[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Condensed summary of CL Cleanup meeting results
- To: CL-Cleanup@SAIL.Stanford.EDU
- Subject: Condensed summary of CL Cleanup meeting results
- From: Kent M Pitman <KMP@STONY-BROOK.SCRC.Symbolics.COM>
- Date: Thu, 6 Apr 89 14:14 EDT
- Cc: firstname.lastname@example.org
Kathy Chapman asked for the information that was in my last burst of
messages in some sort of `summary' form. I had chosen to send out
individual messages because I assumed most people like myself who were
doing per-topic filing would find that simpler. But for those who are
just auditing this stuff and deleting it, here's my summary.
Just a reminder though -- these are just my notes and do not necessarily
represent the official story if it turns out there are discrepancies.
ADJUST-ARRAY-NOT-ADJUSTABLE - Deferred
BREAK-ON-WARNINGS-OBSOLETE - Passed
CLOS-CONDITIONS - Passed
CLOS-MACRO-COMPILATION - Tabled
Motion to replace v7 with v5 passed unamended
COERCE-INCOMPLETE - Withdrawn
COMMON-TYPE - Passed
COMPLEX-RATIONAL-RESULT - Passed
CONDITION-RESTARTS - Deferred
COPY-SYMBOL-COPY-PLIST - Passed
COPY-SYMBOL-COPY-PRINT-NAME - Passed
An incredibly weird vote on the question ``How many favor saying it passed
at the last meeting?'' passed N-0-M.
DEFINE-OPTIMIZER - Failed
There is new information which leads me to believe this might come up
again at the next meeting.
There were hardcopy amendments distributed.
My notes say that we approved amendments A & B (``permit all four'')
and we added an amendment that said that &ENVIRONMENT would not be
duplicated in a DEFMACRO lambda-list.
The amended proposal was passed N-0-1.
An amendment was made (I think by Barrett) to make DESCRIBE deal with
its second argument in the same way as PRINT does (that is, permitting
arguments of NIL and T).
The amended proposal passed 15-0.
Discussion on this was broken over two days with quite a number of
possible amendments discussed. Pitman came up with a written set of
amendments for Thursday which were discarded because Moon submitted a
revised proposal (consistent with those amendments, and adding at least
one other feature not covered in those separate amendments) on Thursday.
The revised proposal was v3, already mailed.
Moon's revised proposal was voted on, and passed 15-1.
Steele and Pitman drafted a revised proposal over lunch Thursday
which passed 17-0.
EQUALP-GENERIC - Withdrawn
ERROR-CHECKING-IN-NUMBERS-CHAPTER - Deferred
Withdrawn in favor of providing Kathy with editorial advice to
`minimize implicit requirements on debuggers' in the presentation of the
debugger in the standard.
EVAL-WHEN-NON-TOP-LEVEL - Option GENERALIZE-EVAL-NEW-KEYWORDS passed
Moon offered some amendments the effect of which were to allow you to throw again
to the same tag as you were already throwing to; specifically:
In the first paragraph of the MINIMAL proposal, delete "or is itself the
target exit" and change "events (c) and (d) at" to "event (c) occurs at".
After the first paragraph add a new paragraph "The event (d) occurs at the
end of the transfer of control."
The proposal was amended, and the amended option MINIMAL passed 11-5.
Withdrawn; editors were instructed to be explicitly vague 16-1.
The net effect is that option LARGE passed with amendments to strike 7,8,9.
GENSYM-NAME-STICKINESS - Passed
Passed 17-0 with hand-written amendments by Pitman (and agreement that the
`obvious' write-o's would be corrected).
HASH-TABLE-SIZE - Deferred
The net effect is that NEW-MACRO passed with the name of SELECT-PACKAGE
changed to IN-PACKAGE.
Version 2, which makes LOAD and *COMPILE-FILE* bind *READTABLE* (and
which does -not- introduce the IN-SYNTAX macro mentioned in version 1)
passed 12-0-3. This version was distributed by KMP in handwritten form.
