\phi versus \varphi
coo...@nospam.verizon.net
My anal compulsive editors (who soend their off-time wondering if anal
compulsive should be hyphenated or not - I digress...) have asked me to
consider a global change of \phi to \varphi. While this is doable
(although it amounts to changing approximately 1100 instances of \phi
over 800+ pages of the book), I'm trying to see if there is a compelling
argument one way or the other beyond aesthetics (which in this instance
don't lead me one way or the other).
In said document, phi (of whatever glyph) represents a parameter (in big
multinomial models and likelihoods and such). My impression had been
that \phi was more appropriately used for such things, and that \varphi
really was the lower-case of the Greek letter phi (in other words, if I
was writing *text* in Greek, I'd use \varphi). But this is math (which
is Greek to some. Again I digress...), so I'd thought \phi.
Any thoughts? I've resisted using my 'get stuffed' email macro to the
editor until I'd thought about it a bit - and got some more learned (or
at least objective) opinions.
Ta much...
lomov.vl
I think if editors asked you to change all \phi (this glyph) to
\varphi (that glyph) you should agree and change them :-)
But do this in "minimal" way:
\let\oldphi\phi
\let\phi\varphi
---
WBW, Vladimir Lomov
Robin Fairbairns
>My anal compulsive editors (who soend their off-time wondering if anal
>compulsive should be hyphenated or not - I digress...) have asked me to
>consider a global change of \phi to \varphi. While this is doable
>(although it amounts to changing approximately 1100 instances of \phi
>over 800+ pages of the book), I'm trying to see if there is a compelling
>argument one way or the other beyond aesthetics (which in this instance
>don't lead me one way or the other).
>
>In said document, phi (of whatever glyph) represents a parameter (in big
>multinomial models and likelihoods and such). My impression had been
>that \phi was more appropriately used for such things, and that \varphi
>really was the lower-case of the Greek letter phi (in other words, if I
>was writing *text* in Greek, I'd use \varphi). But this is math (which
>is Greek to some. Again I digress...), so I'd thought \phi.
my recollection is that none of my (modern) greek textbooks use
\varphi, but that my teacher's handwriting (and certainly mine) do. i
don't have any textbooks here, so can't comment further until i've
taken a look at them. if appropriate, i could dig out the books this
evening. (of course, there's no maths in any of those books.)
apostoulos syropoulos (the author of several packages on ctan) is a
mathematician, iirc, and might be able to provide a more nearly
authoritative answer.
but as vladimir's pointed out, the change to the sources would surely
take less than a minute anyway so you can delay...
--
Robin Fairbairns, Cambridge
G. A. Edgar
> have asked me to
> consider a global change of \phi to \varphi. While this is doable
> (although it amounts to changing approximately 1100 instances of \phi
> over 800+ pages of the book),
No, don't change 1100 instances. That is the Microsoft Word way to do
it. Instead, use the TeX way. Here: I looked at a document I am
working on. Near the top I have these lines...
\renewcommand{\phi}{\varphi}
\renewcommand{\epsilon}{\varepsilon}
\renewcommand{\emptyset}{\varnothing}
I guess only one of them is what you want, but others can be done, too!
> Any thoughts? I've resisted using my 'get stuffed' email macro to the
> editor until I'd thought about it a bit - and got some more learned (or
> at least objective) opinions.
Save your energy for important arguments. In this instance, just do
what the editor wants.
--
G. A. Edgar http://www.math.ohio-state.edu/~edgar/
lomov.vl
on Aug 5, 9:03 pm, "G. A. Edgar" <ed...@math.ohio-state.edu.invalid>
wrote:
>> have asked me to
>> consider a global change of \phi to \varphi. While this is doable
>> (although it amounts to changing approximately 1100 instances of \phi
>> over 800+ pages of the book),
>
> No, don't change 1100 instances. That is the Microsoft Word way to do
> it. Instead, use the TeX way. Here: I looked at a document I am
> working on. Near the top I have these lines...
>
> \renewcommand{\phi}{\varphi}
> \renewcommand{\epsilon}{\varepsilon}
> \renewcommand{\emptyset}{\varnothing}
I recommend to use \let. In case if you want glyph that (old) \phi
produce how you can get it?
\let\oldphi\phi
\let\phi\varphi
Now \phi prints the same glyph as \varphi while \oldphi gives you the
glyph which \phi prints by default.
Actually I input following lines into my document when I need \phi
prints the same glyph as \varphi (as well as epsilon):
\makeatletter
\let\old@phi\phi
\let\old@varphi\varphi
\let\old@epsilon\epsilon
\let\old@varepsilon\varepsilon
\let\phi\old@varphi
\let\varphi\old@phi
\let\epsilon\old@varepsilon
\let\varepsilon\old@epsilon
\makeatother
Don't bother that @ do here these lines just swap \phi<->\varphi,
\epsilon<->\varepsilon.
By the way, if one editor thinks that (old) \phi should be as (old)
\varphi then the lines does job. If another editor thinks
that (old) \varphi should be as (old) \phi just comment lines.
