In Mel Brooks’ classic movie Young Frankenstein, whenever someone speaks Frau Blucher’s name, horses whinny.

Mentioning Linux in this column has a similar effect.

A month or so ago I suggested it’s time for CIOs to study the feasibility of migrating their desktops to Linux, not for technical reasons but as a logical response to where Microsoft seems to be taking its licensing. With a few exceptions, from people who thought they were watching Invasion of the Body Snatchers and thought I’d been replaced by a pod Bob, most of my correspondents welcomed me to the path of truth and righteousness.

Alas, it was not to last, because the very next week I questioned whether the open-source licensing model can move beyond “me-too” products to create significant innovation.

I immediately fell from grace.

I don’t understand why, though. Some drug companies innovate, developing new and innovative pharmaceuticals. Others manufacture inexpensive copies — generic drugs — and serve a valuable purpose by doing so. Thus far, the role of open source, in the desktop marketplace at least, seems parallel to that of the generic drug companies.

Most proponents of open-source software either described whatever happens to have been developed with an open-source license as innovative, which is true in a black-is-the-same-as-white-only-darker kind of way, or they said that this will change over time, in particular because there are more open-source developers than any single software company can ever employ.

But this argument doesn’t hold together very well. Commercial software companies organize their developers into teams that focus on developing specific products. In the open-source community, developers generally contribute as individuals.

Other correspondents pointed out, correctly, that there are a significant number of innovative open-source products. Interestingly, all are development or administrative tools; none they mentioned are end-user-oriented applications.

Which makes sense to me: A single developer can successfully create a language or utility, which other individual developers can extend incrementally. A team of developers, on the other hand, can create coherent products with a consistent user interface and architecture — something that’s awfully hard for the open-source community’s evolutionary model to manage.

Most worrisome to me were the letters describing the importance of the open source movement — that dealt with the open-source vs proprietary licensing question as one of good versus evil. Opinion: If you think in these terms, you need to peel the onion a bit. And if you think in these terms I have one other thing to say to you:

Blucher!

The comedian Steven Wright reported dreaming that everything he owned was stolen and replaced with an exact duplicate.

If you replace Windows with Linux as your organization’s preferred desktop operating system (last week’s column recommended that you carefully assess this option) your end-users just might feel like they’re trapped in Steven Wright’s dream, except the duplicates won’t be exact, and perhaps not quite as good, either.

The conundrum of Linux is that for the most part it’s a me-too operating system. Innovation happens elsewhere – on Windows and the Macintosh for desktop applications and OS enhancements; on Netware, Windows NT/2000, or a commercial Unix for servers – and someone says, “Hey, I can do that on Linux!” Then, through a quasi-Darwinian process, the open-source community quickly refines and debugs those applications and platform extensions it finds useful.

But can the open-source license model support true innovation, or is mimicry the best it can do? We don’t really know.

So long as Linux fails to gain dominant marketshare, this question is unimportant. But imagine for a moment that it succeeds. What then?

One school of thought says the innovative energy now directed at the Windows platform would switch to Linux, following the marketplace wherever it leads. This is a reasonable expectation for a free-market economy.

The distance from reasonable expectation to reality, however, can be pretty long, and another school of thought holds that in a mature marketplace like the personal computer, the open-source model doesn’t promise enough profit to support the risk of investing in truly innovative concepts. Of course, just because Linux is an open-source OS doesn’t mean applications that run on Linux have to be open source as well, but once you start buying non-open-source software for Linux you’re back where you started: MS Office XP for Linux, if it existed, would have licensing terms just as onerous and risky as does MS Office XP for Windows (and presumably for Macintosh).

When Adam Smith first wrote about market-based economics we were a nation of merchants and farmers. Affluence was the most anyone aspired to – the establishment of a wealth-based aristocracy was a century in the future.

The open-source movement is, in a sense, a return to this merchant-based economy (not a communistic plot, as its extreme detractors claim). The open (source) question is whether, in the 21st century, affluence is enough anymore.