Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

What formulae markup/notation do you use?

2 views
Skip to first unread message

Steve

unread,
Nov 30, 2008, 7:01:17 AM11/30/08
to
I'm doing some research into the most appropriate technical solution for a
piece of maths software used by academics. One of the key problems is how to
allow a user to create and manipulate formulae. A TeX-based approach seems
to be the most obvious solution, although MathML is an alternative.

It would be greatly appreciated if you could answer the following questions:

1) How well known/widely used is TeX (e.g. LaTeX) in academia?
2) What proportion of university mathematics/science teachers/lecturers are
familiar with TeX/LaTeX?
3) What tools are generally used to manipulate TeX?
4) Is MathML gaining popularity in academic circles? Is it seen as an
alternative to TeX?

Many thanks,

Steve


Szabolcs

unread,
Nov 30, 2008, 7:18:52 AM11/30/08
to

You probably know that TeX is only meant to *typeset* formulae, and a
piece of TeX notation is not a proper representation of the
*structure* of a formula.

Unlike TeX, MathML *can* be used to describe the structure of
expressions (see Content MathML).

But MathML is not an alternative to TeX, as it was not meant to be
typed by hand. A WYSIWYG MathML + XML editor might be an alternative
for a TeX system, but a web browser, which can do nothing more than
display MathML, is not.

Marc van Dongen

unread,
Nov 30, 2008, 7:51:26 AM11/30/08
to
On Nov 30, 12:01 pm, "Steve" <st...@steve.com> wrote:

Hi Steve,

[questions about TeX MathML]

If I were you I'd also ask these questions in comp.text.tex.

LaTeX has a package called cool, which is meant to provide support for
content. You may find the package at http://www.ctan.org/tex-archive/macros/latex/contrib/cool/.

Regards,


Marc van Dongen

Steve

unread,
Nov 30, 2008, 8:53:11 AM11/30/08
to

> "Szabolcs" <szho...@gmail.com> wrote in message
> news:f96f9a03-1a60-46f1...@w35g2000yqm.googlegroups.com...

I have a requirement to allow user to easily define mathematical formulae.
Unfortunately a WYSIWYG MathML approach would be too costly to develop. As a
simpler alternative, allowing the users to define formulae using TeX seems
more pragmatic. So, for example, the user could enter the following into a
plain text editor:

The quadratic formula is $-b \pm \sqrt{b^2 - 4ac} \over 2a$
\end

The above defines a nice quadratic formula which will be displayed elswhere
in the application. Can you, or anyone else, think of a more efficient
approach (in terms of development effort and usability) than this?

Thanks.


Steve

unread,
Nov 30, 2008, 8:54:13 AM11/30/08
to
> "Marc van Dongen" <don...@cs.ucc.ie> wrote in message
> news:3c21dabc-837d-44bb...@o2g2000yqd.googlegroups.com...

> On Nov 30, 12:01 pm, "Steve" <st...@steve.com> wrote:
> If I were you I'd also ask these questions in comp.text.tex.

Thanks Marc, I'll post on it when I get the TeX concept clearer in my head!
:)


Szabolcs

unread,
Nov 30, 2008, 9:54:33 AM11/30/08
to
On Nov 30, 2:53 pm, "Steve" <st...@steve.com> wrote:
>
> I have a requirement to allow user to easily define mathematical formulae.
> Unfortunately a WYSIWYG MathML approach would be too costly to develop. As a
> simpler alternative, allowing the users to define formulae using TeX seems
> more pragmatic. So, for example, the user could enter the following into a
> plain text editor:
>
> The quadratic formula is $-b \pm \sqrt{b^2 - 4ac} \over 2a$
> \end
>
> The above defines a nice quadratic formula which will be displayed elswhere
> in the application. Can you, or anyone else, think of a more efficient
> approach (in terms of development effort and usability) than this?
>

Is your aim to allow the user to define how a formula should *look
like*, or do you need to do calculations with that formula?

A (perhaps simplified) (La)TeX notation is good for just formatting
the formula. But if you need to do calculations with it, and if the
precise way of formatting doesn't matter so much, then I'd suggest to
invent your own notation (look at the notation used by CAS software
such as Mathematica for inspiration)

Szabolcs

unread,
Nov 30, 2008, 9:56:20 AM11/30/08
to
You can get some inspiration from ASCIIMath, too (use Firefox to view
the page):

http://www1.chapman.edu/~jipsen/asciimath.html

Paul Rubin

unread,
Nov 30, 2008, 10:11:51 AM11/30/08
to
Steve wrote:
>
> 1) How well known/widely used is TeX (e.g. LaTeX) in academia?

