Single letter in subscript
Leo
I am strugling with forming a general guideline on subscripts in math
mode. For example:
$\alpha_\mathrm{H}$ or $\alpha_H$ if H is the first letter of the
author (Henry) who invented this.
$m_\mathrm{e}$ or $m_e$ for mass of an electron
In my field, lots of papers have mixed these styles and they seem to
know what they are doing, ie it does not appear to be randomly
chosen. So I am wondering if such guideline exists on formatting
single letter subscripts. Thanks.
Leo
Philipp Stephani
Guidelines and ISO standards state that subscripts must generally be
printed upright unless the subscript is a variable. See for example
http://www.iupac.org/standing/idcns/italic-roman_dec99.pdf
--
Change “LookInSig” to “tcalveu” to answer by mail.
Leo
[...]
> Guidelines and ISO standards state that subscripts must generally be
> printed upright unless the subscript is a variable. See for example
> http://www.iupac.org/standing/idcns/italic-roman_dec99.pdf
Thanks. That's very helpful.
In terms of getting a roman font, which command to use \mathrm or
\textrm? I use \mathrm for the time being but this FAQ entry[1] seems to
suggest otherwise.
Thanks,
Leo
Footnotes:
[1] http://www.tex.ac.uk/cgi-bin/texfaq2html?label=mathstext
Robin Fairbairns
Leo <sdl...@gmail.com> writes:
>Philipp Stephani <Look...@arcor.de> writes:
>[...]
>> Guidelines and ISO standards state that subscripts must generally be
>> printed upright unless the subscript is a variable. See for example
>> http://www.iupac.org/standing/idcns/italic-roman_dec99.pdf
>
>Thanks. That's very helpful.
>
>In terms of getting a roman font, which command to use \mathrm or
>\textrm? I use \mathrm for the time being but this FAQ entry[1] seems to
>suggest otherwise.
>
>[1] http://www.tex.ac.uk/cgi-bin/texfaq2html?label=mathstext
i'm not sure it does suggest otherwise. it is, however, trying to
cope with any number of different means of doing several different
things.
certainly i would use \mathrm in this circumstance (the faq answer
notes the difficulty of \mathrm playing with spaces, but there aren't
any spaces in this case, and indeed it mentions "upright roman
alphabetic variable names" (which i find difficult to swallow) which
seems to cover what you have.
("upright alphabetic symbols" might be better -- wdyt?)
i have had a lot of difficulty over the years, with this answer. (i
did, at one stage, misunderstand the best recommended techniques,
even: the trouble is that people who _know_ tend not to be reading the
faq, and therefore tend not to put me right.)
--
Robin Fairbairns, Cambridge
Enrico Gregorio
Unfortunately, each method (\mathrm, \textrm or \text) has its
drawbacks. With \mathrm spaces and punctuation signs tend to
be regarded in peculiar ways, while \textrm and \text use the
current text font, which might be wrong (italics in a theorem
statement, for example).
The best thing is \textup (with amsmath).
Ciao
Enrico
Philipp Stephani
The problem is that there are so many possibilities. First of all, one
can choose between math and text mode, with vastly different results
(not only spaces, also things like dashes). Then the \text.. commands
differ in which font features they inherit from the surrounding text.
So the difference between \text and \textrm is not just that \text is
easier to type, as stated in the FAQ. Lars Madsen often points out that
\text inherits the font shape, which can lead to confusion because
italic and upright characters have different semantics in mathematics;
he proposes \textup instead. One could also argue that \textnormal,
which switches to text mode but doesn't inherit any of the font
features, should be used, or \textnormalstyle, a command that inherits
the family, but nothing else. Not to mention the unlimited number of
combinations possible with XeLaTeX. I think these points should be
discussed in the FAQ. Here is an example with seven commands, giving
seven different results:
\documentclass{article}
\usepackage{lmodern}
\usepackage{amsmath}
\DeclareTextFontCommand{\textnormal}{\normalfont}
\DeclareTextFontCommand{\textnormalstyle}{\mdseries\upshape}
\begin{document}
\Large\sffamily\bfseries\slshape
$a_{\mbox{a-a a}} ~ a_{\mathrm{a-a a}} ~ a_{\text{a-a a}} ~
a_{\textup{a-a a}} ~ a_{\textmd{a-a a}} ~ a_{\textrm{a-a a}} ~
a_{\textnormal{a-a a}} ~ a_{\textnormalstyle{a-a a}}$
\end{document}
Leo
[...]
