Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
upright Greek letters for constants as opposed to variables
141 views
Skip to first unread message
Phillip Helbig (undress to reply)
Apr 26, 2020, 5:23:54 AM4/26/20
to
Some journals require upright Greek letters for constants as opposed to=20
variables, for example $\upi$ when used to denote 3.14159... as opposed=20
to a variable ($\pi$ is sometimes used to denote parallax in astronomy,=20
for instance). (Some journals define \upi as "upright pi", \upi as=20
"upright i", and so on.)
I certainly agree that LABELS should be upright (though they are usually
Latin not Greek) and not italic to distinguish them from variables, e.g.
$T_{mathrm{eff}}$ for effective temperature or $\rho_{textrm{g}}$ for
gas density, say, as opposed to $G_{\mu\nu}$ where $\mu$ and $\nu$ are
not constants but variables.=20
And it is not just Greek letters. For example, e for the Euler number=20
or i for the square root of -1 should also not be in math italic, to=20
distinguish them from variables. I tend to agree with that as well. =20
Also, units should be upright, e.g. 5 m and not $5m$ for 5 metres.
On the other hand, I have never seen the gravitational constant $G$,=20
which is even by definition a constant and not a variable, written=20
upright. Ditto for the Hubble constant $H$ and so on.=20
Or is there a difference between mathematical constants and physical=20
constants?
Perhaps because standard (La)TeX provides Greek letters only in math=20
italic, upright Greek letters are less common than upright Latin=20
letters, even when used in the same way (labels, units, symbols which=20
are not variables).
When writing for a specific journal, one usually has to follow the house=20
style. However, if there is no rule, I prefer to do what is generally=20
deemed to be correct. What is generally deemed to be correct here? =20
variables, for example $\upi$ when used to denote 3.14159... as opposed=20
to a variable ($\pi$ is sometimes used to denote parallax in astronomy,=20
for instance). (Some journals define \upi as "upright pi", \upi as=20
"upright i", and so on.)
I certainly agree that LABELS should be upright (though they are usually
Latin not Greek) and not italic to distinguish them from variables, e.g.
$T_{mathrm{eff}}$ for effective temperature or $\rho_{textrm{g}}$ for
gas density, say, as opposed to $G_{\mu\nu}$ where $\mu$ and $\nu$ are
not constants but variables.=20
And it is not just Greek letters. For example, e for the Euler number=20
or i for the square root of -1 should also not be in math italic, to=20
distinguish them from variables. I tend to agree with that as well. =20
Also, units should be upright, e.g. 5 m and not $5m$ for 5 metres.
On the other hand, I have never seen the gravitational constant $G$,=20
which is even by definition a constant and not a variable, written=20
upright. Ditto for the Hubble constant $H$ and so on.=20
Or is there a difference between mathematical constants and physical=20
constants?
Perhaps because standard (La)TeX provides Greek letters only in math=20
italic, upright Greek letters are less common than upright Latin=20
letters, even when used in the same way (labels, units, symbols which=20
are not variables).
When writing for a specific journal, one usually has to follow the house=20
style. However, if there is no rule, I prefer to do what is generally=20
deemed to be correct. What is generally deemed to be correct here? =20
Hendrik van Hees
Apr 26, 2020, 5:39:59 AM4/26/20
to
These are great journals since they care for century-old good practice
in microtypography.
The rules are that mathematical constants like i, e, pi etc. are set
upright. The same holds for operators like the differential d. All units
have to be set upright. Physical constants (even those which are
exactly defined in the new SI like c, h, \hbar, etc.) however have to be
set in italics. The Newton gravitational constant is also to be set in
italics (it's not even a fixed value in the SI for understandable
reasons).
All variable quantities are set in italics. The only exception are
capital greek letters (in the German and the Anglo-Saxxon tradition; the
French also set these letters in italics). The same holds for indices,
which are variable (e.g., running over 1 to 3 in 3D in Euclidean or from
0 to 3 in Minkowski/GR vector calculus). In contradistinction to that if
the index is a descriptor like $k_{\text{B}}$, where the B stands for
Boltzmann and is not variable, it should be set upright.
In LaTeX there are extra packages for a full upright Greek math
alphabeth (it's also included in the exceptionally good font package
mathdesign, where you have the choice full math alphabeths in very nice
fonts like Garamond, Utopia, Charter).
--
Hendrik van Hees
Goethe University (Institute for Theoretical Physics)
D-60438 Frankfurt am Main
http://fias.uni-frankfurt.de/~hees/
in microtypography.
