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: Mon, 13 Aug 90 05:55:00 PDT Reply-To: LaTeX-L Mailing list From: Don Hosek Subject: Re: Some words from the outside To: Rainer Schoepf Status: R X-Status: X-Keywords: X-UID: 217 >here are some comments I got from Cameron Smith : >An example that >was cited to me recently was (La)TeX's making pages one at a time. >It was claimed that the only right way to make pages was to consider >"spreads" (pairs of facing pages), and that a program that irrevocably >made up page 10 before it even knew what would be on page 11 was fundamentally >flawed. I've contemplated this very problem on more than one occasion myself. Making up pages in this fashion would not be too much more complicated than double column setting, the only caveat that I can think of offhand is that the memory requirements will be nearly doubled if pages are complicated (say including a great deal of picture environment graphics) so this sort of setting should be deselectable (but should still be the default for twoside printing; I can't think of any reason to use this approach for simplex setting). As long as we're on this topic, why don't we start dealing with some issues from the bottom up. I'd like to have some idea of what we'll be dealing with in the output routine department by year's end (but then, I'm always the optimist, aren't I?). One other thing I'd like to see: - A mechanism for starting a new page if there is not enough room to typeset X amount of new material on it (\pagetotal et al won't do since they don't take floats into account).