Re: User-defined integers: Asking for feedback

> But when or if I do it, you can't reserve lowtags for user
> code, because compressed pointers (to borrow the JVM term), i.e.
> NIL-relative objects, will become more important than the accidental
> discovery that you can take lowtag N for yourself.

Well, in my local tree I switched to a widetag now:

-  unused-widetag                                  ;  5A   61
+  udef-inttype-widetag                            ;  5A   61


> Since I never did anything, I didn't suggest you stop trying to work on
> this patch, but I'm unconvinced that your effort to land it in the main
> tree is going to reach a consensus.  This is a separate question from
> whether instructing random users how to take advantage of a particular
> detail of the object representation is a generally good idea.

Well, I guess there might be cleaner ways to integrate column-structures
into SBCL, but that would mean either
a) a more or less publicly-defined interface to do so (a kind of MOP),
b) or someone who knows how SBCL does structures now embeds the
    (in the contrib already implemented) accessor-defining code
    into SBCL proper.
    That would reduce the contrib code, as most of the DEFSTRUCT-option
    parsing would vanish.


Do you see these patches as "instructing random users...
particular detail" here as that problematic?
The (8 bit wide-)tagging scheme won't change (I guess),
and taking a free tag and building a _contrib_ so that "random users"
_don't_ need to care about internals (but can just use the contrib)
is the whole point.