Whether you think the show has a future or not, we've discovered one significant drawback to Macworld Expo Boston being such a small affair: when nothing particularly dramatic happens during the Steveless show in the first place, the inevitable post-Expo lull that settles in afterwards is like a freakin' vacuum. Seriously, now that everyone's done their obligatory two cents on "Expo was small, but a success nonetheless" and/or "Expo was an unmitigated disaster of a magnitude not seen since the advent of Oscar Mayer's 'Sack o' Sauce in a Can o' Meat'," we can't remember the Apple world ever having this little to talk about.

Then again, we have really short memories. In fact, we've completely forgotten the point we were trying to make.

So in the absence of anything exciting or, indeed, present in our short-term memory, we suppose now would be a good time to double back a little and cover a development that was eclipsed by the only husky-size Apple news of the week (i.e. those fabulous Q3 results). Do you remember when we mentioned that Apple has projected that "40 per cent of Xserve sales are destined for use in" supercomputing clusters? At the time we noted that the only two announced G5 clusters had been the groundbreaking Virginia Tech project and the Army's even larger "MACH 5" cluster announced just a couple of weeks earlier, but that the two had apparently already been breeding like bunnies given reports of a new "baby cluster" of 256 Xserves newly under construction at UCLA.

Well, evidently the spawning didn't stop there, because MacRumors recently noted the arrival of another baby cluster, also comprised of 256 Xserves-- apparently that's a common cluster birth weight-- that's being stapled together on the opposite coast at the University of Maine. But the lineage gets tricky, because a university press release reveals that the "new 256 node Apple Xserve G5 system" (dubbed "a baby MACH 5") is "funded by the Army under contract with UMaine"-- yet at the same time, the school insists that MACH 5 itself is "based in part on UMaine research funded by the Army to develop more powerful and less expensive computing technology for military research." So the question of which cluster begat which gets a little murky, and right now we're favoring a chicken-and-egg theory that involves time travel and Fry becoming his own grandfather.

(Incidentally, the press release also repeats the same mistake about MACH 5 being "expected to be second in speed only to the $350 million Earth Simulator computer in Japan"; as we've pointed out in the past, MACH 5 almost certainly wouldn't have ranked second even in the current TOP500 listings, but apparently some people will never understand the difference between theoretical peak performance and real-life measured results. Kinda like that whole "immaculate conception" concept that people always get wrong. It makes us lose our patience.)

So let's see, now-- that makes four G5-based supercomputers at least partially constructed in the year since the chip was first unveiled. Even more impressive, it's four Xserve G5-based supercomputers since Apple first started shipping the product less than four months ago. Suppose this trend has legs? Could it be, dare we say, a bona fide phenomenon? Just how many more "large clusters" does Apple "anticipate"? Will all these new clusters suck up all the G5s, leaving Mac users high and dry-- uh, higher and drier? And who took our Pop-Tart that was sitting here on the desk?

Seriously, has anyone seen it?

Aw, man. Now we gotta make toast or something.