Return-Path: LOJBAN%CUVMB.BITNET@vms.dc.LSOFT.COM Received: from SEGATE.SUNET.SE (segate.sunet.se [192.36.125.6]) by xiron.pc.helsinki.fi (8.6.12/8.6.9) with ESMTP id AAA11496 for ; Sun, 10 Dec 1995 00:59:34 +0200 Message-Id: <199512092259.AAA11496@xiron.pc.helsinki.fi> Received: from listmail.sunet.se by SEGATE.SUNET.SE (LSMTP for OpenVMS v1.0a) with SMTP id DA758CE7 ; Sat, 9 Dec 1995 23:59:34 +0100 Date: Sat, 9 Dec 1995 17:45:55 -0500 Reply-To: Jorge Llambias Sender: Lojban list From: Jorge Llambias Subject: Re: comments on CONN.TXT X-To: lojban@cuvmb.cc.columbia.edu, jorge@minerva.phyast.pitt.edu To: Veijo Vilva Content-Length: 743 Lines: 19 And: > > In other words, "act so as to > > make it true that X happens" does not need {do} to be an argument of X > > in general. > Is there no UI that'll do this? Yes, {ei} (obligation) is the closest to a command. Then there are {e'o} (suggestion) and {e'u} (request) for other uses of the imperative. And also there are others for less common uses. > I imagine that {ko} was influenced > partly by English imperatives and partly by the wish to make imperatives > that contain references to the addressee one cmavo shorter. Yes, I'm not saying that {ko} is not useful. Many commands contain {do} in the agent place, so {ko} may be a good shortening device. I'm just commenting that it is not the general way of doing imperatives. Jorge