On 19/02/2016 2:39 a.m., Uwe Stöhr wrote:
Original Message
From: Andrew Parsloe
Sent: Donnerstag, 18. Februar 2016 02:50
To: Uwe Stöhr; [email protected]
Subject: Re: Referencing subnumbered equations
It's possible we are talking about different things Uwe. Extracting the
example from 19.3 in the maths manual and compiling it separately, the
numbering in the pdf is (1), (1a), (1b), equation (1) being what you
mean by the outer equation (I assume). In my example documents there is
no outer equation. They are treated equally and so are numbered (1a) and
(1b). And yet I would still like to be able to refer to them
collectively as "the equations (1)". (I can imagine some discussion
like: "The answer to this question is contained in the equations (1). By
substituting from (1b) into (1a), ....")
Yes, I understood you correctly. But the point is still that math.lyx doesn't
try to explain what you do, it only explains the subnumbering.
Besides this I am still wondering why you make it so complicated to reference
1, 1a, 1b. If you need this you can simply use 1, 2, 3. (you don't need
subequations)
Regards Uwe
Yes, you can "simply use 1, 2, 3" but there are many occasions when that
feels wrong. For example, a coordinate transformation, which is a single
entity but involves two (planar) to four (space-time) individual
equations. Maxwell's equations are often listed as a single grouping (or
two groups of two) and deductions from them often involve similar
groupings. Until my "discovery" of labelling the subequation environment
I wasn't aware of a standard method of handling this situation in which
one label references more than one equation, and yet it recurs in books
time and time again. Often the equations making up the group have a
large right brace to show their "solidarity", followed by an equation
number. (I can supply many references, many from eminent figures of 20th
century physics. Some authors, like Feynman in his "Lectures on
Physics", dispense with the right brace, but use a single number,
generally centrally placed, to refer to the group of equations.) One can
implement this with an array between a \left . \right } pair in an
equation environment, and that's what I've done previously -- even
written a personal LaTeX package to make this simpler -- but it puzzled
me that the AMS environments seemed to be silent on this.
What I was looking for was something like the cases environment, but
with the brace on the right, rather than the left. But then I found that
this situation can be handled in the AMS environments by using the
subequations environment. By labelling *this*, rather than the
individual equations, it's possible to refer to "the coordinate
transformation (4)" or "Maxwell's equations (27)" and yet still, if
needed, label and reference the individual equations.
Perhaps this is obvious to others, or perhaps there is some AMS
environment which handles this situation and I've overlooked it. But
discovering that the subequations environment made this possibility
available, and then that I could write a simple layout for it and
incorporate it in a module so that no ERT was needed, was exciting (yes,
such small things excite me) and prompted me to write to you and the
list suggesting that the math manual should mention this possibility. As
I've indicated, it's not hard, looking through my own (somewhat dated)
library of books on physics in particular to find plentiful examples of
this grouping of equations into a "whole" and referencing them as a whole.
Andrew
---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
