Right. So, where is it?
I'll give you WiBro in S. Korea, and a few niches in the US. But that isn't enough of a deployment to contradict my overall point.
Right, for the tablet, the two approaches that work are: accommodate a modular approach: PCI-Express-Mini card (could be done, with some re-design of the internals), have a well placed USB port (it doesn't; along the top, in a way that wouldn't interfere with the sliding screen, would have been better), or invent a new modular connector (bad idea, IMO), or have multiple devices: one with no WWAN, ones with existing widely deployed WWANs (1xRTT/EVDO version, GPRS/EDGE/HSPA version), ones with niche/emerging WWANS (WiMAX version, LTE version).
The latter would have been reasonable if they had remembered the "ones with existing widely deployed WWANs". They didn't, so that has failed, IMO (and in the opinions of several other people in this discussion; but clearly that's the core of the disagreement -- whether or not skipping that item constitutes a marketing failure, thus far).