By Phil Nadeau.
This letter was sent to Scott Berinato, the author of the ZDNet
Ed Muth Interview. I wouldn’t consider this a polished or finished
rebuttal, but it makes some points that thus far have not been made
in the favor of Linux and OSS. Share and enjoy.A point-by-point
rebuttal would take me all night, so I’ll just touch on the main
points.
I do not understand why Linux should have a higher TCO (Total
Cost of Ownership) than NT, when a lot of software that is ‘value
added’ under NT (and even other Unices such as SGI or Sun) is
included in almost every Linux distribution. Therefore, I do not
understand why Mr. Muth’s perceived lack of commerical software
support for Linux would affect the cost of ownership for a Linux
based server. The most expensive component of any software
installation is the human operator, and these are required just as
much with NT as they are with any other OS. This is the real issue
for determining TCO.
While I do not have hard numbers available as to which requires
more maintainence, Linux or NT, seven years of sysadmin experience
would make me bet on Linux. NT works as long as you don’t push it
outside the narrow design envolope that MS set for it. Linux
continually exceeds all design expectations, and several or my
clients are strictly Linux/FreeBSD shops – because one admin can
manage sixty mission-critical servers without breaking a sweat.
Let’s see NT host a quarter million web sites without a small army
rebooting each host every fifteen minutes.
Regarding the (lack of) a Linux ‘Long-Term Development Roadmap’:
I know of no such roadmap for Microsoft, either. At least, I know
of no roadmap they wish to share with the public. This is because
Microsoft is not an innovator or leader, merely a fast follower.
Their roadmap is determined by whatever everyone else is doing to
make money… then MS tries to leverage its market clout to get
everyone to buy their cheap knockoff.
If we grant Mr. Muth the point that Linux lacks integration,
then that can only further our argument because anyone – you, me,
the CEO of Boeing, or a potato farmer in Mexico – can crack open
the system and tailor it to his needs. Nobody has to worry that
modifying the web browser will break the word processor (a
misfeature of Win98 that Microsoft is using in their defense in the
DoJ trial).
Nobody but Microsoft thinks that such tight integration is a
good idea, because it isn’t. It’s bad engineering. Good engineering
– true integration – divides a software system into layers, each of
which has interfaces for above and below, heavily over-engineered
for expansion. That’s how good software is made. Microsoft uses
‘integration’ as their new excuse to cut corners and muscle out the
competition.
Further, I do not see how the alleged lack of a roadmap could
lead to technical risk in using an OS. People use computers to
accomplish specific tasks. No company can anticipate perfectly the
needs of the entirety of their potential market. By using the
distributed open-source model, Linux is actually in a better
position to meet customer needs, because it is the customers
themselves who drive the development. The open, loose nature of the
platform offers the user a greater ability to tailor his system to
his needs, outside of the limited design envelope of the original
system.
Mr. Muth’s comments on the quality of the personnel in the Linux
community is almost insulting:
“I find it hard to believe that some of the best computer
scientists in the world will want to do their work for free,” he
said. “Without a long-term technical road map, without
multimillion-dollar test labs, someone wants me to believe these
visionary programmers and developers will want to do the best work
of their lives and then give it away. I do not believe in that
vision of the future.”
Mr. Muth is apparently not familiar with how science (real
science) works. A scientist’s goal is to obtain scientific credit,
and it is often said in academic circles that one must ‘publish or
perish’. Publishing results – good results – is how one obtains
scientific credibility, something more valuable to scientists than
money.
In short, we AREN’T giving our work away for nothing. We get
paid in credibility. Linux developers have no respect for
Microsoft, because they haven’t garnered any credibility, only a
hell of a lot of money.
Linux developers are also ‘paid’ by the very existance of Linux.
Because of our efforts, we have a stable, cost-effective, flexible
operating system available that we can use to get things done and
make a living. I could not make a living using Microsoft as a
platform, because I cannot afford their exorbitant costs for
(frequently inadequate) solutions, and I cannot trust them to put
my clients’ interests ahead of the Microsoft corporate agenda. I
can trust Linux.
Mr. Muth’s argument appears to be ‘We’re Microsoft, We’re Big,
We’re Bad, We’ve got Market Share. Buy our product because everyone
supports us. Don’t buy linux because you don’t know what you are
getting’.
It is true that nailing down exact numbers on Linux
installations worldwide is impossible. However, using that to build
a case against any software is a flawed argument, based on playing
to your audience’s fears.
Where I come from, we call this FUD, and it stinks.
–Phil Nadeau
Freelance Software Wizard
[Feel free to quote me, as long as it is in appropriate context,
or to distribute this memo as you see fit. I welcome your
response.]