M E M O FROM: David Sklar, Director of Applications, EBT TO: All those attempting to publish Rainbow documents with DynaText SUBJECT: Why can't DynaText directly support Rainbow ?? As the essay stored in the file "rainbow.why" explains, Rainbow is not intended for storage, distribution, or publication. Its sole purpose is to be the starting point for conversion to "rich" SGML. In particular, the DynaText system cannot effectively support Rainbow documents. This is because the DynaText stylesheet technology depends upon content markup. The primary stylesheet technique is that you create a formatting style for each tagname. But a Rainbow document has only a very small set of tagnames in it; for example, all of the paragraphs are called PARA. Thus, although it is possible to create a stylesheet for a Rainbow document by using the "property-value functions" feature, it is extremely difficult and certainly is not proper use of DynaText or Rainbow. So how can you convert your Rainbow documents to "good" SGML? You could write a program to do it, using a programmable conversion system like FastTag, OmniMark, etc. Or, you can consider a new product: EBT will be releasing a product called DynaTag that allows conversion of Rainbow to DynaText electronic books; that product operates by converting the Rainbow to "rich" SGML, and auto-creating DynaText stylesheets. DynaTag will be released in September of 94; an alpha version was demoed at Seybold (Mar 94) and at SGML Europe (May 94).