<-- Back to Ranger Rick's Projects
This is a reprint of a (now gone) article at 32bitsonline talking about my reverse-FUDification of a PCWeek article talking about Linux's "weak value proposition". The original's pretty hard to read now, since the layout gets messed up in modern browsers.
By Jeff Alami
Many Linux users are familiar with the concept of FUD (Fear, Uncertainty, and Doubt). FUD involves disseminating exaggerated and sometimes downright wrong information about one's competitor. IBM did it ("No one got fired for buying IBM"), and Microsoft's doing it ("No one got fired for buying Microsoft"). We all know the claims: "Linux has no support," "Linux is extremely hard to install," and more recently, "Linux has fragmented and incompatible distributions."
On March 4, PC Week conducted an interview with Microsoft group product manager Ed Muth, in an article entitled "Microsoft exec dissects Linux's 'weak value proposition'". Not surprisingly, this article was filled of FUD, with only a few sprinkles of reality. One Slashdot reader, Ben Reed (using the name "Ranger Rick"), posted a modification of the article that is essentially the opposite of the PC Week article. I found it to be one of the best posts I've ever read on Slashdot. Here is the post:
Linux user dissects Microsoft's 'weak value proposition'
By Ranger Rick, Slashdot Reader
THE INTERNET, Worldwide -- Think of it as a burst of cold international rain on the Microsoft parade. In a far-reaching interview this week with himself, Slashdot reader Ranger Rick sounded off on the closed-source operating system, outlining what he considers are fundamental flaws with the Microsoft business model and the OS itself.
"I see it as more of a threat to Macintosh. Windows is a challenge, a competitor," Rick said. "The more I study Windows, the weaker I think the value proposition is to customers."
Technical misgivings
Rick delineated two main technical reasons why he believes Windows will not succeed with corporate customers.
First, a broad base of support for applications -- especially small, interoperable, easily-customizable, instantly-available applications, with modern internet distributions channels -- is necessary for an operating system to compete in today's market, he said.
"Five years ago, everything was shrink-wrapped, and the trend since then has been to customize standard shrink-wrapped software for individual business needs," Rick said. A "closed-design" ethos will not work in corporations, he said, where open standards ease interoperability and customizability.
The second failing, Rick believes, is an extreme level of integration between the OS and its applications. It's a point that touches on ground where holy wars are fought.
Indeed, some Microsoft advocates say Microsoft's integration of one large codebase with everything in one package is what makes it appealing. Rick disagrees.
"People want less integration," he said. "They want a choice between tools to use, with an open standard of interoperability. On the server side they want strong queuing and security. This is all done through a comprehensive set of tools that can be customized to their needs, which use open protocols for talking to each other. Microsoft has a high degree of integration, and therefore is more rigid and uncontrollable. Microsoft is basically a big step backward for those two reasons plus others."
Economies of scale
Rick next turned to the economics of Microsoft. He said his preliminary cost analysis showed Microsoft actually costs end users more than Linux.
"We have very little concern we can't compete with Microsoft on a TCO level," Rick said. "We think the total cost of ownership of Linux is lower than NT, but it's still hard to do good TCO studies because at the moment they're hard to compare since a large majority of Linux applications are free and have been developed, debugged, and improved upon for years, while NT supports so few Internet standards out of the box."
"Let's say, for discussion, they are equally scalable," he said. "And let's assume applications are available for both, and setup time is the same. Given all these factors, the best you could hope for is about the same cost per transaction between servers."
But Ranger Rick turned that argument on its ear.
"The problem with that is there are fewer applications available with the base NT install, there's a shaky development road map (with the Windows 2000 release date being pushed back again and again), and there's a higher technical risk in using it," he said. "You could cut NT some slack if it were sharply lower in cost per transaction than Linux, but that's not the case."
Acknowledging the phenomenon
Ranger Rick did acknowledge the myriad marketing forces that have propelled Windows NT into the spotlight, to the point that the OS was even featured in major print and online publications.
Ranger Rick attributed the closed-source hype to a number of factors, including a lack of fairness in media coverage of Windows NT.
"We're all in the business of wanting the customer to have the information needed to make informed choices," Rick said. "We haven't seen a flavor of NT coverage that addresses that. Some criticalness is needed.
For example, "some people say positive things about NT when their message is anti-Linux," he said. "But I wonder, in 36 months is this the next [Network Computer] or is it a viable OS? We don't see people question the NT numbers."
Ranger Rick pointed out that it's hard to track shipments of an open-source platform and its applications when they can be downloaded for free from any number of Web sites. "We feel that 2 to 20 percent of Windows NT shipments turned out to be 'shelfware,'" he said.
"From what we can tell, many servers come bundled with Windows and then have Linux installed instead. In fact," he continued, "many supposed Windows NT file and print servers are actually running Linux, and the users can't tell the difference!"
As for the recent vendor support at Windows conferences, Rick wasn't suprised, adding that the media needs to apply the same critical eye to this trend.
"Take IBM," he said. "They have long supported multiple OSes on X86. They are fundamentally in the service business. You would expect them to support Windows and Linux. But the deep investment is in services. It's a human investment -- developers, support -- the operating system is irrelevant. You have to separate out what OS they will install if you ask them and what investments they make."
But Ranger Rick's most passionate argument came on the development, where he said skill and enjoyment is the wild card that many observers have been ignoring.
"I find it hard to believe that some of the best computer scientists in the world will want to do their work in a stuffy corporate environment, missing deadlines, making code compromises, sitting in meetings," he said. "Without a long-term road map, programmers are free to make the best possible technical decisions, without worrying whether deadlines must be met, and worrying if the marketing department is happy. I have a hard time believing these visionary programmers and developers would get the same satisfaction just being another cog in the machine. I do not believe in that vision of the future."
But hey, it's competition
For all its shortcomings, Windows NT is part of a larger competitive landscape in the server realm, Rick said. "In the OS market, a fair person would see extraordinary barriers to competition," he said. "And the competition exists in terms of business model and channel model, and NT is a very interesting case."
For linux, that's key, since the question of whether it faces "real" competition in the OS market has been a contentious issue. Linux has done well in server market share, Muth said, but "we have lots to do in some parts of the market. Take the $100,000 to $1 million server range."
"There is extraordinary competition," he reiterated. "The market is a rich mosaic of parry and thrust from the vendors, with competing OEM service deals in the way of a free OS marketplace where choice rules. We have to earn our stripes every day. That's how it should be."
One of the comments regarding this post pondered if all FUD can be inverted, is a sort of "Inverted FUD Theorem." In many ways, similar exaggerations and mistruths that are used against Linux can be used against many other computer products, including Microsoft's. This article proves the fact that many of the undocumented comments in the PC Week interview, such the use of TCO and cost per transaction, could be reversed with barely any real modification. IBM used FUD, Microsoft used FUD--it's about time the Linux community shows how it's done.
<-- Back to Ranger Rick's Projects