Issue LOOP-PRESENT-SYMBOLS-TYPO Writeup

Status:         Accepted 3/31/92

Issue:          LOOP-PRESENT-SYMBOLS-TYPO

References:     X3J13/92-101, page 6-13, 6-25; X3J13/89-004, page 2-14

Related issues: 

Category:       CHANGE

Edit history:   24-Feb-92 Version 1 by JonL


Problem description:


  Iteration over "present" symbols by the loop schema PRESENT-SYMBOLS

  excludes all present but external symbols.



Proposal (LOOP-PRESENT-SYMBOLS-TYPO:FLUSH-WRONG-WORDS):


  Remove the words:

    "... but not \term{external symbols} of that \term{package}."

  from the first sentence of the draft description on page 6-13.


  Add the line:

        "\OUT THIS"

  to the output of the draft example at the top of page 6-25.


Rationale:


  This is clearly a typo since the phrase "present symbols" has had

  a precise technical meaning since the early 1980's;  furthermore,

  it has a significant adverse impact on the intended semantics.


Current practice:


  All of Lucid's implementation use the meaning of "present" rather 

  than "internal"; apparently one of Symbolics' implementations uses

  "internal".


Cost to Implementors:


  Surely, very small (probably more to marketing types.)


Cost to Users:


  Again, small if any at all.  


Cost of non-adoption:


  Occurance of many obscure bugs (looping over present symbols misses

  some of those present); egg on face for such a misnomered feature.



Performance impact:


  None.


Editorial impact:


  KMP says, in an email msg of Thu, 13 Feb 1992 22:57-0500:


    I estimate the editorial impact to be about 2 minutes of work ..."


Benefits:


  See Cost of non-adoption.


Esthetics:


  Say what you mean.


Discussion:


  Possibly some user might think that this odd wording is simply a means 

  of resurrecting the non-proposed schema for INTERNAL-SYMBOLS; but this 

  schema was considered and rejected on its own (non) merits.


  Some "bibliographic archeology" of the actual LOOP proposal --

  X3J13/89-004 -- shows that the wording in the original sources were 

  very confusing, leading to the conculsion that the offensive phrase 

  was inserted (by someone not knowing what they were doing?) into

  X3J13/89-004 to lessen the "confusion"


  Moon has questioned how the committee could have voted upon

  X3J13/89-004 without reading it more carefully and noticing this 

  glitch.  JonL compares the scrutiny of X3J13/89-004 with that given 

  to X3J13/88-002R (i.e., probably no on re-read either before voting.)