From nobody@digitalkingdom.org Thu Dec 14 07:53:23 2006 Received: with ECARTIS (v1.0.0; list lojban-list); Thu, 14 Dec 2006 07:53:24 -0800 (PST) Received: from nobody by chain.digitalkingdom.org with local (Exim 4.63) (envelope-from ) id 1GustJ-000798-BD for lojban-list-real@lojban.org; Thu, 14 Dec 2006 07:53:01 -0800 Received: from silene.metacarta.com ([65.77.47.18]) by chain.digitalkingdom.org with esmtps (TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.63) (envelope-from ) id 1Gust8-00078Q-A4 for lojban-list@lojban.org; Thu, 14 Dec 2006 07:53:01 -0800 Received: from localhost (silene.metacarta.com [65.77.47.18]) by silene.metacarta.com (Postfix) with ESMTP id 2B57214C850D for ; Thu, 14 Dec 2006 10:52:43 -0500 (EST) Received: from silene.metacarta.com ([65.77.47.18]) by localhost (silene.metacarta.com [65.77.47.18]) (amavisd-new, port 10024) with ESMTP id 22807-08 for ; Thu, 14 Dec 2006 10:52:35 -0500 (EST) Received: from [65.77.47.178] (cheyenne.metacarta.com [65.77.47.178]) by silene.metacarta.com (Postfix) with ESMTP id D7DBD14C852A for ; Thu, 14 Dec 2006 10:34:33 -0500 (EST) Message-ID: <45816F09.4060602@ropine.com> Date: Thu, 14 Dec 2006 10:34:33 -0500 From: Seth Gordon User-Agent: Debian Thunderbird 1.0.2 (X11/20060926) X-Accept-Language: en-us, en MIME-Version: 1.0 To: lojban-list@lojban.org Subject: [lojban] Using continuations to model quantifiers, focus, and coordination Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new-20030616-p10 (Debian) at metacarta.com X-Spam-Score: -2.6 (--) X-archive-position: 13360 X-ecartis-version: Ecartis v1.0.0 Sender: lojban-list-bounce@lojban.org Errors-to: lojban-list-bounce@lojban.org X-original-sender: sethg@ropine.com Precedence: bulk Reply-to: lojban-list@lojban.org X-list: lojban-list In computer science, the "continuation" of some computation (say, a function call deep within the bowels of some program) is whatever the rest of the system is going to do with the results of that computation. (The Wikipedia article http://en.wikipedia.org/wiki/Continuation has an overview and some links.) One thing you can do with continuations is rewrite a program in "continuation-passing style": every time you have a function of n arguments that returns a value to its caller, you can replace it with a function of n+1 arguments that passes a value to its last argument. There is a theory of grammar that maps sentences onto logical functions, but this theory has trouble handling quantifiers like "everyone", "someone", and "only". If I want to parse "Matt spoke", I can treat "Matt" as an individual and treat "spoke" as a function that maps an individual onto a truth value (given X, did X speak?). But if I use the same system to parse "Everyone spoke", "everyone" is not an individual but a function in itself (given an individual X and an action Y, did X do Y?). Furthermore, if we turn to the sentence "Matt spoke to everyone", it seems like we need to interpret "everyone" as a function of another type. A recent paper by Chris Barker at UCSD has an elegant solution to this problem: translate this theory of grammar into continuation-passing style, so that "Matt", too, is a function (given an action Y, did Matt do Y?). Some folks on this list might be interested in the paper, for obvious reasons. http://www.cs.bham.ac.uk/~hxt/cw04/barker.pdf To unsubscribe from this list, send mail to lojban-list-request@lojban.org with the subject unsubscribe, or go to http://www.lojban.org/lsg2/, or if you're really stuck, send mail to secretary@lojban.org for help.