Standard maxima installation under Linux?

I've been looking at it.  The streaming behavior is extremely low
level, which is actually good news because it means I will definitely
be able to catch the fact that a question has been asked, as long as
the program is smart enough to consistently recognize that it is a
question.  (What to do with it is the next question.  Passing it on to
the user looks like the only option.  I really prefer Mathematica's
approach, but I understand why it may be a bit tricky to implement.)

So, do all questions that interrupt evaluation in Maxima have at least
one line ending in a question mark?  (All the ones I've seen have, but
I haven't been comprehensive by any stretch of the imagination.)

On 2/13/06, Andrey G. Grozin <A.G.Grozin at inp.nsk.su> wrote:
> On Mon, 13 Feb 2006, Robert Dodier wrote:
> >>> well, it should be, but it is not -- at odd times maxima will try
> >>> to read something from the console, and waits forever if noone
> >>> is there to type something ... but if your program already works
> >>> on windows, you must have solved that problem.
> >>
> >> Actually, this still worries me.  So far, I've only used Maxima for
> >> factorization and solving equations, and I haven't had any trouble
> >> with either of those.  I'm thinking that when I add integration, it
> >> will be more likely to pop up.
> >
> > so far as i know, working around this behavior is completely a mess.
> > however, there are several front-end programs so i guess that it
> > has been solved in different ways. see:
> > http://maxima.sf.net/relatedprojects.shtml
> I don't think it's really a mess. A question like
> Is a positive or negative?
> is just a special kind of prompt. A front-end program (e.g., TeXmacs)
> treats it like any other prompt, like, e.g.,
> (%i123)
> and waits for a user input after it. This can be done easily by redefining
> *prompt-prefix* and *prompt-suffix*.
>
> Andrey
>