- From: Curt Arnold <carnold@houston.rr.com>
- Date: Sun, 10 Jun 2001 13:51:16 -0500
- To: <www-dom-ts@w3.org>
> My understanding of what we're doing is that people will not take the > transorm and change it. If they want another transform, they'll probalby > write it. Another ambition we have is to provide as many transforms as > possible (except for the two default ones, Java and ECMA) in order to > increase use. The Apache Batik project has already expressed interest in running DOM tests as part of their development process using their own test framework. The best path for them would be to take the transform that produces Java code for the JUnit framework and make slight modifications to produce code that works with their internal framework. In addition, I see most of the logic and structure of the transform being similar between the various target languages and test frameworks. For example, I would expect that the JavaScript or Python transform would be developed by modifying the Java transform. If someone wanted to port the code generation transform to a unanticipated language, the best approach would be to start with the closest W3C transform and to modify it. Having the code generation transforms under the Software license would allow these types of uses. Having the code under the Document license would appear to require a clean-room implementation of the transform which could result in people just not using the tests due to potential legal liability.
Received on Sunday, 10 June 2001 14:49:55 UTC