It varies by discipline and perhaps by geography. My impression is that
people in math and physics tend to know of it and frequently use it; my
colleagues in our business school generally give me a blank stare (or
think I'm referring to a paint base, if I spell it out). I think it's
also somewhat widely known in areas (such as chemistry or linguistics)
where writing involves frequent use of funky diagrams or formulas, even
if they're not mathematical formulas. I also have a sense that it might
be more widely used in Europe than in the US (possibly because they're
less addicted to M$ Word).

> 3) What tools are generally used to manipulate TeX?

That depends on how comfortable you are rooting around in raw (La)TeX.
If you want to work at the base level, several editors such as Notepad++
(Windows) and I think emacs (*nix) have some built-in syntax
recognition. There are a bunch of text editors with tool buttons or
menus designed to let you run latex against the current file.

There are also several programs that look more like conventional word
processors, but typeset with LaTeX, either on-the-fly (which I think
tends to be slow) or on command. I use, and highly recommend, LyX
(www.lyx.org), which is open source. There's also Scientific Word
(LaTeX based clone of M$ Word), which is commercial, and a couple of
others I can't remember off-hand.

Lastly, IIRC there's a plug-in for the Eclipse IDE to develop LaTeX
documents.

> 4) Is MathML gaining popularity in academic circles? Is it seen as an
> alternative to TeX?

I haven't seen much use of it myself, other than occasionally in web
pages. It's not a full typesetting system, so it's not a true
alternative to TeX/LaTeX, other than in special circumstances.

/Paul

rjf

unread,
Nov 30, 2008, 11:48:16 AM11/30/08
to
On Nov 30, 4:01 am, "Steve" <st...@steve.com> wrote:
> I'm doing some research into the most appropriate technical solution for a
> piece of maths software used by academics. One of the key problems is how to
> allow a user to create and manipulate formulae. A TeX-based approach seems
> to be the most obvious solution, although MathML is an alternative.
>
> It would be greatly appreciated if you could answer the following questions:
>
> 1) How well known/widely used is TeX (e.g. LaTeX) in academia?

As others have said, it is widely known in some disciplines.

> 2) What proportion of university mathematics/science teachers/lecturers are
> familiar with TeX/LaTeX?

Ditto.


> 3) What tools are generally used to manipulate TeX?

Text editors, previewers, computer algebra systems, interactive
editors.

> 4) Is MathML gaining popularity in academic circles? Is it seen as an
> alternative to TeX?

Probably not.


If you want a user to create and manipulate formulas, then the kind of
software
you need is a Computer Algebra System (CAS) not a typesetting/display
system.
Manipulation would include simplification, symbolic operations like x
+x -->2x, etc.

A CAS can typically provide TeX versions of any formulas that are
produced
as intermediate or final expressions.

For CAS, see, for example.
http://en.wikipedia.org/wiki/Comparison_of_computer_algebra_systems

G. A. Edgar

unread,
Nov 30, 2008, 1:42:20 PM11/30/08
to
In article <egvYk.28379$hy4....@newsfe11.ams2>, Steve
<st...@steve.com> wrote:

> I'm doing some research into the most appropriate technical solution for a
> piece of maths software used by academics. One of the key problems is how to
> allow a user to create and manipulate formulae. A TeX-based approach seems
> to be the most obvious solution, although MathML is an alternative.
>
> It would be greatly appreciated if you could answer the following questions:
>
> 1) How well known/widely used is TeX (e.g. LaTeX) in academia?

Among mathematicians and a few related fields, very well.

> 2) What proportion of university mathematics/science teachers/lecturers are
> familiar with TeX/LaTeX?

mathematics: virtually all
science (other than mathematics & a few related fields): virtually none
other academics: virtually none

> 3) What tools are generally used to manipulate TeX?

What do you mean by "manipulate"?

> 4) Is MathML gaining popularity in academic circles?

I have seen no evidence of it.

> Is it seen as an alternative to TeX?

No

>
> Many thanks,
>
> Steve
>
>

--
G. A. Edgar http://www.math.ohio-state.edu/~edgar/

Steve

unread,
Nov 30, 2008, 5:54:07 PM11/30/08
to
"G. A. Edgar" <ed...@math.ohio-state.edu.invalid> wrote in message
news:301120081342204297%ed...@math.ohio-state.edu.invalid...
<snip!>

Many thanks to all who have answered.


