Re: Generators/coroutines in future Ada?

["Randy Brukardt" <randy@rrsoftware.com>]

"Victor Porton" <porton@narod.ru> wrote in message 
news:ok3meg$fg6$1@gioia.aioe.org...
> Dmitry A. Kazakov wrote:
>
....
>>> A task is a perfect abstraction for a thread of control that maintains
>>> its own execution state!
>>
>> Except that no control is required.
>>
>> I don't understand what is wrong with:
>>
>>     type Stateful is limited private;
>>     function Generate (X : not null access Stateful) return Things;
>
> The thing "wrong" with this, is that it is sometimes hard to implement
> Generate(). In Python this can be done in an easier way by using 
> generators
> (Python's "yield" keywords).

The thing "wrong" with this is that it forces sequential execution. A Python 
generator is by its nature a serial bottleneck (indeed, when this was 
discussed on the ARG list, that was suggested as an advantage!!! of the 
Python approach).

The coming problem with "classic" programming languages like Ada is that 
they don't map well to architectures with a lot of parallelism. The more 
constructs that require sequential execution, the worse that Ada programs 
will perform. With that in mind, it is hard to imagine why anyone would want 
to invest the hundreds or even thousands of hours required for a new 
language version on still more sequential constructs. Especially constructs 
that by their nature require sequential execution.

                               Randy.