From nobody@digitalkingdom.org Mon Jun 23 18:53:08 2008 Received: with ECARTIS (v1.0.0; list lojban-list); Mon, 23 Jun 2008 18:53:08 -0700 (PDT) Received: from nobody by chain.digitalkingdom.org with local (Exim 4.69) (envelope-from ) id 1KAxiW-0008UV-D5 for lojban-list-real@lojban.org; Mon, 23 Jun 2008 18:53:08 -0700 Received: from eastrmmtao104.cox.net ([68.230.240.46]) by chain.digitalkingdom.org with esmtp (Exim 4.69) (envelope-from ) id 1KAxiQ-0008U4-TP for lojban-list@lojban.org; Mon, 23 Jun 2008 18:53:08 -0700 Received: from eastrmimpo01.cox.net ([68.1.16.119]) by eastrmmtao104.cox.net (InterMail vM.7.08.02.01 201-2186-121-102-20070209) with ESMTP id <20080624015258.UXDP7964.eastrmmtao104.cox.net@eastrmimpo01.cox.net> for ; Mon, 23 Jun 2008 21:52:58 -0400 Received: from [192.168.1.100] ([70.187.229.105]) by eastrmimpo01.cox.net with bizsmtp id hdsv1Z00c2H5Hc602dsvap; Mon, 23 Jun 2008 21:52:56 -0400 Message-ID: <48605377.3010103@lojban.org> Date: Mon, 23 Jun 2008 21:52:55 -0400 From: Robert LeChevalier User-Agent: Mozilla Thunderbird 1.0.7 (Windows/20050923) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Lojban List Subject: [lojban] Random Sentence Generator - Visual basic help needed Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Score: 0.0 X-Spam-Score-Int: 0 X-Spam-Bar: / X-archive-position: 14543 X-ecartis-version: Ecartis v1.0.0 Sender: lojban-list-bounce@lojban.org Errors-to: lojban-list-bounce@lojban.org X-original-sender: lojbab@lojban.org Precedence: bulk Reply-to: lojban-list@lojban.org X-list: lojban-list All the recent work on a glosser and the PEG Grammar inspired me to corner Nora during her last vacation, and we finally finished updating the Random Sentence Generator to the baseline grammar, something I started only 10 years ago %^). This involved converting the program from the old DOS-based Basic compiler, which wouldn't compile it any more, probably due to the abuse we inflicted with string handling, to Visual Basic which can handle it. The program has been tested more or less using Visual Basic Express 2005, which we downloaded from Microsoft. However our understanding is that we cannot create a standalone executable from VB Express - anyone who wants to use the RSG would have to download VB Express onto their machine in order to have the necessary run time library, and this isn't satisfactory. The best solution would be to have someone who has Visual Studio, or one of the comparable Visual Basic packages prepare a standalone executable, and perhaps even a website executable (I am not sure what is involved in that). We could zip up the project directory and the data files, and email them to a willing person who has more of a clue than we do (Nora might eventually learn how to do it at work, but rarely has any time to work on such things, and so far as we know the program is usable now). People who detest Microsoft are of course free to port the program to another version of Basic, or even another language, after it is released (we'll ask Robin make the source available on the Lojban website, but be forewarned that this is incredibly ancient code, much of which were written on a TRS-80 and is loaded with minimally commented 2 letter variables), but we want to make sure this version works first on some machine other than ours. For those unfamiliar with the RSG, it has 6 levels of complexity, which were based on the outline for the original Lojban textbook that I started back in 1989, but completed only 6 and a fraction chapters of (this was rewritten by John Cowan into a 20+ lesson collection which I think is the version now found on the Lojban website, preceded by an introductory lesson that I wrote anew. Alas, the rewrite no longer corresponds to the original outline, but it still serves as a basis for generating relatively easy (if nonsensical) sentences at the lowest levels of complexity. The first 4 levels correspond to what I had intended to cover by certain chapters of the textbook. The 5th level of complexity uses the full baseline grammar, and a somewhat reasonably weighted set of cmavo (so that "pa" is a more frequent numeral than "ci'i"). These can be used by learners of various levels of competence. There are options to show the text parenthesized according to the source grammar, and which perform the simplest removals of elidable terminators (someone might be able to improve that, but I'm not sure it is worth it) The 6th level of complexity is intentionally unnatural to emphasize some of the less common constructions, and all cmavo of a selma'o are equally likely. If someone wants to test a parser (or the PEG grammar) that is the ultimate stress test - every sentence is grammatical in theory, but the RSG is quite capable of producing a 20 line humdinger with nested termsets inside of relative clauses inside of a MEX construction, that I cannot imagine could be translated. I'm not sure it has much other use than as a stress test of a human or computer parser. The last release of the RSG reflected the grammar of 1991, which evolved considerably before the baseline (the cmavo also changed as well). http://www.lojban.org/files/software/rndsen28.zip and you can see some discussion of the outputs on http://www.lojban.org/tiki/tiki-index.php?page=random+sentence+generator&bl though I'm not sure which program was used to create the various examples. lojbab 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.