Steve

unread,
Nov 30, 2008, 5:55:13 PM11/30/08
to
> "Szabolcs" <szho...@gmail.com> wrote in message
> news:5fb0f5bf-813e-4f48-a471->24aee9...@n10g2000yqm.googlegroups.com...

> On Nov 30, 2:53 pm, "Steve" <st...@steve.com> wrote:
> >
> > I have a requirement to allow user to easily define mathematical
> > formulae.
> > Unfortunately a WYSIWYG MathML approach would be too costly to develop.
> > As a
> > simpler alternative, allowing the users to define formulae using TeX
> > seems
> > more pragmatic. So, for example, the user could enter the following into
> > a
> > plain text editor:
> >
> > The quadratic formula is $-b \pm \sqrt{b^2 - 4ac} \over 2a$
> > \end
> >
> > The above defines a nice quadratic formula which will be displayed
> > elswhere
> > in the application. Can you, or anyone else, think of a more efficient
> > approach (in terms of development effort and usability) than this?
> >
>
> Is your aim to allow the user to define how a formula should *look
> like*, or do you need to do calculations with that formula?

It's just a visual thing. No calculations.


Rene Grothmann

unread,
Dec 1, 2008, 3:30:12 AM12/1/08
to
I agree that you can safely say that Latex is used in Math by
virtually everybody on the University level, and in other sciences by
virtually nobody. But even most high school math teachers would rather
use Word and its formula editor than LaTeX, even when they are aware
of its inferior output. The reason is, of course, that LaTeX is not
really user friendly, especially when we want images or a more fancy
page layout.

Some remarks about MathML: TeX has been invented before the Net came.
The markup is very different from the now favored XML style. It is
easier, shorter and more human readable (esp. with packages like the
above mentioned cool), but less stringent and exact. MathML was
designed to do the same as TeX for web documents. It was not really
accepted due to its cumbersome and longish expression style and the
missing browser support, nor did the popular programs for web design
implement visual tools for it. Thus, we now have various LaTeX style
extensions for math in web pages, or even use images to represent math
in the Web. It is a shame.

Jon G.

unread,
Dec 1, 2008, 10:08:54 AM12/1/08
to
I use Math Type to create formulas, then use Excel to do calculations. One
application is the Trisection Solution page
http://mypeoplepc.com/members/jon8338/math/id8.html This gives the correct
solution. Somebody could copy my Excel Workbook and embed it in HTML to
make a web site that calculates the answer, or embed it in a scientific
application to crank out continuous solutions. It has *portability* The
Math Type equations are just aesthetic to visualize the solution and aren't
used in the calculations.

"Steve" <st...@steve.com> wrote in message
news:egvYk.28379$hy4....@newsfe11.ams2...

Nasser Abbasi

unread,
Dec 2, 2008, 2:21:47 PM12/2/08
to

"Rene Grothmann" <mga...@gmx.de> wrote in message
news:1b852838-3bbf-4efd...@t3g2000yqa.googlegroups.com...

> The reason is, of course, that LaTeX is not
> really user friendly, especially when we want images or a more fancy
> page layout.
>

Scientific word (SW) helps with the user friendly bit. It has an easy to use
interface, and generates portable Latex code in the background.

Currently it is at version 5.5, I have been waiting for version 6.0 of SW
for years now, they promised more enhancements. May be they will release
version 6.0 this century I hope.

But it is a very good product. I use it to typeset all my homework's and
reports since the output is Latex, it looks good.

Check it out, they have free 30 days trial.

Nasser


Paul Rubin

unread,
Dec 2, 2008, 4:41:00 PM12/2/08
to

Or try the open source LyX (www.lyx.org).

Robert Dodier

unread,
Dec 2, 2008, 6:21:09 PM12/2/08
to
On Nov 30, 6:53 am, "Steve" <st...@steve.com> wrote:

> I have a requirement to allow user to easily define mathematical formulae.

> The quadratic formula is $-b \pm \sqrt{b^2 - 4ac} \over 2a$

Well, as a possibility, you could allow the user to type
stuff in more conventional notation, and then parse that,
and generate TeX or MathML or whatever for display.

e.g. user types in (-b +- sqrt(b^2 - 4*a*c))/(2*a)
and some TeX is generated from that expression.

Incidentally the program Maxima can do the parsing
and TeX generating. See: http://maxima.sourceforge.net

HTH

Robert Dodier

Rene Grothmann

unread,
Dec 3, 2008, 2:03:49 AM12/3/08
to
I tried LyX. Found it not good enough to be usable for me (your
mileage may vary, try yourself!). I do not want to write a review
here, since this is a newsgroup. I will do so on my math blog.

