Hi Joel,
you wrote:
>
> a: next "123456"
> b: next next a
>
>I suggest that there are three entities of interest:
>
>i) one which we get at via the variable 'a
>ii) one which we get at via the variable 'b
>iii) one which we can't get (directly) but which corresponds to a
> copy of the literal string "123456"
>
>These three entities are distinct. I think it's accurate to state
>that iii) is an anonymous data entity (whether it even HAS an index
>is moot, as it is inaccessible), while i) and ii) are named
>entities with individual/distinct indices that share the data
>contained in iii).
I believe that in essence we differ on whether the two example functions,
next and insert (and with them all other series related functions), can be
fully, and can only be explained by reviewing how they apply to your case
iii). You say about case iii)
>I think it's accurate to state
>that iii) is an anonymous data entity (whether it even HAS an index
>is moot, as it is inaccessible),
What is iii)?
>iii) one which we can't get (directly) but which corresponds to a
> copy of the literal string "123456"
See, if insert and next can be demonstrated to apply exclusively to your
case iii) then it is no longer true that:
>
>Performing {a: next a} changes something that is specific to 'a,
>while performing {append a "?"} changes something shared by 'a and 'b
>without changing anything that is specific to 'a.
If there is no 'a or 'b or 'c, and still 'next and 'insert continue to act
as they do when you apply them to 'a and 'b and 'c, then certainly what
'next changes canoot be something that is specific to 'a and so on.
In order to follow your example of keeping messages short, let me first
verify that you agree with the logic of my argument, before I proceed to
demonstrate that not only 'insert but also 'next is effective on the
"anonymous data entity" and all that 'a and 'b do is record the state of
that entity and nothing else.
Elan
