J3/00-211

Date:        30 May 2000
To:          J3
From:        R. Maine
Subject:     terminology of "interoperates with"

The terminology "interoperates with" is misleading as it is used in
the existing draft.  The active terminology implies some relation
between a Fortran and a C thing that interoperate with each other.
But the usage in the draft only implies a form of compatability.  For
example, when the draft says that two variables interoperate, this
means only that they have appropriate types.  I might suggest that
some form of the term "compatable" would be appropriate, but both
the Fortran and C standards already use the word "compatable" in
multiple senses; adding another sense might also add possibilities
for confusion.  I suggest the term "is interoperable with" as a
replacement for "interoperates with".  To me, this terminology
seems to fit our useage better.

The current draft does not use the term "interoperable" (as confirmed
by grep).  The following edits cover every place that the term
"interoperate" appears, including those places where no change is
needed.

EDITS

intro

  no change needed - [xvii:19]

c04

  "interoperates" -> "is interoperable" - [41:30]
  (This sentence is pretty ugly in both the original and the revision,
  but I'll ignore that for now.  Might be easier if we could talk
  about types being interoperable.  Might go there later.)

  "interoperates with" -> "is interoperable with" - [58:15,33]

c05

  "interoperates with" -> "is interoperable with" - [78:37]
  (Its not clear to me that the usage here is actually consistent
  with with usage elsewhere, but I'll leave that question).

  "does not interoperate with" -> "is not interoperable with" -
  [79:16]

c12


  "does not interoperate with" -> "is not interoperable with" -
  [253:39], [267:1,5,6]

  "interoperate" -> "be interoperable" - [273:5]
  (This section needs work, but that's a separate issue).

  "than" -> "that" - [273:5] (Unrelated typo).

c16

  ignore because they are in J3 notes - [385:29,387:24,28,30],
  [388:28], [392:41]

  "<<interoperate>> with" -> "be <<interoperable>> with" - [387:7]
  Also change the index entry referencing this area.

  "interoperates with" -> "is interoperable with" - [387:8,15,17],
  [388:4], [389:11,21,24,37], [390:33,42,43], [391:2,19,22,29,40],
  [392:25,33,37], [393:42], [394:24,27]

  (paper 00-199r1 moves the [393:42] occurance.  Do both this
  change and the move.)

  "interoperate with" -> "be interoperable with" - [387:10],
  [394:10]

  "interoperate with" -> "are interoperable with" - [390:35],
  [391:14], [394:26]

  "interoperate" -> "are interoperable" - [392:28]

  "does not interoperate with" -> "is not interoperable with" -
  [387:18-19]

  "would interoperate with" -> "is interoperable with" - [391:32]

  "cannot interoperate with" -> "is not interoperable with" -
  [391:36]

  no change needed - [389:19]

  no change now - [394:8]
  I don't know what this is trying to say anyway.  There's already
  a J3 note on it.

  no change now - [394:31,32]
  These uses are already inconsistent.  I'm not sure what the best
  fix is, but its not a subject for this paper.  I think they are
  trying to talk about being "linked" instead of about interoperating.