From nobody@digitalkingdom.org Wed Nov 16 11:28:30 2005 Received: with ECARTIS (v1.0.0; list lojban-list); Wed, 16 Nov 2005 11:28:31 -0800 (PST) Received: from nobody by chain.digitalkingdom.org with local (Exim 4.54) id 1EcSx8-00057Q-W2 for lojban-list-real@lojban.org; Wed, 16 Nov 2005 11:28:19 -0800 Received: from sabre-wulf.nvg.ntnu.no ([129.241.210.67]) by chain.digitalkingdom.org with esmtps (TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.54) id 1EcSwv-000571-IX; Wed, 16 Nov 2005 11:28:18 -0800 Received: from hagbart.nvg.ntnu.no (hagbart.nvg.ntnu.no [129.241.210.68]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by sabre-wulf.nvg.ntnu.no (Postfix) with ESMTP id 467BF94789; Wed, 16 Nov 2005 20:27:32 +0100 (CET) Date: Wed, 16 Nov 2005 20:27:34 +0100 (CET) From: Arnt Richard Johansen X-X-Sender: arj@hagbart.nvg.ntnu.no To: lojban-list@lojban.org, lojban-beginners@lojban.org Subject: [lojban] A ranking system for Lojban speaking proficiency? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-NVG-MailScanner-Information: Please contact the ISP for more information X-NVG-MailScanner: Found to be clean X-MailScanner-From: arj@nvg.org X-Spam-Score: -2.6 (--) X-archive-position: 10783 X-ecartis-version: Ecartis v1.0.0 Sender: lojban-list-bounce@lojban.org Errors-to: lojban-list-bounce@lojban.org X-original-sender: arj@nvg.org Precedence: bulk Reply-to: lojban-list@lojban.org X-list: lojban-list It has come to my attention that many beginners (and some advanced learners) are in search of a way of comparing their spoken Lojban skills with others. One way of doing this, that was suggested on IRC yesterday, is to set up a system based on the ELO ranking system[1], which was originally designed for the game of chess. The way we are thinking of adapting this to Lojban is roughly as follows: Each "match" is a conversation. After the conversation, both parties throw a secret ballot which indicates whether they thought themselves better or worse than their "opponent". If the ballots agree, a winner is designated appropriately. If the ballots disagree, the conversation is treated as a draw for the purposes of the ELO system. There are two reasons that I am bringing this to the mailing list: 1) What are the flaws of this scheme, and how can they be resolved? I suggested yesterday that it might be possible for someone to game the system by voting for themself every time. I don't know how serious this is. At worst, it will cause the malicious participant to better their score by getting a false draw against a stronger opponent, instead of the deserved loss that would lower their score. 2) Design and implementation of the software infrastructure. Many of the Lojban newbies are reportedly eager to volunteer their computer skills to Lojbanic causes. To get this idea flying, we would need some way of anonymously submitting ballots for conversation, *and* automatically update scores. There would also have to be some delay and/or obfuscation, to prevent people from inferring the vote of the opponent by comparing their current rating to the previous one. [1] http://en.wikipedia.org/wiki/ELO_rating_system -- Arnt Richard Johansen http://arj.nvg.org/ Many familiar with Descartes' work are likely to remember him from philosophy courses as that French guy who was wrong a lot. --Daniel Harbour 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.