I also support the position that most Tkx users will be coming from Perl/Tk, which is well documented through books like Mastering PerlTk. So, the first problem we get when trying to use Tkx is the limited documentation. Basically, we have to read between the lines to figure out how to do things. When I first started using Tkx I was trying to figure out the translation from Perl/Tk but I've been increasingly disappointed that the translation is not that simple. These days, I'm reading the Tcl documentation and trying to figure out how to translate that into Tkx syntax. Surprisingly, this isn't always easy either.
I think redundancy is the key to documentation. Even if Tkx::eval() is just like any other Tkx::foo if it's useful, then it would be nice to have in the documentation, even if just as an example of how powerful Tkx::foo can be. We also need to generate a strong Tkx community (to cover cases not in the documentation). I think the fact that there's isn't a proper Tkx discussion group is a serious problem. Why isn't one created? A first good step would be to at least explain in the documentation where people should post questions about Tkx. (Is the tcltk mailing list the right place?) Just my thoughts... Laurence.