> i'm not sure it does suggest otherwise. it is, however, trying to
> cope with any number of different means of doing several different
> things.
>
> certainly i would use \mathrm in this circumstance (the faq answer
> notes the difficulty of \mathrm playing with spaces, but there aren't
> any spaces in this case, and indeed it mentions "upright roman
> alphabetic variable names" (which i find difficult to swallow) which
> seems to cover what you have.
>
> ("upright alphabetic symbols" might be better -- wdyt?)
>
> i have had a lot of difficulty over the years, with this answer. (i
> did, at one stage, misunderstand the best recommended techniques,
> even: the trouble is that people who _know_ tend not to be reading the
> faq, and therefore tend not to put me right.)
Enrico Gregorio <greg...@math.unipd.it> writes:
[...]
> Unfortunately, each method (\mathrm, \textrm or \text) has its
> drawbacks. With \mathrm spaces and punctuation signs tend to
> be regarded in peculiar ways, while \textrm and \text use the
> current text font, which might be wrong (italics in a theorem
> statement, for example).
>
> The best thing is \textup (with amsmath).
>
> Ciao
> Enrico
Thanks to both of you for the fine answers.
Since there isn't any space or punctuation in my case, I am going to
use \mathrm which is readily bound to a key in auctex.
Best,
Leo
zugzwang
> > http://www.tex.ac.uk/cgi-bin/texfaq2html?label=mathstext
...
> The problem is that there are so many possibilities.
Somewhat off topic, there are two other possibilities that the faq
might mention:
1) From the relsize package, something like
{\relsize{-2}\raisebox{-.6\height}{mytext}\relsize{2}}%
as an alternative to the faq's $z = a_{\mbox{\scriptsize other end}}$
2) \textsubscript
Lars Madsen
I'd use \textup to switch to the text font, and enable various chars
that are not available in math
\mathrm and \textrm is no good if your are working in a sans serif context.
/daleif
Guenter Milde
Here is a selection of publicly available guides (from the package
isomath.sty):
References
**********
.. [ISO-31] International Standard 31 `Quantities and units`,
(ISO 31-0:1992 to ISO 31-14:1992):
http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_tc_browse.htm?commid=46202
.. [typefaces] `Typefaces for Symbols in Scientific Manuscripts`:
http://physics.nist.gov/Document/typefaces.pdf
.. [checklist] `SI Unit rules and style conventions`
Check List for Reviewing Manuscripts:
http://physics.nist.gov/cuu/Units/checklist.html
.. [fonts-for-symbols] `On the use of italic and roman fonts for symbols
in scientific text`, (Revised December 1999):
http://old.iupac.org/standing/idcns/fonts_for_symbols.html
.. [SI-brochure] `The International System of Units (SI)`:
http://www.bipm.org/en/si/si_brochure/
Günter
Leo
> References
> **********
>
> .. [ISO-31] International Standard 31 `Quantities and units`,
> (ISO 31-0:1992 to ISO 31-14:1992):
> http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_tc_browse.htm?commid=46202
>
> .. [typefaces] `Typefaces for Symbols in Scientific Manuscripts`:
> http://physics.nist.gov/Document/typefaces.pdf
>
> .. [checklist] `SI Unit rules and style conventions`
> Check List for Reviewing Manuscripts:
> http://physics.nist.gov/cuu/Units/checklist.html
>
> .. [fonts-for-symbols] `On the use of italic and roman fonts for symbols
> in scientific text`, (Revised December 1999):
> http://old.iupac.org/standing/idcns/fonts_for_symbols.html
>
> .. [SI-brochure] `The International System of Units (SI)`:
> http://www.bipm.org/en/si/si_brochure/
>
>
> G�nter
Thanks.
Leo