The notion of a Linux workbench as a kind of Wright Brothers bicycle shop, a space for inventors with low barriers to entry, other than a willingness to learn, is somewhat new (gnu). In the early days of AT&T Unix, the only low cost POSIX was FreeBSD whereas most NGOs, worthy causes, had to make do with the "PC revolution's" DOS-based solutions, later Windows.
A GUI on UNIX meant Solaris or SGI or something else equally unaffordable. The rich and spoiled used UNIX, VMS, CMS etc., (more cathedral types, than bazaar types), the idea of "programming for everybody" not yet having traction. Apple went POSIX only after the Jobs @ NeXT chapter, having been more generous with NGOs pre OS X (not free to this day).
My own career as a tech savvy guy committed to helping NGOs, an idealist of sorts, took me through the Microsoft experience up through my using Ubuntu in tandem, by way of cygwin along the way. Although familiar with and respectful of the Bourne Again religion, I'm in no way the bash guru some are.
When meeting the econometrician recently, I was extolling the spare starkness of the version control systems (cvs, svn, bzr, git... hg), but I'm not one to memorize all their switches. I fumble at the command line much as I fumble with Tinkerbell, and no Boeing 747s have emerged from my humble garage of late e.g. I have not a single line of code at CodePlex, home of "Python Fe" (IronPython).
Thanks to cygwin, I'm thinking our early math labs, stocked with hand-me-downs in many cases, won't have to play second fiddle all the time, even to private industry's more opulent setups.
Our students have the same hunger for command line skills as their forebears, but maybe their day care center doesn't offer any real numeracy (gnumeracy) training. A laptop in a coffee shop works just as well, or maybe the shop provides its own workstations.
Kids learned the "forbidden math" on their own, while at "school" it was all about dinking around with doofy calculators (remember those?).
This was the early 21st century recall, with many 1900s holdovers squeezing out this "analog math" from their gruel sack, tree killer textbooks, meanwhile turning their backs on our best heritage. Our guy's 42 PhDs weren't enough or whatever (I forget all the reasoning of which there was precious little ("stonewalling" was their game, laughably weak in retrospect, but nevertheless devastating to many a promising career)).