---
WBW, Vladimir Lomov
Marc Olschok
I think \varphi looks better most of the time (partly because it seems
to come up more often in handwriting). The \phi might in fact be
confused with \emptyset by a casual reader. For the big letter there
is always \Phi available.
--
Marc
coo...@nospamverizon.net
of doing something like
\renewcommand{\phi}{\varphi}
would save the search and replace time (although on a good machine, not
too odious...).
Cheers...
coo...@nospam.verizon.net
> I think \varphi looks better most of the time (partly because it seems
> to come up more often in handwriting). The \phi might in fact be
> confused with \emptyset by a casual reader. For the big letter there
> is always \Phi available.
>
I tend to agree (about the aesthetic merits of \varphi), and understand
your point about \phi - n one would 'casually' read the MS in question,
though. ;-)
As for cap \Phi, I thought about that. If (i) the vast majority of the
other parameters weren't lower case, and (ii) if I was using cap symbols
for matrices and vectors (by and large), I'd consider it.
Thanks again.
Dan
> Greetings -
>
> My anal compulsive editors (who soend their off-time wondering if anal
> compulsive should be hyphenated or not - I digress...) have asked me to
> consider a global change of \phi to \varphi. While this is doable
> (although it amounts to changing approximately 1100 instances of \phi
> over 800+ pages of the book), I'm trying to see if there is a compelling
> argument one way or the other beyond aesthetics (which in this instance
> don't lead me one way or the other).
>
> In said document, phi (of whatever glyph) represents a parameter (in big
> multinomial models and likelihoods and such). My impression had been
> that \phi was more appropriately used for such things, and that \varphi
> really was the lower-case of the Greek letter phi (in other words, if I
> was writing *text* in Greek, I'd use \varphi). But this is math (which
> is Greek to some. Again I digress...), so I'd thought \phi.
>
> Any thoughts?
I see \phi and \varphi in roughly equal numbers in math
journal/monographs/textbooks. The result of \phi (in CM fonts
anyway) somehow looks more like a symbol than a variable,
so I personally prefer \varphi. Also, a lot of beople (incorectly,
IMO)
use \phi for the empty set.
Since \varphi takes twice as long to type, my articles all begin with
\let\oldphi\phi % optional
\let\phi\varphi
If I had to deal with really anal retentive editors, I'd run a
search-and-replace operation: just in case they were so
AR as to require the _source_ be changed. The total time
to change 1100 instances? No more than it takes to type
the above two lines. Really.
Dan
Mathias Lindner
A lot of books in my field (electrical machines) use both symbols for
different values. While \varphi often represents an angle, \phi
represents a factor.
So, if your editor tells you what to do you'll prob. have no other
chance. But you both should look what's convenient in your field...
Mathias
Leo
> A lot of books in my field (electrical machines) use both symbols for
> different values. While \varphi often represents an angle, \phi
> represents a factor.
BTW, what does 'var' stand for in \varphi?
--
Emacs uptime: 10 days, 13 hours, 59 minutes, 6 seconds
Philipp Stephani
> On 2009-08-06 08:34 +0100, Mathias Lindner wrote:
>> A lot of books in my field (electrical machines) use both symbols for
>> different values. While \varphi often represents an angle, \phi
>> represents a factor.
>
> BTW, what does 'var' stand for in \varphi?
Variant.
--
Replace “READ-MY-SIG” by “tcalveu” to answer by mail.
Leo
> Leo schrieb:
>> On 2009-08-06 08:34 +0100, Mathias Lindner wrote:
>>> A lot of books in my field (electrical machines) use both symbols for
>>> different values. While \varphi often represents an angle, \phi
>>> represents a factor.
>>
>> BTW, what does 'var' stand for in \varphi?
>
> Variant.
Thank you for the quick reply.
--
Emacs uptime: 10 days, 14 hours, 8 minutes, 11 seconds
Dr Engelbert Buxbaum
> Greetings -
>
> My anal compulsive editors (who soend their off-time wondering if anal
> compulsive should be hyphenated or not - I digress...) have asked me to
> consider a global change of \phi to \varphi. While this is doable
> (although it amounts to changing approximately 1100 instances of \phi
> over 800+ pages of the book), I'm trying to see if there is a compelling
> argument one way or the other beyond aesthetics (which in this instance
> don't lead me one way or the other).
Not really. Either do a global search and replace with your text editor,
or simply switch the meaning of \phi and \varphi in the limbo. Each takes
only a minute.
> In said document, phi (of whatever glyph) represents a parameter (in big
> multinomial models and likelihoods and such).
Check what the conventions in your field are, in my case the IUPAC "Green
Book" (Quantities, Units and Symbols in Physical Chemistry,
http://old.iupac.org/publications/books/gbook/index.html) is the
autoritative source.
> My impression had been that \phi was more appropriately used for such
> things, and that \varphi really was the lower-case of the Greek letter
> phi (in other words, if I was writing *text* in Greek, I'd use \varphi).
no, an uppercase of \phi is \Phi