There's two major classes of open source myth: the "Open source is evil and strange" myth, and "Open source makes everything perfect forever" myth. It's easy to see how notions so far off-center are far from being universally true, but I'm heartened whenever someone debunks myths on both sides by taking a more moderate stance. Such is the case, intriguingly enough, with a paper entitled "Ten Myths About Running Open Source Software In Your Business," courtesy of ActiveState.

I've learned to be automatically wary of papers like this, if only because their publishers usually have a vested interest in the specific angle being taken. ActiveState's no exception, given that it sells and distributes interpreters for a number of languages on top of which open source solutions are built -- namely, Perl, Python, and Tcl. But they did the smart thing and approached the subject from as close to the center as possible, and the only sales pitches in the paper itself are fairly diffuse and easily separated from the bulk of their claims.

An example: Myth 6 -- Open Source Software Equals Open Standards. "Open source software is simply a licensing model; it does not equal best practices, like incorporating open standards." I'd argue that it's a development model more than it is a licensing model -- but I do agree that openness of code is no guarantee of anything else. It doesn't guarantee interoperability or security, although it sure makes them easier to implement. Another example, from the other side: Myth 5: Licensing Is Always A Nightmare (typically only if you're reusing code in a publicly deployed project), or Myth 1: You Have To Choose Between Open Source And Proprietary Software (the two can interoperate nicely).

What I wonder is, how many of the myths in question are strawmen by now? It would've been interesting to poll a bunch of CIOs and see how much they agree or disagree with said myths -- and maybe in the process find out what myths about open source, positive or negative, are circulating in the wild that don't make it to the pages of whitepapers like these. If they did glean their list of 10 through such a methodology, it isn't clear.

Still, it's useful to have documents like this on hand as a starting point from which to build a good defense of open source in an organization -- should you need to make one.