Abstract patterns

[ke han]

Richard,
I've just read your paper on abstract patterns (found it in-lined in  
a mailist post from a year or two ago).
After a day or so of letting it sink in, I am coming around to your  
point of view on its usefulness in encapsulating term structure.
The problems with records is obvious but the record update and  
matching syntax is much more readable than the update function of  
your abstract patterns (see your quote below).

<quote>
     L2 = L#log{blocked_by = none, status = ok}

which is arguably prettier than

     L2 = L#log_blocked_by(none)#log_status(ok)
</quote>

<quote>
     #log{status == S, users = N} = L

which is arguably prettier than

    #log_status(S) = #log_users(N) = L
</quote>

So the question I have is, is this syntax as good as it gets?  Is  
there any way to improve on this syntax?  particularly the update and  
matching.

thanks, ke han