I also have used Scientific Word long time ago. It produces standard
LaTeX, if you do not count the huge package it needs. I stopped using
it, when I tried to compile its output with my standard LaTeX (and the
package needed), and failed. So this was no option for me. Maybe I
should try again.

After all experimenting, I will just go ahead with a simple editor and
a standard LaTeX compiler.

And yes, MathML was probably not designed to be human readable.
Someone mailed that to me personally. Maybe we should start a
discussion about MathML editors. So far, I am only aware of some CAS
system that can export MathML.

Kamil Dedecius

unread,
Dec 3, 2008, 8:40:08 AM12/3/08
to
Hello Steve,

a few remarks from me:

> 1) How well known/widely used is TeX (e.g. LaTeX) in academia?

Widely when the subject is math, otherwise I think some wysiwyg SW is
preferred.

> 2) What proportion of university mathematics/science teachers/lecturers are
> familiar with TeX/LaTeX?

I think that almost everyone who is interested in higher mathematics
is familiar with TeX/LaTeX. Note, that most issuers of math.
scientific papers accept contributions in own latex style. Anyway, a
lot of them have a stylefile for MS Word as well.

> 3) What tools are generally used to manipulate TeX?

I prefer the unix editor ViM, however you can use any text editor
(even M$ Word :o)). A lot of people like the LyX editor (http://
www.lyx.org), as it is partly wysiwyg and available for free. When
recommending latex, I usually notice texmaker ( http://www.xm1math.net/texmaker),
which is very user-friendly and free as well.

> 4) Is MathML gaining popularity in academic circles? Is it seen as an
> alternative to TeX?

I have never seen any of my colleagues to use MathML.

Generally, LaTeX is a great software for typesetting of documents,
presentations, mathematics and whatever. However, it is not very easy
to learn to use it, but once you learn it, it's likely that nothing
else will satisfy you. The alternatives are then either too
complicated to learn/use, or they do not meet your expectations.

Best regards,
Kamil

Victor Porton

unread,
Dec 3, 2008, 10:50:53 AM12/3/08
to
On Nov 30, 2:01 pm, "Steve" <st...@steve.com> wrote:
> I'm doing some research into the most appropriate technical solution for a
> piece of maths software used by academics. One of the key problems is how to
> allow a user to create and manipulate formulae. A TeX-based approach seems
> to be the most obvious solution, although MathML is an alternative.

I suggest to use TeXmacs (http://www.texmacs.org). It is a powerful
WYSIWYG editor.

TeXmacs has export to LaTeX and HTML (with MathML or with images)
formats.

Philip A. Viton

unread,
Dec 3, 2008, 11:17:38 AM12/3/08
to
In article <22a667ad-a916-428b-a68e-b0f2d3fef0d7
@t3g2000yqa.googlegroups.com>, mga...@gmx.de says...


SW's "huge package" for standard latex support is really just a
few megs worth of files: LyX has the same kind of needs. (Or
did you mean the size of the entire SW installation, which *is*
huge? That's because comes with its own complete TeX system.)
In general it is possible to compile with another system like
MikTeX: what you lose in that case is the ability (a) to
support a huge number of graphics types and (b) TrueType fonts.
On the other hand because SW works only with TrueType fonts,
you can't use Metafont-generated ones; you may consider that a
serious problem, I don't know. Recent editions of SW come with
pdftex support built-in, which solves an old problem of
somewhat flaky support for type-1 fonts.

--


Phil Viton
Ohio State University

Michael Press

unread,
Dec 4, 2008, 8:44:33 PM12/4/08
to
In article
<8cbd572a-1031-4405...@u14g2000yqg.googlegroups.com>,
Kamil Dedecius <kamil.d...@gmail.com> wrote:

I am surprised that LaTeX is not widely adopted by physicists.

--
Michael Press

chle...@gmail.com

unread,
Dec 31, 2008, 6:30:22 PM12/31/08
to
If you are a Mac user, you may give MathMagic a try.
It is a WYSIWYG equation editor. It also support typing TeX or LaTeX
expression directly.
Equations can be copied out as Plain TeX, LaTeX, AMS LaTeX MathML, as
well as other common graphic formats such as PDF or PICT or JPEG.

So whether you are fluent in TeX or not, MathMagic will be a good
starting point to make great looking equations and learn MathML or TeX
expressions.

www.mathmagic.com offers fully working trial versions.

Charles

0 new messages