Microsoft has a long history of interleaving really terrible releases with cleaned up offerings and therefore, given Vista, I had high hopes for Windows 7.
On quick review it appears to meet expectation: it's a pig, but it's a pig that works within the expectations and limits of the Wintel market place. What it feels like, in fact, is exactly what it seems to be: the hybridized 2003 minwin core from 2008 Server with improved driver recognition and handling on the inside and the most recent Office improvements in copying the Mac interface somewhat generalized to the GUI on the outside.
Overall, not too bad - and likely, I think, to appeal to those members of the Windows/XP crowd looking for a politically and emotionally acceptable excuse to jump back on the upgrade band wagon.
Notice, please, that I'm saying it's better suited to Wintel community use than Vista in large part because it seems to have fewer internal corruptions and a more Mac like exterior - not that it's in the same league with any of the major Unix players. Relative to MacOS X or SuSe's latest Linux desktop, Windows 7 is just another three legged horse, differing from Vista only in not also being lame on one side.
In many ways the improvement relative to Vista reminds me of Windows 98's improvements over Windows 95 - itself, of course, similarly MacOS inspired. But remember what came after 98? Right: ME and the end of the line for the tottering tower of complexity Microsoft had layered on top of DOS.
The original "Longhorn" ideas were largely based on PICK and might, I think, have led to a new foundational core OS for Microsoft had they been pursued after XP came out. But they weren't with Microsoft choosing, instead, to continue layering complexities on on the object VMS ideas Dave Cutler brought to Microsoft in the early 1990s - and like DOS with Windows 98, I think that process has pretty much reached its limits.
Thus while I see even the beta for Windows 7 offering clear advantages over Vista, there just isn't any "next" here - meaning that its expected success will make the inevitable transition to whatever really does come next (OpenBSD with a Windows shell?) much harder for Microsoft and its loyalists than it would have been had the company taken the opportunities created by Vista's market failure to introduce some new ideas now.