Re: Generic Package

["Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>]

On Thu, 03 May 2007 20:27:59 +0200, Georg Bauhaus wrote:

> On Wed, 2007-05-02 at 15:12 +0200, Dmitry A. Kazakov wrote:
>> On Wed, 02 May 2007 13:48:26 +0200, Georg Bauhaus wrote:
>> 
>>> On Wed, 2007-05-02 at 12:29 +0200, Dmitry A. Kazakov wrote:
>>> 
>>>>> Memory is not abstract, addresses aren't abstract,
>>>> 
>>>> In what sense? 
>>> 
>>> When you write a Set implementation for a PC, you can specifying
>>> addresses and refer to addresses in a consistent way.
>> 
>> But I am not required to do so.
> 
> (I'll listen to Randy who said: "In any case, this is an Ada forum,
> and abstractions that you cannot describe in Ada are simply not
> relevant", and be brief, for one more time only.)

There exist legal Ada programs which don't refer to the type
System.Address.

(which statement is an exact equivalent to "I am not required to do so"
[use addresses in Ada])

>>  Now, let you have a container holding 1, 2, 3, can you
>> point me the address of 2 there?
> 
> Yes. It is between 1 and 3, the docs say.

(Where RM says anything about 2'Address?)

> Whether it actually is a
> computer address between 1'Address and 3'Address is a matter of
> implementation,

Also it is not memory address. What is then, a postal address?

> I had given an argument about the correctness of the program WRT
> some model of how Foreach works.

You didn't specified that "model." I am especially interested in the formal
meaning of "once."

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de