The rules are that mathematical constants like i, e, pi etc. are set
upright. The same holds for operators like the differential d. All units
have to be set upright. Physical constants (even those which are
exactly defined in the new SI like c, h, \hbar, etc.) however have to be
set in italics. The Newton gravitational constant is also to be set in
italics (it's not even a fixed value in the SI for understandable
reasons).
All variable quantities are set in italics. The only exception are
capital greek letters (in the German and the Anglo-Saxxon tradition; the
French also set these letters in italics). The same holds for indices,
which are variable (e.g., running over 1 to 3 in 3D in Euclidean or from
0 to 3 in Minkowski/GR vector calculus). In contradistinction to that if
the index is a descriptor like $k_{\text{B}}$, where the B stands for
Boltzmann and is not variable, it should be set upright.
In LaTeX there are extra packages for a full upright Greek math
alphabeth (it's also included in the exceptionally good font package
mathdesign, where you have the choice full math alphabeths in very nice
fonts like Garamond, Utopia, Charter).
Hendrik van Hees
Goethe University (Institute for Theoretical Physics)
D-60438 Frankfurt am Main
http://fias.uni-frankfurt.de/~hees/
Phillip Helbig (undress to reply)
Apr 26, 2020, 6:07:06 PM4/26/20
to
In article <hgl318...@mid.dfncis.de>, Hendrik van Hees
do.
> The Newton gravitational constant is also to be set in
> italics (it's not even a fixed value in the SI for understandable
> reasons).
True, but it is certainly not a variable in the mathematical sense.
> All variable quantities are set in italics.
My question is really why mathematical, but not physical, constants
should be upright. In maths is clear: italic means variable, upright
means constant or label or operator or unit. In physics, italic means
variable OR constant, upright means label or operator or unit.
> The only exception are
> capital greek letters (in the German and the Anglo-Saxxon tradition; the
> French also set these letters in italics).
What is a common example of a capital Greek letter being used as a
variable? Perhaps the Gamma factor in special relativity? (Zel'dovich
used Delta = 1 - 1/(1+z) for the cosmological redshift, but that has
gone out of fashion.)
> The same holds for indices,
> which are variable (e.g., running over 1 to 3 in 3D in Euclidean or from
> 0 to 3 in Minkowski/GR vector calculus). In contradistinction to that if
> the index is a descriptor like $k_{\text{B}}$, where the B stands for
> Boltzmann and is not variable, it should be set upright.
Right; it is a label.
> In LaTeX there are extra packages for a full upright Greek math
> alphabeth (it's also included in the exceptionally good font package
> mathdesign, where you have the choice full math alphabeths in very nice
> fonts like Garamond, Utopia, Charter).
Yes, I've used some of those. At least in some cases, the upright
letters have new commands, since the usual ones are used for the italic
characters. This makes it a pain when recycling something like a
complicated formula. In general, I try to have all differences between
journals which are not handled by the corresponding LaTeX class handled
by my own macros. So first I load my own macros, which are my
defaults, then journal-specific macro packages which redefine things
which differ from my defaults. (That applies to many things, such as
spelling, punctuation, italicization (or not) of foreign expressions,
forms of dashes, and so on. Using \newcommand, \renewcommand, and
\providecommand gives the necessary flexibility (e.g. one can define
something (sensibly, with the same definition) if it hasn't been defined
by the class, otherwise use the definition from the class).
So basically, my question is what should the defaults be, but it seems
that you agree with the great journals, so I'll go that route.
But what is the origin of the difference between mathematical and
physical constants and for the exception to set capital Greek variables
upright (but not in France)?
Of course, $W$ is the variable for energy, W the unit watt, $m$ the
variable for mass, m the unit meter, and so on. Could that be the
reason?
What about alpha, beta, and gamma rays? Should probably be upright?
<he...@fias.uni-frankfurt.de> writes:
> These are great journals since they care for century-old good practice
> in microtypography.
>
> The rules are that mathematical constants like i, e, pi etc. are set
> upright. The same holds for operators like the differential d. All units
> have to be set upright. Physical constants (even those which are
> exactly defined in the new SI like c, h, \hbar, etc.) however have to be
> set in italics.
Right; that seems to be what the journals who care about those things
> These are great journals since they care for century-old good practice
> in microtypography.
>
> The rules are that mathematical constants like i, e, pi etc. are set
> upright. The same holds for operators like the differential d. All units
> have to be set upright. Physical constants (even those which are
> exactly defined in the new SI like c, h, \hbar, etc.) however have to be
> set in italics.
do.
> The Newton gravitational constant is also to be set in
> italics (it's not even a fixed value in the SI for understandable
> reasons).
