Sgolds, excellent post, but you fail to address the point made by the IBM exec about Opteron outselling Itanium 2 on a server unit basis.
Re: First, Intel (and HP) set the bar for themselves back in 1994 when they claimed that IPF would effectively displace x86 by the start of the decade. Not only in servers, but on desktops also. It is not really accurate to say that Itanium was positioned at >4P, rather it is more accurate to say that Itanium has be repositioned at >4P. This point does not seem to be lost on the buying public - Itanium plays in a smaller market each year (with promises of a larger market in the future).
I really doubt that an executive from IBM would be misled by Intel's positioning of IPF. Maybe this particular executive is out of the loop, but I would expect Intel to spend a great deal of time explaining the prospects of Itanium, and what would be realistic in terms of sales.
As for the market shrinking, I responded to this in an earlier post on this forum. I think the RISC market is due for a comeback, given that the performance has vastly improved over x86 in the last couple years, and availability of Power4 and IPF based systems has grown considerably.
Re: Second, Intel (and HP) claimed that x86 would not be an effective way to migrate to 64-bits. AMD disproved that convincingly. AMD introduced an architecture which is a roadmap away from messy segmented, register-starved, stack-oriented legacy x86 and into a world of simple, flat architecture x86-64. As discussed here in length about a week ago, the AMD64 design is a roadmap to cut loose the parts of x86 which so irritate compiler writers and programmers. All done with compatibility.
This is largely irrelevant. Intel has reversed their position on x86-64, but they have not eased off of their IPF positioning in the least. The market seems to be suited to both architectures.
Re: Third, x86 has a history of defeating every attempt to displace it in the market. If the best way to predict the future is to apply the lessons of the past then Itanium is running against history. Computer-literate people sense that (even if they haven't fully vocalized the concept). Smaller, cheaper, faster - that is the mantra. Itanium violates the first two items. AMD64 rides with the wave of history, even those not familiar with the details expect it to overtake Itanium later this decade.
That's just the thing. IPF is not trying to displace x86, at least at this point. I think it might have the prospects to do it as some point, just like the PPC970 has shown itself to be a viable competitor. The same cannot be said for solutions in the past.
Re: Fourth, this is the traditional David vs. Golliath battle. As long as David looks competitive, David always gets better press! The world roots for Golliath to stumble.
The press may root for the underdog, but end users just want the best value. Money changes hands several times before a processor gets into the system of an end user, and all the layers are trying to maximize value. The common mistake I see on these kinds of forums is that everyone is end-user centric, rather than aiming to grasp the bigger picture.
Market Data 
Markets 
AI
