Requesting input on some possible low-level changes

On Wed, 2003-09-03 at 15:11, Richard Fateman wrote:
> The debugging REPL in Allegro Common Lisp is rather more elaborate, with 
> backtraces, access to local variables, and a zillion other things that I 
> personally never use but might use if I learned more about them.

Right. That's why I was trying to make a strong distinction between
access to the basic REPL and the debugging REPL. The plan I described
wouldn't leave an easy way to get to the basic REPL, but it would be
easy to restore the native debugging REPL -- just set *debugger-hook* to
nil. Using the native debugger would preclude using the TeXmacs (or
other similar) interface; one would have to rely on text mode.

>  I would 
> not be surprised if the actual break loop could be instrumented to print 
> (say) a particular character at line-position 0 to indicate the fact 
> that you were not talking to Maxima.

I don't see how to do this in any of the other Lisps, but I haven't
looked very hard. In any event, I don't see any way to do so within the
ANSI standard. I was thinking that there would be some way to implement
a generic debugging REPL such that access to at least some of
implementation-specific functionality in the underlying lisp would be
exposed. I haven't tried to implement anything along these lines yet,
however.

> The proposed solution represents something of a "dumbing down" of
> debugging, and I doubt that I would find something without
> a backtrace and local variable access  useful for debugging.

Keep in mind that I'm not proposing to remove the ability to use the
real lisp debugger in text mode. In fact, I'm not proposing to remove
the ability to use the debugger from TeXmacs, either -- such ability was
never there.

--Jim