>>>>> "bzg" == Bastien <b...@altern.org> writes:
bzg> Hi Nick, Nick Dokos <nicholas.do...@hp.com> writes:
>> $$...$$ is plain TeX, not LaTeX, and according to Lamport (p.
>> 233), "... does not work properly." See also
>>
>> http://www.tex.ac.uk/cgi-bin/texfaq2html?label=dolldoll
bzg> $$...$$ worked for me for years in LaTeX (using tetex on
bzg> GNU/Linux) so I'm curious: is this FAQ entry still accurate?
bzg> The web page says "2012-03-29" but maybe that's just an
bzg> automated time-stamp, not telling whether this entry in
bzg> particular is still relevant.
It is, but the reasons are not so obvious. You can find some arguments
here :
<http://tex.stackexchange.com/questions/21173/display-math-deadly-sin-in-l2tabu>
In short :
- without amsmath (but who does not load amsmath?), vertical spacing is
wrong in some circumstances (which should never happen anyway : a
displayed equation can never begin a new paragraph), and
- with amsmath, some other problems remain, most notably the global
option "fleqn" (flush left equations) does not work.
In general, $$ should never be modified by any package, so if you load a
package that should do something with displayed equations, you'd better
not use $$ $$. Unfortunately I don't know which packages are in this
situation.
So, is it worth changing your code ? I personally never use $$ (nor \[ \]) (i)
because I like having meaningful environment names, (ii) because I was
taught not to use them, and (iii) because amsmath makes it easy to
change a environement into another (e.g. change between
numbered/unnumbered, or change between {equation} and {align}).
--
Nicolas.
