How can you tell a treasure. We can tell by the reaction. And after that. And after that we are pleased - Gertrude Stein

Home About Automatic poetics Directed poetics Poems on file Source code The Prosthetic Imagination Stats To students The author Contact

To build and run the etc3 project, you will need the source files and the MySql databases that contain the semantic and grammatical models the code uses to build texts. Please note that the source available here is for version 1.0.2 of the software. This Web site is running version 1.0.3, the difference being new structures in the grammar and some minor enhancements to the semantic model as defined in the database. Although structurally small, I recognize that these changes have had an enormous influence on the quality of the generated texts. I will post the newer source after getting some production experience with it. Here's the source: etc3source.zip. Here's the jar file of the compiled source: etc3.jar. Here are the databases: SourcePosting_2007-11-20.zip. You will have to unzip this file and use the MySql restore utility to build your local copy. You will need MySql of course. Get it here: http://www.mysql.com/ You will also need MySql's jdbc driver. Here's the one I've been using: mysql-connector-java-5.0.3-bin.jar There is a kludge in the code that searches the jar to extract class hierarchy information. But it has to know where the jar is, so that information is stored in the DB. To make this work, set the DB field Etc3.EtcConfig.etc3jarpath to the fully qualified path name (including the filename etc3.jar). Use "/" for root, even on Windows, and forward slashes for the folder delimiter (even on Windows). I've left the field I use in the DB as an example. I have some Javadoc files: Etc3javadoc.zip. These contain all of the classes, but the explanations are woefully incomplete. I had planned to finish this documentation before posting the source, but you know how that goes. I am also posting the grammar Xmls: Etc3Grammars.zip and a little utility form (GrammarMaintenance.zip) that facilitates creating and maintaining the grammar-related tables in the database. It uses the Etc3.jar and the MySql jdbc jar (above). There is a HelloWorld.java source file in the etc3.jar (in the etc3 directory) that explains how the code constructs a poem, which should be helpful in getting started. You can find some of the design considerations I used in building this monstrosity on my blog: http://theprostheticimagination.blogspot.com/. Just root around. I am releasing all of this source as public-domain work. The onnly restriction is that you may not copyright it. You may modify it, distribute it in any fashion, use it for commercial purposes, sell it, rent it. Whatever you want. You can claim authorship. You don't even have to mention me. Just don't copyright it. Finally, I can respond to quick questions, but I cannot provide in-depth support. Nor can I provide Java language or MySql instruction. Time just won't allow that.