> All variable quantities are set in italics.
should be upright. In maths is clear: italic means variable, upright
means constant or label or operator or unit. In physics, italic means
variable OR constant, upright means label or operator or unit.
> The only exception are
> capital greek letters (in the German and the Anglo-Saxxon tradition; the
> French also set these letters in italics).
variable? Perhaps the Gamma factor in special relativity? (Zel'dovich
used Delta = 1 - 1/(1+z) for the cosmological redshift, but that has
gone out of fashion.)
> The same holds for indices,
> which are variable (e.g., running over 1 to 3 in 3D in Euclidean or from
> 0 to 3 in Minkowski/GR vector calculus). In contradistinction to that if
> the index is a descriptor like $k_{\text{B}}$, where the B stands for
> Boltzmann and is not variable, it should be set upright.
> In LaTeX there are extra packages for a full upright Greek math
> alphabeth (it's also included in the exceptionally good font package
> mathdesign, where you have the choice full math alphabeths in very nice
> fonts like Garamond, Utopia, Charter).
letters have new commands, since the usual ones are used for the italic
characters. This makes it a pain when recycling something like a
complicated formula. In general, I try to have all differences between
journals which are not handled by the corresponding LaTeX class handled
by my own macros. So first I load my own macros, which are my
defaults, then journal-specific macro packages which redefine things
which differ from my defaults. (That applies to many things, such as
spelling, punctuation, italicization (or not) of foreign expressions,
forms of dashes, and so on. Using \newcommand, \renewcommand, and
\providecommand gives the necessary flexibility (e.g. one can define
something (sensibly, with the same definition) if it hasn't been defined
by the class, otherwise use the definition from the class).
So basically, my question is what should the defaults be, but it seems
that you agree with the great journals, so I'll go that route.
But what is the origin of the difference between mathematical and
physical constants and for the exception to set capital Greek variables
upright (but not in France)?
Of course, $W$ is the variable for energy, W the unit watt, $m$ the
variable for mass, m the unit meter, and so on. Could that be the
reason?
What about alpha, beta, and gamma rays? Should probably be upright?
Hendrik van Hees
May 2, 2020, 5:38:05 PM5/2/20
to
Of course, after all these are conventions and often just a tradition.
Why the physical fundamental constants are in italics and not upright
is something I'm also puzzled about, but on the other hand, while
mathematical constants are really constant and what they are, the
values of the fundamental constants depend on the choice of the
units. E.g., with the official change to the new SI the magnetic
field constant became a derived quantity with an error, while before
it was defined as an exact quantity by the old definition of the
Ampere (two infinite straight wires...). Now it's to be measured
based on the fixed value of the elementary charge and the definition
of the second in terms of the Cesium hyperfine splitting frequency.
Maybe that's the reason, why these fundamental constants are set
in italics.
Also the fixed names of elementary particles, hadrons, nucleons,
etc. are set upright (roman). That's, because they are naming fixed
specific entities.
Last but not least, not all journals follow these rules. Sometimes
they even don't like to set the differential d upright. Sometimes
you find hints about their conventions in the LaTeX templates/class
files of the journal.
If you use Word, it's impossible to get a good microtypography
anyway. There you can only hope that the journal transforms it
into a readable text ;-)).
On 27/04/2020 00:07, Phillip Helbig (undress to reply) wrote:
> In article <hgl318...@mid.dfncis.de>, Hendrik van Hees
> <he...@fias.uni-frankfurt.de> writes:
>
Why the physical fundamental constants are in italics and not upright
is something I'm also puzzled about, but on the other hand, while
mathematical constants are really constant and what they are, the
values of the fundamental constants depend on the choice of the
units. E.g., with the official change to the new SI the magnetic
field constant became a derived quantity with an error, while before
it was defined as an exact quantity by the old definition of the
Ampere (two infinite straight wires...). Now it's to be measured
based on the fixed value of the elementary charge and the definition
of the second in terms of the Cesium hyperfine splitting frequency.
Maybe that's the reason, why these fundamental constants are set
in italics.
Also the fixed names of elementary particles, hadrons, nucleons,
etc. are set upright (roman). That's, because they are naming fixed
specific entities.
Last but not least, not all journals follow these rules. Sometimes
they even don't like to set the differential d upright. Sometimes
you find hints about their conventions in the LaTeX templates/class
files of the journal.
If you use Word, it's impossible to get a good microtypography
anyway. There you can only hope that the journal transforms it
into a readable text ;-)).
