On 08/25/2015 11:29 AM, Jamison Hope wrote:
> On Aug 25, 2015, at 8:37 AM, John Cowan <xxxxxx@mercury.ccil.org> wrote:
>
>> Bradley Lucier scripsit:
>>
>>> I'm leaning towards "simple-array", so what do you think are the issues?
>>
>> I was concerned that it might confuse Common Lispers, but on reflection
>> I don't think it will, even though it's possible for a simple-array to
>> be displaced to (i.e. share bodies with) with another simple-array.
>> The more important point is that it is not adjustable (that is, the
>> dimensions can be mutated without breaking eq?).
>>
>> So sure, "simple-array" is fine.
>
> Just to muddy up the water a bit, what about "ordinary-array"?  They're
> ordinary in the sense that they're what we usually think of as arrays in
> programming languages, and they're also ordinary in that elements are
> accessed by ordinal position.  And, it doesn't conflict with existing
> terminology from Lisp.

The more I think about it, these arrays are neither "simple" nor "ordinary".

They have a backing store, an indexer that might jump around in the
backing store, a storage-class that itself has several different
properties, and a setter in addition to the minimal getter and domain
(which is the kind of simple array that array-map, mutable-array-curry,
array-curry, array-distinguish-one-axis,
mutable-array-distinguish-one-axis, ... return).

So in many ways what the SRFI currently calls fixed-arrays are
"special", or "specialized".

So how about

special-array

or

specialized-array

???

Brad