DYNAMIC-WIND vs. multi-processing

[Aubrey Jaffer <jaffer@martigny.ai.mit.edu>]

   Date: Wed, 12 May 93 14:59:51 -0400
   From: Gerald Jay Sussman <gjs@martigny.ai.mit.edu>


	If DYNAMIC-WIND should be used only for
	dynamic binding I suggest that we should either say so in the
	description or just scrap it in favor of an explicit dynamic binding
	construct.

   The problem with this is that fluids are not really sufficient in an
   operating system.  You also have things that must be done when
   switching users, such as changing the states of I/O buffers and
   interlocks.

Would saying "DYNAMIC-WIND should be used only for reversible
operations" fix it?

   I really don't see how you can make an operating system
   without such a thing.

DYNAMIC-WIND is not a primitive.  It can be implemented in terms of
R4RS Scheme (by redefining CWCC).  Therefore it must be possible to
write any program which uses DYNAMIC-WIND without using it.

I like DYNAMIC-WIND.  I am just pointing out that it screws up the
semantics of CWCC.