On 27/04/2020 00:07, Phillip Helbig (undress to reply) wrote:
> In article <hgl318...@mid.dfncis.de>, Hendrik van Hees
> <he...@fias.uni-frankfurt.de> writes:
>
>> These are great journals since they care for century-old good practice
>> in microtypography.
>>
>> The rules are that mathematical constants like i, e, pi etc. are set
>> upright. The same holds for operators like the differential d. All units
>> have to be set upright. Physical constants (even those which are
>> exactly defined in the new SI like c, h, \hbar, etc.) however have to be
>> set in italics.
>> in microtypography.
>>
>> The rules are that mathematical constants like i, e, pi etc. are set
>> upright. The same holds for operators like the differential d. All units
>> have to be set upright. Physical constants (even those which are
>> exactly defined in the new SI like c, h, \hbar, etc.) however have to be
>> set in italics.
> Right; that seems to be what the journals who care about those things
> do.
>
> do.
>
>> The Newton gravitational constant is also to be set in
>> italics (it's not even a fixed value in the SI for understandable
>> reasons).
>> italics (it's not even a fixed value in the SI for understandable
>> reasons).
> True, but it is certainly not a variable in the mathematical sense.
>
>
>> All variable quantities are set in italics.
> My question is really why mathematical, but not physical, constants
> should be upright. In maths is clear: italic means variable, upright
> means constant or label or operator or unit. In physics, italic means
> variable OR constant, upright means label or operator or unit.
>
> should be upright. In maths is clear: italic means variable, upright
> means constant or label or operator or unit. In physics, italic means
> variable OR constant, upright means label or operator or unit.
>
>> The only exception are
>> capital greek letters (in the German and the Anglo-Saxxon tradition; the
>> French also set these letters in italics).
>> capital greek letters (in the German and the Anglo-Saxxon tradition; the
>> French also set these letters in italics).
> What is a common example of a capital Greek letter being used as a
> variable? Perhaps the Gamma factor in special relativity? (Zel'dovich
> used Delta = 1 - 1/(1+z) for the cosmological redshift, but that has
> gone out of fashion.)
>
> variable? Perhaps the Gamma factor in special relativity? (Zel'dovich
> used Delta = 1 - 1/(1+z) for the cosmological redshift, but that has
> gone out of fashion.)
>
>> The same holds for indices,
>> which are variable (e.g., running over 1 to 3 in 3D in Euclidean or from
>> 0 to 3 in Minkowski/GR vector calculus). In contradistinction to that if
>> the index is a descriptor like $k_{\text{B}}$, where the B stands for
>> Boltzmann and is not variable, it should be set upright.
>> which are variable (e.g., running over 1 to 3 in 3D in Euclidean or from
>> 0 to 3 in Minkowski/GR vector calculus). In contradistinction to that if
>> the index is a descriptor like $k_{\text{B}}$, where the B stands for
>> Boltzmann and is not variable, it should be set upright.
> Right; it is a label.
>
>
>> In LaTeX there are extra packages for a full upright Greek math
>> alphabeth (it's also included in the exceptionally good font package
>> mathdesign, where you have the choice full math alphabeths in very nice
>> fonts like Garamond, Utopia, Charter).
>> alphabeth (it's also included in the exceptionally good font package
>> mathdesign, where you have the choice full math alphabeths in very nice
>> fonts like Garamond, Utopia, Charter).
> Yes, I've used some of those. At least in some cases, the upright
> letters have new commands, since the usual ones are used for the italic
> characters. This makes it a pain when recycling something like a
> complicated formula. In general, I try to have all differences between
> journals which are not handled by the corresponding LaTeX class handled
> by my own macros. So first I load my own macros, which are my
> defaults, then journal-specific macro packages which redefine things
> which differ from my defaults. (That applies to many things, such as
> spelling, punctuation, italicization (or not) of foreign expressions,
> forms of dashes, and so on. Using \newcommand, \renewcommand, and
> \providecommand gives the necessary flexibility (e.g. one can define
> something (sensibly, with the same definition) if it hasn't been defined
> by the class, otherwise use the definition from the class).
>
> So basically, my question is what should the defaults be, but it seems
> that you agree with the great journals, so I'll go that route.
>
> But what is the origin of the difference between mathematical and
> physical constants and for the exception to set capital Greek variables
> upright (but not in France)?
>
> Of course, $W$ is the variable for energy, W the unit watt, $m$ the
> variable for mass, m the unit meter, and so on. Could that be the
> reason?
>
> What about alpha, beta, and gamma rays? Should probably be upright?
