mailing-list for TeXmacs Users

[Joris van der Hoeven <address@hidden>]

On 4 Mar 2002, Michael John Downes wrote:

> Joris van der Hoeven <address@hidden> writes:
> 
> > 1. The multiplication should be entered using "*".
> >    Example: "x * y" or "a * ( b + c )"
> > 
> > 2. A function or symbol operating on a value or something else
> >    should be entered using brackets or "space".
> >    Examples: "f ( x )" or "d space x"
> > 
> > 3. All successions of roman letters of length > 1
> >    are considered as operators.
> >    Example: "s i n space x" or "L i space z"
> 
> You might want to consider the Scientific Word/Workplace way of doing
> this, which is similar, but rule 3 is applied only when the sequence of
> letters typed by the user is recognized as a defined operator name;

Yes, but the concept of a "defined operator name" is very flexible;
what about "Li" or "mon"?

> otherwise it is treated as a product of one-letter variables (rule 1)
> and shows in italic rather than roman.

This is a problem with dx, because that is "d space x" and not "d * x".
Also "Lf" usually means "L space f" and not "L * f".
I do not like schemes which are too heuristic (a specialty of M$;
anyway I do not have the money to buy Scientific Word);

> This is perhaps less consistent from a purely logical point of view,
> but it may be more useful for ordinary users.

I prefer the current scheme which may be slightly longer
in certain circumstances, but it has the good property that
it forces users to think about what they type.

> It does mean that some users simply leave unusual operator names to
> print in italic instead of giving them a proper definition. Because they
> don't notice the distinction. But this problem also affects the TeXmacs
> approach, in the opposite direction: some users will type "abc"
> intending the product "a*b*c", and expect it to print in italic instead
> of roman. In fact this is where this thread started, I believe, with
> "dx" :-)

Well, I guess that the users do have to spend five minutes at least
in order to learn typing mathematics in TeXmacs. In any case,
the current conventions are simple, unambiguous and systematic.

> Also I believe there is a mechanism in SW for defining that certain
> letters in the current document should be treated (by default) as
> functions, which affects the interpretation of things like
> 
>   a(b+c)    versus     f(b+c)

I think that they also implemented a functionality which
magically types the text that you had in mind when hitting F12 :^)

-Joris-


----------
P.S.: from version 0.3.4.14 on, there will be a tiny space
for operator application, e.g. "L space f" or "f space x".
There will also be a symbol \big. to "close" a \sum (for instance).