Rick,

I've done a number of iterations on my dissertation introduction.
While I am pretty happy with the big-picture flow of it,
at a small and medium level I'm not at all happy with it.
However, I'm not exactly sure what is the best way to go,
and I could easily thrash a week on this until (maybe)
getting it right.  I would thus appreciate your opinion.

Following are my structural problems with the paper.  In general,
it should be readable, but I have not polished it to a great
extent.  It would seem a waste for me to do this, when
half the things I polished could get removed or moved to another
location and thus have to get rewritten to fit the new context. 

- I note that 1.1 takes up a lot more than 1.2 -- 1.5.  This
  may not be wrong, but it seems a bit odd to me...


- 1.1.1 seems a bit too detailed, but I'm not exactly sure
  what to cut; though the virtuous circle part doesn't
  seem to add too much.	


- 1.1.2 seems a good deal too complicated, and maybe 
  should be cut down to 50%-70% of its current size.
  However, I'm not sure what the best way to proceed is:
  
	- It seems that the discussion of granularity and
	  latency is nice to have before the 3 architectures.
	  However, this seems to break up the flow of the section.

	- It seems helpful to give the ballpark latencies so the reader
	  can understand why certain granularities don't work well
	  on certain architectures, etc.  However, these numbers
 	  seem to get in the way of the flow of the subsection.

Thanks for any comments on this!

Dave