Talk:LaTeX formatting tips and tricks

Um .... "condensed intertext" is one of my pet hates! I don't think it's a good idea to employ these sort of ways to preserve the exact look of the page (which may have been printer stinginess rather than author intent). As a PPer I find these spacing tricks really annoying and change them all back to ordinary. Laurawisewell 11:35, 20 July 2006 (PDT)

As a reader I find that the sea of white space intoduced by putting short bits of text on a separate line in the middle of a display interferes with the flow of the argument. Words which are just there to make the mathematics read grammatically (things like "hence", "so", "therefore", "and" etc) can coexist with the next clause of the display, kind of like equation tags do. (Obviously I think equation numbers belong on the right ;-) Anything which is part of the argument/explanation, or longer than a few words, deserves its own line. I am sufficiently ancient to have had the joy of getting galley proofs of papers to correct prior to publication, and seeing the typescript with its colour-coding for italics, bold, Greek, etc transformed into "real" mathematics. In my experience copy editors rarely make changes to the layout of displays, so what is printed usually corresponds with what the author wanted.

I'm not totally enamoured of the flalign approach because it feels like an abuse of the structure, especially when you need a \phantom to get it to work; this doesn't mean that as a PPer I won't retain a working flalign lovingly crafted in the F rounds, but sometimes I have resorted to using a plain TeX halign-based construction instead. Fortunately PG is a broad church. Dcwilson 03:23, 21 July 2006 (PDT)

I too have had the joy of correcting galley proofs, and have had my displaying/inlining altered. But I really don't think the behaviour of today's copy editors is any guide to the past. Whatever. Semantically, these are intertexts, not pieces of an equation. If you really want to keep the layout of the book, maybe you should redefine the intertext command to do what you want? Laurawisewell 10:00, 21 July 2006 (PDT)

I don't understand why the semantic distinction between "intertexts" and "pieces of an equation" implies that they must be vertically distinguished. The choice between putting the text on the same line or on its own line is not one that would be easily automated, so re-implementing \intertext is just too daunting, especially when there are ways to achieve the desired effect using existing code. "Whatever." Perhaps we should abandon all attempts to preserve idiosyncrasies of the books and make them uniformly awful with default LaTeX for everything... Dcwilson 17:13, 23 July 2006 (PDT)

It doesn't mean they should be vertically distinguished. It does mean that you shouldn't be putting your text inside an equation environment and then having to switch back to text mode. Like you said above, "it feels like an abuse of the structure". As I'm sure you know, I wouldn't have a clue how to reimplement \intertext. I like to keep things simple. If that makes my PP projects awful, well that's your opinion. I do preserve interesting idiosyncrasies (weird notation for intervals, upside-down \leq's in Infinitesimal analysis for example) but in my view, spacing tricks are unnecessary. Laurawisewell 01:14, 24 July 2006 (PDT)

The mathtools package provides \shortintertext, which is way better than \intertext. Dfeuer 14:23, 28 December 2006 (PST)