The net effect is that this passed with amendment to rename package USER to
COMMON-LISP-USER with nickname CL-USER.
GZ wanted an amendment to strike item 8 from this list.
Sandra had some concern about the penultimate paragraph where she wanted a
prohibition on the ability to trace local functions (in implementations that
permit that). Moon thinks the proposal was amended to explicitly allow
tracing of such local function bindings.
We went round in circles about item 8. A straw poll to send this back for
more work failed 6-10, so we kept on.
A motion was made to terminate discussion. This passed by 2/3 vote.
Moon's notes say item 8 may need further refinement, as for instance by GLS's
amendment. The goal is to separate properties into the ones the user can
bash and the ones the user cannot bash.
Ultimately, I have written in my notes that we voted on
``proposal replaced by RPG, item 8 struck, w/ Sandra's prohibition
to trace local functions''
and that it passed 14-3.
Item 8 might be revisited at the next meeting.
Moon proposed friendly amendment to use the name MAKE-LOAD-FORM-SAVING-SLOTS.
The amended proposal passed 18-0.
LOAD-TRUENAME - Deferred
LOCALLY-TOP-LEVEL - Passed
LOOP-AND-DISCREPANCY - Passed
Tabled Tuesday. Not re-raised Thursday.
Might or might not come up at next meeting.
MACRO-ENVIRONMENT-EXTENT - option DYNAMIC passed 15-1.
MAKE-STRING-FILL-POINTER - Withdrawn
This was on the agenda but not discussed.
I expect it will come up at the next meeting.
PATHNAME-CANONICAL-TYPE - Deferred
PATHNAME-COMPONENT-CASE - Deferred
PATHNAME-COMPONENT-VALUE - Deferred
PATHNAME-LOGICAL - Deferred (but not seriously expected to come up again)
This was deferred from Tuesday to Thursday but then didn't come
around for discussion due to lack of time.
It might come up at the next meeting.
PATHNAME-SUBDIRECTORY-LIST - Deferred
PATHNAME-SYNTAX-ERROR-TIME - Deferred
PATHNAME-WILD - Deferred
Kim Barrett mentioned again that he wants to reopen this, but nothing was done.
PRETTY-PRINT-INTERFACE - Deferred
This was deferred from Tuesday to Thursday but then didn't come up
for discussion. Being a clarification, I expect it to come up at the
This was deferred from Tuesday to Thursday but then didn't come up
for discussion. Since some people consider this almost a clarification,
this might come up at the next meeting.
PROCLAIM-LEXICAL - Failed
READ-CASE-SENSITIVITY - Deferred
This issue was mentioned but was not on my list of possible topics.
No action was taken. It's probably a clarification and I guess it might
still come up later.
REAL-NUMBER-TYPE - Passed (the `union' of the two proposals)
REMF-DESTRUCTION-UNSPECIFIED - Passed
Reconsidered but no change made. Previous vote stands.
This was on the agenda but never gotten to.
I think it might come up at the next meeting.
SYMBOL-MACROLET-SEMANTICS - Version 6 passed.
SYNTACTIC-ENVIRONMENT-ACCESS - Deferred
This didn't come up. Being a clarification, I assume it might come up next meeting.
Moon proposed time zones be multiples of 1/3600 so that they were
even numbers of seconds. (Some people suspected this was a subtle
marketing ploy for Symbolics.) This amendment was accepted 11-0.
Pitman proposed that we limit time zone to the range [-24,24], inclusive.
The inclusive was to allow countries to disagree on which end was open,
since it was agreed that the correct value here is a largely political,
rather than technical issue. The amendment was accepted 8-5.
The full proposal with both amendments passed 18-0.
Some people had some concerns about a previous version of this that passed.
We wanted to reconsider this, but didn't have the right hardcopy.
No action was taken. This will probably come up next meeting.
UNDEFINED-VARIABLES-AND-FUNCTIONS - Deferred
This passed 11-6 with an amendment to say it defers "warnings" rather than "actions"
and with an amendment to say it does not apply to the COMPILE function (only to