From rinehuls@access.digex.net Fri Apr 4 20:16:14 1997 Received: from access4.digex.net (qlrhmEbBUV1EY@access4.digex.net [205.197.245.195]) by dkuug.dk (8.6.12/8.6.12) with ESMTP id UAA16001 for ; Fri, 4 Apr 1997 20:15:59 +0200 Received: from localhost (rinehuls@localhost) by access4.digex.net (8.8.4/8.8.4) with SMTP id NAA28245; Fri, 4 Apr 1997 13:12:23 -0500 (EST) Date: Fri, 4 Apr 1997 13:12:22 -0500 (EST) From: "william c. rinehuls" To: sc22docs@dkuug.dk cc: fjh@cs.mu.oz.au Subject: SC22 N2447 - SC22 Review of Two Vocabulary DISs Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII ____________________beginning of title page _______________________ ISO/IEC JTC 1/SC22 Programming languages, their environments and system software interfaces Secretariat: U.S.A. (ANSI) ISO/IEC JTC 1/SC22 N2447 April 1997 TITLE: SC22 Review of Two Vocabulary DISs SOURCE: Secretariat, ISO/IEC JTC 1/SC22 WORK ITEM: N/A STATUS: N/A CROSS REFERENCE: SC22 N2387, N2388 DOCUMENT TYPE: N/A ACTION: To SC22 Member Bodies and WG Conveners for information. Address reply to: ISO/IEC JTC 1/SC22 Secretariat William C. Rinehuls 8457 Rushing Creek Court Springfield, VA 22153 USA Tel: +1 (703) 912-9680 Fax: +1 (703) 912-2973 email: rinehuls@access.digex.net ___________________end of title page; beginning of text ________________ ISO/IEC JTC 1/SC22 Programming languages, their environments and system software interfaces Secretariat: U.S.A. (ANSI) April 5, 1997 Ms. Lisa Rajchel Secretariat, ISO/IEC JTC 1 American National Standards Institute 11 West 42nd Street New York, NY l0036 SUBJECT: SC22 Review of Two Vocabulary DISs Dear Lisa: At the request of ISO/IEC JTC 1, SC22 conducted a review of the following DISs currently in the JTC 1 letter ballot stage: DIS 2382-15 (Programming Languages) and DIS 2382-7 (Types of Languages). The SC22 review was hampered considerably but our inability to obtain pages 28-32 of DIS 2382-15. Those pages appear to contain taxonomy and examples of data types for Ada, Pascal and COBOL. Several attempts were made to obtain these pages from the Secretariat of SC1 to no avail. Taking that into consideration, SC22 does offer one specific comment dealing with the definition of "declarative language" in DIS 2382-7 and a number of comments dealing with the definitions in DIS 2382-15. These comments is attached. SC22 would still be interested in reviewing pages 23-32 of DIS 2382-15 if those pages can be made available to us. Sincerely WILLIAM C. RINEHULS Secretariat, JTC 1/SC22 Attachments cc: E. Blanchon, Secretariat SC1 R. Follett, Chairman, SC22 ______________________________________________________________________ Attachment A SPECIFIC SC22 COMMENT ON DIS 2382-7 The definition of "declarative language" in DIS 2382-7 states: "A programming language for expressing declarations". That definition is very much at odds with existing usage. The Oxford Dictionary of Computing states: "Since declarative languages are concerned with static rather than dynamic concepts (i.e., what rather than how), they do not depend on any inherent notion of ordering and there is no concept of flow of control and no assignment statement. Ideally a program in a declarative language would consist solely of an unordered set of equations, sufficient to characterize the desired result." This notion is entirely different from the definition in DIS 2382-7. The term "declarative" has already lost any real meaning it might have had even as a buzzword of the 1980's. It has been used to mean many different things; however, the Oxford Dictionary statements above best characterize the currently accepted definition of what a declarative language is. Therefore, SC22 suggests that the definition either be deleted or revised to reflect its current usage as defined in the Oxford Dictionary. ________________________________________________________________________ Attachment B Comments on ISO/IEC DIS 2382-15 15.01.09 disambiguation proposal Change the word "identifier" into "sequence of lexical tokens". explanation The word disambiguation is applicable to more complex constructs. The programming language C contains disambiguation rules for the construct "((identifier)())". Pascal has rules for which "if" the "else" belongs in the following construction: if ... then ... else ... else ... 15.02.04 implicit declaration proposal Remove the word "data" from the definition. explanation This standard does not seem to differentiate between declaration and definition. In our view a definition is a creation of a new entity while possibly naming it. A declaration is the naming of an existing entity. The programming language C allows one to use an identifier in a certain context in an expression that also declares the identifier as referring to an existing function. To be exact: an old-style function returning an int. We have chosen not to propose a change from "data" into "data or instruction". This would probably cause confusion. To allow for implicit declarations of non- data, we propose to delete the word "data". 15.02.10 declarative region proposed definition A portion of a program consisting of declarations. explanation In our view a declarative region is a portion of a program consisting solely of declarations. These declarations also determine scope and visibility. A different interpretation would be a region of a program that determines scope and visibility for declarations contained in that portion. The use of the word "other" in the original text is contradictory to our view of a declarative region. All declarations should be contained in the portion of the program that is referred to. 15.03.02 data object proposal Delete "and that is named ... program." explanation There are languages with unnamed data objects. So called pointers are used to refer to such objects. Pointers are not names and do not consist of characters from the program's character set. 15.04.06 logical type proposal Add Boolean type to the heading. Explanation It seems a logical thing to do. Index The index contains a reference to 15.1.11 for numerical constant and to 15.1.17 for string constant. 15.1.11 is a number used for a totally different definition and 15.1.17 is not present in this DIS. We would like the standard to contain definitions for both concepts. We suggest that the index be checked for consistency. ______________________end of SC22 N2447 _____________________________