From - Thu Mar 14 11:54:08 1996 Received: from access1.digex.net (ql/6O0AY1b.Cw@access1.digex.net [205.197.245.192]) by locke.ccil.org (8.6.9/8.6.10) with ESMTP id CAA00688 for ; Thu, 14 Mar 1996 02:50:01 -0500 Received: (from lojbab@localhost) by access1.digex.net (8.6.12/8.6.12) id CAA09876 ; for ; Thu, 14 Mar 1996 02:51:40 -0500 Date: Thu, 14 Mar 1996 02:51:40 -0500 From: Logical Language Group Message-Id: <199603140751.CAA09876@access1.digex.net> To: veion@XIRON.PC.HELSINKI.FI Subject: Re: TECH: PROPOSED GRAMMAR CHANGE 2$i Cc: cowan@ccil.org X-Mozilla-Status: 0011 Content-Length: 824 I willl leave it to Cowan to decide whether he finds the work needed to make termsets work without the mniddle NUhU worthwhile. I think I am still skeptical that it is, and do not mind the middle NUhU as much as the plurality of rules, though it seems that what you are saying is that all you eliminate is the direct equivalent of sumti_94 GIks. I think that termsets and their power have never been well explored, and they will come to be more useful as people get more fluent in the language, and so I would rather have a more consistent set of rules, and a broader range of things called termsets, including single sumti_90, not merely modifier_82 permitted (I can imagine there to be times when brevity might lead to the BAI put only on the first term in the termset, and not on the others.) lojbab