We know, we know-- this is traditionally Wildly Off-Topic Microsoft-Bashing Day, and you're jonesing for your anti-Redmond fix. But we're hankering for a little something different today, so just for a change of pace, are there any objections to us celebrating Mildly On-Topic Microsoft User-Bashing Day instead? No? We thought not.

We should probably clarify, though: it's not us doing the bashing, it's Paul Murphy over at LinuxInsider (as pointed out by faithful viewer Sleeper)-- and actually, he's not doing any explicit bashing, either. But before you give up on this whole thing as a waste of time and resort to looking up ways to insult your Windows-user cohorts in Icelandic at Swearsaurus, check out what Paul did: in an attempt to quantify whether Windows users were any less intelligent than Mac users, he fed a bunch of text from Windows- and Mac-centric web sites into a UNIX-based text-processing utility called "style," which allegedly "produces readability metrics on text."

Now, assuming that you accept the premise that someone's "ability to read and write his or her native language" is a reasonable criterion by which to judge his or her intelligence, then the results are, at the very least, interesting if not necessarily conclusive. The highest set of readings Murphy provides as an example come from Henry James's The Golden Bowl and include something called a "Kincaid" score of 18.2 and a "Lix" mark of "64.4 = higher than school year 11." Another set of readings comes from articles in the respected publication The Christian Science Monitor (Kincaid: 10.4, Lix: 48.8 = school year 9). Now, compare those to a sample of Windows user text-- the scores of "reader comments hosted by PC Magazine" (Kincaid: 5.9, Lix: 32.3 = below school year 5) and "an MSN forum host" (Kincaid: 2.9, Lix: 21.5 = below school year 5).

That looks like a pretty wide gulf to us-- but why wouldn't it be? After all, readers posting to web sites aren't generally expected to hold themselves to the same standards as professional news publications and novelists writing for publication. But now consider the scores from "reader comments hosted by the MacInTouch site": Kincaid: 8.9, Lix: 40.5 = school year 6-- and the other five metrics given (ARI, Coleman-Liau, etc.) are generally much higher for the Mac sample versus the Windows samples. Granted, most MacInTouch readers probably aren't going to be writing for the Monitor or publishing psychological novels with themes of intense alienation anytime soon, but the difference between the Mac and Windows samples is clearly noticeable.

We should stress that we know absolutely nothing about what these metrics generated by the "style" utility are in the first place ("Fox Index"? "SMOG-grading"? Is this text-processing or some sort of meteorological app?) or what validity they possess; indeed, Murphy himself suggests that "you don't take any of it too seriously" because plenty of smart people claim that these readability scores are "meaningless." Then again, plenty of other smart people disagree, so it's up to you how you'd like to interpret the results-- especially since, as Mac users, you're obviously smart people yourselves. Right?