Re: List container strawman

[Ted Dennison<dennison@telepath.com>]

In article <9sdnb2$dd4$1@news.huji.ac.il>, Ehud Lamm says...
>
>Now I don't see any reason why newcomers shouldn't be able to reuse standard
>containers, but I really don't want to start explaining "controled type must
>be declared at library level".

To make matters worse, that isn't even where you'd have to start. You'd have to
start by talking about tagged types and declaration levels, then move to the
fact that controlled is a tagged type declared at the library level. :-)

>Are we coming to the conclusion that Ada95 approaches to memory management
>(Controlled and storage pools), simply aren't enough for creating reliable
>components, without imposing inessential restrictions?

I don't think there's anything horribly wrong with the rest of the language, but
we can certianly come to the conclusion that the way Controlled is currently
implemented blows. I've said as much to any who would listen for about the last
year or so. However, I would like to agree with Nick's point that my contempt is
aimed solely at the feature in its current state, not the folks who worked on
the language design. I could easily see where this consequence was either not
forseen, or it wasn't realised how big of a problem the restriction would be.

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html

No trees were killed in the sending of this message. 
However a large number of electrons were terribly inconvenienced.