By the way, David, my markup already includes page numbers, sidenotes and math via latex. Of course, since it's only conceptual markup it's up to the exporter and the display device to instantiate the realization. But that exactly is the way eb.tbicl.org stores, transforms and displays them now. Look up the fourier article I referenced (should be eb.tbicl.org/fouriers-series ( think) earlier for an extensive latex example. All the articles that had page numbers marked somehow in the PG texts have been transformed upon loading to this msrkup and are functional. Same for sidenotes. It's just a stylesheet change, as you would expect, to have offset or inset page numbers. There's a lot of unsimplified html still left from the PG html files. Tables are pretty much untouched. Even most of the illustrations are not easily converted yet. But what's used for the article sources is substantially simpler yet more malleable than what PG provides, given software that understands the markup - none of which exists other than what I've got in the app. On Fri, Feb 3, 2012 at 6:08 PM, don kretz <dakretz@gmail.com> wrote:
Hi keith,
Of couse xml can handle a book if you constrain your definition of your book to only include things that can be described by xml. What if I choose to include footnotes that extend across multiple pages, including both where they are referenced in the text flow, and on what physical pages they are found? You can't just declare the possibility to be invalid because it can't be described in a strict hierarchy.
On Fri, Feb 3, 2012 at 4:22 PM, Keith J. Schultz <schultzk@uni-trier.de>wrote:
Hi Don,
I very sorry , but are completely wrong. XML is perfectly well of handling the structures of book. The problem is how you handle and define the structure which is not predefined by the XML standard.
As far as the output is concerned that that is a matter of parsing the tree and reacting to the semantics for the entities defined therein.
Please do not forget that a paragraph is an entity of a text, where as a page is an entity of a book!
On another side. HTML actually has no concept of a page, except maybe the rendering of the file itself. Meaning that i would have to have a file per page. Then you have the same problem. Naturally, that is not how it is done. You can simulate the same semantics for page breaks in XML as in HTML, because the page break is just simulated.
regards Keith.