Why avoid a monoculture of code in schools?

Anyone interested in free and open source software in education should read this. Dr. Couros has set out and tied together a rich body of research and observations. I'm learning a lot as I go through his work.

On page 132 of Couros' dissertation something struck me as odd as a participant argues for "compudiversity" over "monoculture" in schools. Generally, I agree. However, diversity/monoculture is a quality of code, not copyright and patent licensing. If an organization feels a need for diversity, then that can be accomplished by a variety of exclusively free systems, a variety of exclusively proprietary systems, or both.

This is why I see a flaw in the following argument -

"Linux monoculture is not necessarily a good thing either ‘cause we want kids to be able to do things with computers, and not just Linux boxes and not just Windows boxes..."

Linux monoculture is risky not because "we want kids to be able to do things with computers" (though true in its own right), but because all computers would be exposed to any critical kernel flaw.

This reasoning is sometimes supported by an unsound argument I've encountered that "justifies" the acquisition of MS Windows and MS Office for K-12 schools. That is,

"We need kids to use what is popular. If we don't, we won't be preparing them for the 'real-world'"

If this were an actual fact, then monoculture avoidance for the reason stated would make sense. But it's not. Any student encouraged to critically think instead of memorize will know, or figure out quickly, how to operate any GUI or office suite thrown at them. So if unskilled GNU/Linux students (is that an oxymoron?) are actually being created by monoculture, then that is a pedagogical issue, not a choice of software issue.

I think it's critical to understand this as decision makers within K-12 schools may easily fall for this thinking. Basing a decision on an argument like this can needlessly set an organization back years.