RE: [External] : Re: Setting up abbrev

[Drew Adams <drew.adams@oracle.com>]

> Anyhow, the so-called dotted pair notation tells us that the
> universal data structure actually isn't a list, but a pair!

Yes.

> Only a pair is a list as well ...

Yes.  But a list isn't a pair. ;-)

> In practice - here, someone maybe disagrees? - but in my
> experience I don't see why that should be used ...
> 
> And I don't know why it was ever essential to the point it
> even got to symbolize Lisp on a couple of occasions?
> 
> Maybe that box diagram with arrows to make up a cons cell
> link-list-fragment just looked neat and interesting ...

1. Key-value pairs are quite common - far beyond Lisp.

2. A cons is a key-value pair.

But the constructor function `cons' is more/other than
that.  As a key-value pair, (cons a (cons b c)) has key
`a' and value key-value pair (b . c).  Not your typical
key-value pair use case - not typically thought of that
way, at least.

3. A fundamental structure-building thingie in our
universe is the unary constructor, `s' (successor
function).  Together with the nullary constructor, `0',
it gives you the Natural numbers - pretty foundational.

The next structure-building thingie is the binary
constructor, `cons'.  Together with nullary constructor
`nil' it gives you key-value pairs, or if you prefer,
essentially all structures/constructions - sculptures
of all sorts.

Syntax-tree and function-application are fundamental
structures for programs and (other) data.  In Lisp,
both are directly, unabashedly, handled as conses.

If you want lists then you have conses.  If you want
trees then you have conses.

As for the (a . b) _notation_: Why not?  JSON uses
`:'.  Prolog uses `|'.  Doesn't matter much what
notation you choose.  OOPs often use a dot notation
for method application.  Six of one; half a dozen of
another...