Return-Path: Received: from SEGATE.SUNET.SE by xiron.pc.helsinki.fi with smtp (Linux Smail3.1.28.1 #1) id m0tN5s9-0000ZUC; Wed, 6 Dec 95 00:30 EET Message-Id: Received: from listmail.sunet.se by SEGATE.SUNET.SE (LSMTP for OpenVMS v1.0a) with SMTP id 0F589116 ; Tue, 5 Dec 1995 23:30:00 +0100 Date: Tue, 5 Dec 1995 17:39:13 -0500 Reply-To: John Cowan Sender: Lojban list From: John Cowan Subject: TECH: PROPOSED GRAMMAR CHANGE 39: change BO to BIhE in MEX operators X-To: Lojban List To: Veijo Vilva Content-Length: 1383 Lines: 37 CHANGE 39 CURRENT LANGUAGE: A MEX operator can be made high-priority by prefixing it with BO. PROPOSED CHANGE: Eliminate this use of BO in favor of a new cmavo "bi'e" of selma'o BIhE. RATIONALE: The current use of BO is ambiguous, but this ambiguity is masked from YACC by the compounder. Consider the improbable, but grammatical, sentences: 1) li re su'i jebo pi'i re du li vo The number two plus or times two equals the number four. 2) li re su'i je bopi'i re du li vo The number two plus or times two equals the number four. The parser will always assume Example 1 rather than Example 2, because the compounder tries to compound the longest possible string. This particular example does not amount to any semantic difference, because neither tight-binding logical connection between operands (as in Example 1) nor tight-binding operands (as in Example 2) are needed in this sentence; however, it would be possible to concoct examples where the difference mattered. Even if this could be made to work by trickery, the use of BO as a prefix is highly unlojbanic; it occurs only on MEX operators, and was done there because a postfix BO would require compounding, and not all MEX operators can be compounded (some include full bridi). -- John Cowan cowan@ccil.org e'osai ko sarji la lojban.