| Double Toil and Trouble.
Three weeks of dedicated research.
Thirty hours of long-distance phone calls.
It was the first time I had been involved in designing and building a web server and it was turning into a long haul. I had spent the past three years surrounded by Linux users and had come to understand the power of Unix-based web servers. And while I was determined to serve websites from a Macintosh, I also knew a solution that met my needs as a developer and my clients' requirements for power, stability, and speed was a must-have.
I put together a team to assist me. My Perl programmer, a convert from Windows 95/NT to Linux, never held back his opinions and stated with semi-religious fervor.
"Fine for graphics and video editing," he said of the MacOS, "but I wouldn't trust it for a second with mission critical services."
While I argued with him, I could see his point: the MacOS beats Windows 95 hands down for everything I do, from hacking Perl scripts in BB Edit to video editing to running my business in FileMaker Pro. But one crash and I could already hear my web clients requesting their removal from my system and possibly from my design services.
The second person on my team was a quiet proponent of Macintosh, as well as a Linux developer. He researched MKLinux and LinuxPPC but had never tested his research. He believed a web server should be based on an architecture that allows rapid growth, maintenance from remote locations, and the support of more than one company.
And I was looking for a system that felt close to home—something I could understand without a huge learning curve (I couldn't afford to hire a full-time system administrator to maintain a Unix-based web server). Even the concept of a command-line interface (instead of a mouse) was, at the time, threatening and quite alien. We needed, it seemed, something that could serve many masters.
|