> letters have new commands, since the usual ones are used for the italic
> characters. This makes it a pain when recycling something like a
> complicated formula. In general, I try to have all differences between
> journals which are not handled by the corresponding LaTeX class handled
> by my own macros. So first I load my own macros, which are my
> defaults, then journal-specific macro packages which redefine things
> which differ from my defaults. (That applies to many things, such as
> spelling, punctuation, italicization (or not) of foreign expressions,
> forms of dashes, and so on. Using \newcommand, \renewcommand, and
> \providecommand gives the necessary flexibility (e.g. one can define
> something (sensibly, with the same definition) if it hasn't been defined
> by the class, otherwise use the definition from the class).
>
> So basically, my question is what should the defaults be, but it seems
> that you agree with the great journals, so I'll go that route.
>
> But what is the origin of the difference between mathematical and
> physical constants and for the exception to set capital Greek variables
> upright (but not in France)?
>
> Of course, $W$ is the variable for energy, W the unit watt, $m$ the
> variable for mass, m the unit meter, and so on. Could that be the
> reason?
>
> What about alpha, beta, and gamma rays? Should probably be upright?
Phillip Helbig (undress to reply)
May 3, 2020, 3:04:06 AM5/3/20
to
In article <hh2lqp...@mid.dfncis.de>, Hendrik van Hees
> E.g., with the official change to the new SI the magnetic
> field constant became a derived quantity with an error, while before
> it was defined as an exact quantity by the old definition of the
> Ampere (two infinite straight wires...).
As you say, in the old system it was just what it was, an exact value.
One reason might be that roman (upright) latin letters are used for
units. So the unit tesla is upright T but italic T is temperature, say.
> Now it's to be measured
> based on the fixed value of the elementary charge and the definition
> of the second in terms of the Cesium hyperfine splitting frequency.
Like the meter: it used to be 1/40,000,000 of the circumference of the
Earth, but now the actual size of the Earth has to be measured.
Similarly, at one time a liter of water weighed exactly 1 kilogram.
> Last but not least, not all journals follow these rules. Sometimes
> they even don't like to set the differential d upright.
Some journals just have no style. One thing which I find rather strange
is the lack of a final full stop after a caption in some journals. Now
if the caption is just a phrase or a word and not a full sentence, then
that makes sense. But when it is a paragraph consisting of several
sentences and there is no full stop after the last one, that is just
strange.
> If you use Word, it's impossible to get a good microtypography
> anyway. There you can only hope that the journal transforms it
> into a readable text ;-)).
I always use LaTeX myself (or, for really simple things, just a text
file). Some journals which accept or even require submissions in LaTeX
convert it to something else when setting the final version. Some
journals even use WORD. (By the way, apparently new versions of WORD
can open PDF files and the result is a relatively good approximation of
what the PDF file looks like.
<he...@fias.uni-frankfurt.de> writes:
> Of course, after all these are conventions and often just a tradition.
> Why the physical fundamental constants are in italics and not upright
> is something I'm also puzzled about, but on the other hand, while
> mathematical constants are really constant and what they are, the
> values of the fundamental constants depend on the choice of the
> units.
I don't think that that's the reason.
> Of course, after all these are conventions and often just a tradition.
> Why the physical fundamental constants are in italics and not upright
> is something I'm also puzzled about, but on the other hand, while
> mathematical constants are really constant and what they are, the
> values of the fundamental constants depend on the choice of the
> units.
> E.g., with the official change to the new SI the magnetic
> field constant became a derived quantity with an error, while before
> it was defined as an exact quantity by the old definition of the
> Ampere (two infinite straight wires...).
One reason might be that roman (upright) latin letters are used for
units. So the unit tesla is upright T but italic T is temperature, say.
> Now it's to be measured
> based on the fixed value of the elementary charge and the definition
> of the second in terms of the Cesium hyperfine splitting frequency.
Earth, but now the actual size of the Earth has to be measured.
Similarly, at one time a liter of water weighed exactly 1 kilogram.
> Last but not least, not all journals follow these rules. Sometimes
> they even don't like to set the differential d upright.
is the lack of a final full stop after a caption in some journals. Now
if the caption is just a phrase or a word and not a full sentence, then
that makes sense. But when it is a paragraph consisting of several
sentences and there is no full stop after the last one, that is just
strange.
> If you use Word, it's impossible to get a good microtypography
> anyway. There you can only hope that the journal transforms it
> into a readable text ;-)).
file). Some journals which accept or even require submissions in LaTeX
convert it to something else when setting the final version. Some
journals even use WORD. (By the way, apparently new versions of WORD
can open PDF files and the result is a relatively good approximation of
what the PDF file looks like.
0 new messages