How to gauge the risks of WWW publishing

My favorite WWW horror story involves a joint venture funded by two well-healed traditional media companies (full story). They launched a modest enterprise into cyberspace with a $180,000 year 1 budget. By the time the dust had cleared, $750,000 had been spent with no official launch and no revenue.

This enterprise, run by otherwise capable people, was the victim of the pace of technology. Their best plans were eclipsed by the march of technological change. One key factor was that this group had settled on a UNIX platform that was not only expensive to buy, but very expensive to maintain and modify.

For all its power, UNIX is nevertheless hideously difficult to use. For some leading-edge applications in the sciences, there probably is no other solution. By comparison, Macs are extremely flexible and easy to use, and there's a huge pool of Mac savvy workers (unlike UNIX).

But, for a relatively straightforward application like Web publishing, does it make sense to lock oneself into expensive administrators who basically babysit the machine?

The Apple Internet Solution is a software package that turns a PowerPC server into a potent Web server. Prices start at $2995 for the 6150, which makes a great WWW server.

Anyone who can use a word processor on a Mac can administer a Mac Web machine. The Internet connection may be difficult, but service providers will take care of that. (I speak from experience, having set up 2 Mac Web servers - it was easy!).

The best news is that the Mac limits the enterprise risk in branching into Web publishing. Publishers whose content is already digital, particularly desktop publishers, can probably add a Web server without adding technical staff (though it may be worthwhile to add a Web-oriented creative person if only as a contractor or consultant).

The Mac does not limit the upside: Mac Web server can be ganged to give performance levels (no to mention redundant backup) that will match any high-end platform.

The Mac Web server makes it possible for a publisher to assess maximum downside risk in real terms. And, since it's a Mac, if the enterpise doesn't perform as expected, the publisher has a machine that can be converted easily to other common publishing and business tasks.

Chris Gulker cg@gulker.com