer users install their own operating systems these days; they buy their computers with the system pre-loaded. There are plenty of vendors who sell Intel boxes with Linux pre-installed, including Dell, VA Research, and so on.

Truitt writes, "compare that with Windows 98: $199 for a full setup, $89 for an upgrade, or bundled for 'free' as part of nearly every non-Macintosh computer." I think that it is pretty nonsensical to make the statement that Windows is "free" under any circumstances. Nor does Truitt note that when you get Linux from any of a variety of sources you get far more than just the OS that Windows 98 users are accustomed to. You get a full development system with an unlimited server-based OS. The equivalent software would cost upward of $15,000 if you were to try to duplicate it from Microsoft--an unlimited client version of Windows NT costs $5,000. The databases, programming tools and applications included in Linux are worth many thousands more--if you can find them at all for Windows NT.

If you expect to attract readers to Slate, you really need to do a better job in presenting balanced, factual reporting.


--Eric Larson

Free for All

I just read Andrew Shuman's "Geek Love," and I feel I have to write to make a few points. The very first paragraph contains a stunning number of errors. Linus Torvalds is the original author of the Linux kernel, but he is by no means the author of Linux. There are literally thousands of people, both programmers and non-programmers, who have made contributions to the Linux distributions. Linux is not derived from an operating system called GNU. GNU is not an operating system but a software suite that runs on many operating systems (mostly Unix-based OS's, but Windows is also supported by many of the packages). The proper name for Linux is Linux, not "GNU/Linux." And just for your information, there is a "GNU operating system" in development, it's called the HURD and it is different in a number of respects from Unix.

Shuman makes the claim that just because software is freely distributable, the authors have no means of recompense for their work. That is simply untrue--there are many ways that software can remain free but still allow the authors to get paid (just look at RedHat or Netscape).


Then Shuman claims that Linux provides no graphical user interface. The truth is that you have a choice--if you want a text interface you can have it, if you prefer a graphical interface that's fine too. Shuman also makes the totally unsubstantiated (and untrue) claims that there is very little in the way of application software for Linux and that very few people use it: The latest estimates of the number of Linux users run into the millions--how can that be "so few"?

Then Shuman claims that free software is less tested than commercial software. I use both commercial software and free software in the course of my work (I'm adminstrator of a large number of NT and Unix computer systems), and it's pretty clear to me that it's the free software that is rigorously tested and the commercial software that gets released just as soon as it appears to run. Why? Because the incentive to an author of free software is to make her package the best, so releasing inadequately tested software will do the author's personal reputation no good at all. The incentive to a commercial author is to get the application on the shelf so she can derive income from it; if Version 1 happens to have bugs in it, the bug fixes become a selling point for Version 2.

I'm not anti-commercial software or even anti-Microsoft. I am in favor of choice and freedom though, and your article doesn't exactly encourage readers to exercise their choice and freedom.

--Steve Bennett


Lancaster, England

Zero Zero Sum

It doesn't make a big difference to me--or, I suspect, to most Linux users--whether Truitt, Shuman, or any other Windows users like Linux or not (see "SlateDoes Linux"). Most Linux contributors don't primarily care about competing with Microsoft, or decreasing Bill Gates's wealth, or giving away something for free. Linux contributors simply are creating systems they themselves enjoy using and find productive.

The main reason Linux users are increasingly concerned about Microsoft is because Microsoft increasingly attempts to set proprietary standards for how data are exchanged. That is, technical and legal documents are often found in Microsoft Word formats, access to online and banking services often requires the use of Microsoft-only software, multimedia data often come in Microsoft-only formats, and hardware often only ships with Microsoft-only drivers. This is not because Microsoft-based systems are necessarily better, but because Microsoft is the biggest player, and that's what companies are going to focus their efforts on when writing drivers and software; there are, after all, only limited resources for writing drivers and front-end software.


So, the concerns Linux users have about Windows are related to whether network effects involving proprietary content formats are going to exclude them from some areas of public and commercial life. Just like Truitt would not enjoy being forced to struggle with Linux, most Linux users don't enjoy being forced to do their work on Windows using its "WIMP" interface.

Microsoft needs to become more sensitive to concerns about interoperability and open standards. Recent efforts by Microsoft in the areas of XML and other Web standards are encouraging. Clear support for Sun Java would also be good (and would, in the long run, benefit Microsoft). I hope that, as Microsoft matures, so will its attitude toward standards and cooperation. The computer market is big enough that it does not need to be winner-take-all.

--Thomas M. Breuel

San Jose, Calif.

Simple Elegance

Eliza Truitt's article, "A Labor of Linux," on her experiences with her Linux installation provides wonderful inspiration for the intimidated. But a word to Truitt: The term "elegant" in science and technology has a special meaning apart from common use. Elegance is not synonymous with "simple." Rather, it refers to the most terse solution to a problem. Such a solution will tend to be cryptic to the uninitiated. Therefore, "elegant" is most often used by those after the fact of having used a procedure successfully. For the others, I would venture to say, their word for it would be arcane.

--Roger Imai

Nashville, Tenn.

Emperor Has No Clothes

Please note in your fascinating "Explainer" on hara-kiri that there is one point on which I believe you are mistaken. In feudal Japanese society, the person who would order suicides in place of executions was the shogun, or military leader of Japan. Emperors, excepting the period from the late 19th century to the end of World War II, have yielded little if any power in Japanese society.

One fascinating example of ritual suicide in 18th century Japan that still regularly appears in movies and television in Japan is the story of the 47 samurai. According to this actual historical account, an adviser to a Tokugawa period shogun antagonized a rival to the point where this rival drew his sword. This all occurred in front of the shogun, which meant serious trouble, because displaying a weapon in the shogun's presence was punishable by death. Despite the fact that he was forced into this situation, the offending adviser was sentenced to death and committed suicide as you described in your magazine. After this, 47 of the condemned official's samurai decided to take revenge. They coordinated an attack on the enemy adviser and murdered him. Honor called on them to avenge their master's death, but the laws forbade murder. After regaining their master's honor, all 47 samurai went to their master's grave and also committed hara-kiri to take responsibility for breaking the law. The site of these samurai's graves is still a well-visited location in Tokyo.

--Mike Cojerian

Princeton, N.J.