Re: Testing dangerous change in tomorrow's snapshot (was: Re: Guile segv)

[Mikael Djurfeldt <mdj@nada.kth.se>]

Jim Blandy <jimb@red-bean.com> writes:

> > I agree on everything except the naming. What you call (guile-user)
> > should be called (guile), and what you call (interaction) should be
> > called (guile-user). It is a long-standing Lisp tradition that the
> > user (or foo-user) module is where the user _is_ by default, not what
> > the user _gets_ by default. In fact, my proposal would be perfectly
> > analogous to common lisp, which has package "CL" which exports all the
> > system bindings (possibly importing them from elsewhere and
> > re-exporting them), and package "CL-USER" which is the package where
> > the user gets dumped initially, and which starts with no bindings of
> > it's own, but importing everything from "CL". I think this is a good
> > naming convention to follow, i.e. that user is where the user is, not
> > what the user gets.
> 
> Okay.  I don't know CL, and I didn't realize there was such strong
> precedent.  Let's go with your suggestion.
> 
> Okay, Mikael?

Well, since what you call (guile-user) contains the bindings which
should be provided by an interaction environment, and since what you
call (interaction) contains the bindings created by the user, I think
this isn't all too bad.

Also, I did this change yesterday.

/mdj