[sword-devel] Making Import Easier

[Daniel Owens]

Since this thread began (morning in SE Asia, now evening) there has been a bewildering array of discussion, some of it very interesting. Forgive me if I don't respond to specific emails ...

The idea of suggested or automatically created conf files is excellent (thanks for that in tei2mod--it was a welcome sight the first time I ran it). If a conf file is automatically generated, it would be great if the files and folders expected by the conf file could be put in the SWORD path automatically or created in some way that makes it easy to put them in the right place (such as in a zip archive identical in structure to those one downloads and installs). Certainly a web-based interface that turns an uploaded xml file into a compiled module sounds GREAT to me. 

Adding module creation to each frontend sounds like more work than it's worth--better for engine and module developers to spend more time on support for OSIS and TEI than have to worry about how to make it easy for module creators to prepare content. 

On the other hand, a cross-platform offline GUI to create OSIS and TEI xml files would be fantastic, especially if one could set up a configuration file to find the right biblical book abbreviations and set up the right OSIS markup for Scripture references. I just wonder what kind of user-created content is likely to be common. It seems that with Bibles a WYSIWYG interface doesn't make sense (the texts are too long, and proper versification is too important), but with genbooks (and possibly commentaries), it would make more sense. Could those with better xsl skills than I just improve the OpenOffice export filter and create an import filter too? I mean, why not capitalize on an already-existing program?

Earlier this month I had hoped to be able to send out an announcement about a content creation effort I am involved I, but we weren't quite ready. We are trying to create content (first a modern replacement of Strongs) that is released under a creative commons license to encourage distribution and translation. Check it out at www.textonline.org. In our effort we plan to use a wiki and MDF field markers (similar to usfm), and we will manage the conversion of content to pdf or SWORD, freeing up content creators to focus on content. It strikes me that this sort of approach is a more realistic way to provide opportunities for non-techies to contribute content, though I expect that even our "simple" approach may meet with some resistance from non-techies. 

For what it's worth,
Daniel