So, does this mean that I now not only have to download the master xml file, the css, and a set of conversion tools? You must be kidding, right? If it came to that, I would rather have the plain text and forget the page numbers. It is already inconvenient to use "lynx -dump -nolist filename.htm." Why in the world would I want to run it through a conversion tool and still have to do that anyway? OK, so a plain text file can be output directly from the xml. I still have to go through at least one extra conversion step that I wouldn't have to otherwise.
Why? The whole idea behind PG moving to XML is not to complicate things, it's to give more flexibility while retaining simplicity. How about this situation: PG files are, by default, coded in XML. All other formats are then automatically generated from that XML format. There would still be TXT versions, there would still be HTML versions. Getting those would be no harder than it is for you to retrieve a TXT file now. All this conversion stuff should be done by the PG back-end, not the end-user (why make a human do a machine's job?). That way, instead of manually preparing every different format like what goes on at PG for the most part now, we could make every format available with only the effort of creating a super-format from which every other format could be derived and a set of tools which could automatically generate other formats from the super-format. If someone wants the entire PG library to be available in some obscure format, then it could be if they can just write a converter which outputs that format. Cheers, Holden