COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	All references are to the August 5 printing of the EXCELSIOR edition.
C00011 ENDMK
C⊗;
All references are to the August 5 printing of the EXCELSIOR edition.


p. viii:  Eric Benson works at Symbolics, Palo Alto.  Rod Brooks has moved to
Stanford.

p. 2 (paragraph marked "Expressiveness"):  Change "...not only from MACLISP but
from INTERLISP,..." to "...not only from MACLISP, but also from INTERLISP,...".

p. 5:  Here, and throughout the manual, the function definition format has no
blank line before the definition.  All the other descriptions (for variable, 
macro, etc.) have one.

p. 14 (1st paragraph):  "description" spelled "decription".

p. 15 (4th paragraph):  The second sentence should pair SHORT-FLOAT with SINGLE-
FLOAT, and pair DOUBLE-FLOAT with LONG-FLOAT.
 
p. 23 (4th paragraph):  Scribe typo in sentence which reads "See chapter PACKAG
(page PACKAG)."  Also occurs on pages 129, 269, and 274.

p. 30 (CONTORTED-EXAMPLE trace):  In 3rd call to CONTORTED-EXAMPLE, shouldn't 
the last argument be 0?

p. 31 (2nd paragraph):  "causing" spelled "cuasing".

p. 39:  "(STRING size)" should not be indented.

p. 50 (1st paragraph):  &AUX variables are bound sequentially, so the example
should use LET* instead of LET.

p. 62 (examples for EQL):  Mention that (EQL -0.0 0.0) is false.

p. 62 (paragraph after the above):  "numbers" spelled "numebrs".

p. 68 (1st paragraph of FUNCTION definition):  Where is "lexical closure" 
defined?  There is no "lexical closure" or "closure" type.  None of the sections
describing functions (p. 24, 47) talk about it.  Neither are they mentioned 
under APPLY or FUNCALL.  This is the same problem as noted in the Memorial Day 
ballot, and it has not been "dealt with".

p. 73 (table at the bottom):  The tabulation of the heading is confusing.

p. 75:  The arguments for SHIFTF should read "SHIFTF {place}+ newvalue".

p. 75:  The examples for SHIFTF don't demonstrate any shifting.

p. 83:  The examples for APPLY and FUNCALL should use FUNCTION, not QUOTE.

p. 89 (3rd example for WHEN):  NIL should not be quoted.

p. 92 (3rd paragraph):  First sentence should read "The lexical scoping of the 
block name is fully general,...". [inserted "is"]

p. 105 (paragraph labeled "Implicit PROGN contexts"):  "but" spelled "bu".

p. 112 (4th paragraph):  "causes" spelled "cuases".

p. 113:  In the example for DEFMACRO, it seems confusing to be defining a
special form, IF, in terms of a macro, COND.  Similarly, perhaps AND and OR
should be defined in terms of IF rather than COND, even though they are not
special forms. 

p. 117:  The table of forms containing declarations is messed up.  The page
numbers for LOCALLY and MACROLET are missing.

p. 119 (3rd paragraph):  "mentioned" spelled "mnetioned"

p. 144:  In the DO-EXTERNAL-SYMBOLS definition "statement" written "stmt".

p. 145:  "*MODULES*" written "**MODULES**".

p. 158 (Implementation note):  "defined in" written "define din".

p. 159 (EXPT):  Is "↑" not going to be a defined synonym???  It certainly
makes for clearer code.  If the reason was consistancy with the other 
exponential and logarithmic functions, it should be noted that its arguments
are the reverse of LOG's.

p. 165 (top of page):  The word "excluded" shouldn't be indented.

p. 197 (3rd paragraph):  The 1st sentence should read "The type of the result
sequence is specified by the argument `result-type', which must be a subtype
of SEQUENCE, as for the function COERCE."

p. 203 (3rd paragraph):  In MISMATCH definition, shouldn't the wording be 
"relative to sequence1" rather than "within sequence1" to cover the case where 
sequence1 is shorter.
shorter.

p. 205: The example for MERGE at the bottom of the page is missing the 
`result-type' argument.

p. 217 (compatibility note for MEMBER):  In equating Common Lisp's MEMBER with
MacLisp's MEMQ, the documentation has finessed the difference between the use
of EQL (in the former) and EQL (in the latter) for comparison.  Similarly on 
p. 220 for ASSOC.

p. 229 (3rd paragraph):  "unspecified or nil" written "unspecified for nil".

p. 250:  In the definition of the :PRINT-FUNCTION option to DEFSTRUCT, the "#S"
syntax should be mentioned as the default.

p. 278:  In the example for the "#nA" definition, the second row of the matrix
is improperly indented.

p. 283 (3rd paragraph):  "function" spelled "fucntion".

p. 286 (3rd paragraph from the bottom):  The "#S" syntax should be mentioned as the 
default for :PRINT-FUNCTION.

p. 289 (2nd paragraph):  "error" spelled "erro".  "because" spelled "becaue".

p. 290 (4th paragraph):  The CONS backquotes its arguments rather than quoting
them.
 
p. 298 (7th paragraph):  "dependent" spelled "depdent".

p. 318 (1st paragraph):  In the example for MERGE-PATHNAMES, the function is call
"MERGE-PATHNAMES-DEFAULTS".

INDEX:  There is no entry any where for LAMBDA.  Is there going to eventually be
a unified index?
 
GENERAL:  We could not find any place in the manual which explicitly describes
the behavior of variables which are not explicitly declared SPECIAL or are
"lexically apparent".  One would assume that they are declared SPECIAL as in
MACLISP or ZETALISP



Eric Benson
Paul Wieneke