X-VM-v5-Data: ([nil nil nil nil nil nil nil t nil] [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil]) Date: Tue, 20 Mar 90 20:18:11 CET Reply-To: LaTeX-L Mailing list Sender: LaTeX-L Mailing list From: Leslie Lamport Subject: Re: theorem considerations To: Rainer Schoepf In-Reply-To: Your message of Tue, 20 Mar 90 18:06:33 CET. <9003201712.AA12747@decpa.pa.dec.com> Status: R X-Status: X-Keywords: X-UID: 56 The comments by Mike Downes, forwarded by Barbara Beeton, display a lack of understanding of the distinction between form and logical structure. Perhaps the clearest illustration is the discussion of how to produce "7.3 Theorem". Since the "7.3" looks like a section number, the absurd construction \section{}\begin{theorem} is proposed. Mr. Downes apparently believes that all things labeled "Theorem" are the same, and should be produced by the same LaTeX environment. What he fails to realize is that the input should reflect the logical structure of the document. The standard LaTeX document styles assume that theorems appear as isolated units, separate from the section/subsection hierarchy. If authors choose to structure their documents differently, then that structure should be reflected in the input. The standard document styles were meant to handle the structure of most documents, not to be a mold into which all documents must fit. The first question a document-style designer must ask is: What is the logical structure? Trying to hide differences in logical structure with more "flexible" macros will produce a system that combines the power of WYSIWYG with the ease-of-use of Plain TeX. Leslie Lamport