Open University, UK
Abstract: This presentation will investigate some issues related to the evolution of TEX's standard output format for future extensions of TEX.
Many of my thoughts about such a future standard output format are informed by the needs of documents that are designed (to high standards) for multiple uses, in particular for reading both on screen and in paper form.
Many people now have experience of using TEX as the typesetting engine for such documents where the multi-use output form is a `PDF document'--PDF here refers to the Adobe's Portable Document Format.
Although PDF is technically not a Device-Independent language, it does contain a large core of stuff that is, at least potentially, as `device independent' as TEX's DVI language. Both must, of course, be parsed by an application that understands the language and its underlying document- and page-models; and although they look very different at the detailed level, their page-models and abstract semantics also have a lot in common. This is why the part of pdfTEX that handles classic TEXfiles is only very locally and minimally different from classic TEX.
However, PDF has a far richer document model, and hence pdfTEX has a number of extra primitives. Many of these (such as those which support hyper-links and pop-ups) clearly implement particular abstractions that are (as seen by modern practitioners) essential parts of on-screen documents, whilst others (such as thumb-nails) are peculiar to the somewhat old-fashioned and arbitrary interface to document structure chosen by the designers of various Adobe products. There probably also exist necessary extensions to the currently used models that are not supported well by either language.
By August 1999 I hope, with a bit of help from my friends, to have further analysed the models and concepts that need to be supported for multi-use documents and how well PDF does this. I am confident that this will prove useful in considering in more detail the necessary extensions of TEX's DVI standard.