This is the mail archive of the davenport@berkshire.net mailing list for the Davenport project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

DAVENPORT: Re: Maintaing documents for multiple languages


Norman Walsh <ndw@nwalsh.com> writes:

|   Since I don't think I made those arguments on this list, let me
|   recapitulate two of the disadvantages of marked sections:
|   
|   1. They aren't valid in XML.

"marked sections" are valid SGML.  If you depend on XML tools, there are
esay ways to translate your text into XML.

|   2. After parsing, the IGNOREd sections are *gone*. The parser throws
|      them away. This means that you can never write a tool that uses
|      a standard parser that can check for the 1-to-1 correspondence of
|      chunks or do anything with both languages.

I don't see the point.  If you say "IGNORE" you'll mean "IGNORE", I
guess.  Process your document twice (e.g., with sgmlnorm); than you'll
have two SGML/XML documents (one in english and one in french) and you
can run queries on them -- I don't know whether those "next generation"
tools are available; but since SGML/XML databases are already announce,
I'd think the needed tools are there.

To make this work, you've to think about needed attributes carefully.

-- 
                                             work   :        ke@suse.de
Karl Eichwalder                              private: ke@gnu.franken.de


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]