Return-Path: Received: from SEGATE.SUNET.SE by xiron.pc.helsinki.fi with smtp (Linux Smail3.1.28.1 #1) id m0tF0Du-0000ZTC; Mon, 13 Nov 95 16:51 EET Message-Id: Received: from listmail.sunet.se by SEGATE.SUNET.SE (LSMTP for OpenVMS v1.0a) with SMTP id 4B665B32 ; Mon, 13 Nov 1995 15:51:01 +0100 Date: Mon, 13 Nov 1995 09:37:59 -0500 Reply-To: John Cowan Sender: Lojban list From: John Cowan Subject: Re: buffer vowel X-To: Lojban List To: Veijo Vilva In-Reply-To: <199511121421.JAA16150@locke.ccil.org> from "ucleaar" at Nov 12, 95 02:14:52 pm Content-Length: 1104 Lines: 32 la .and. cusku di'e > I vote for buffer-hyphen equivalence. Unfortunately, it won't work in naive form; I found a counterexample: If the "y" is omissible from: bastyrapli < "basti rapli" = "replace-repeat" then it becomes the different lujvo bas-tra-pli < "basti tarti pilno" = "replace-behave-use" (Don't ask me what they mean, but they are both valid lujvo.) > As for the slinkui test, I don't see how it's relevant here. > Also, I've never understood the rationale for it. The tosmabru > test says if a string is potentially ambiguous between lujvo > and cmavo+lujvo, then the latter wins. The slinkui test says > if a string is potentially ambiguous between lujvo and cmavo > +lujvo, either lujvo is banned. I am confused. No, the slinku'i test says that if a string is potentially ambiguous between le'avla and cmavo+lujvo, the le'avla is banned. > Or do as Chris suggests, and scrap buffering. I'm beginning to think that this is the best alternative available. -- John Cowan cowan@ccil.org e'osai ko sarji la lojban.