Re: Type safety on wikipedia

[Martin Krischik <krischik@users.sourceforge.net>]

Larry Kilgallen wrote:

> In article <17514424.vPDbTzKvie@linux1.krischik.com>, Martin Krischik
> <krischik@users.sourceforge.net> writes:
>> Larry Kilgallen wrote:
>> 
>>> In article <1138363035.664717.262330@g47g2000cwa.googlegroups.com>,
>>> "Martin Krischik" <krischik@users.sourceforge.net> writes:
>>> 
>>>> You can manualy free memory - provided all pointers pointing to the
>>>> freed memory are set to null at the same time. And access to a null
>>>> pointer is handled in orderly manner.
>>> 
>>> If a program is correct, having the pointers become null is not an
>>> issue. Referencing through a null pointer or a stale pointer is equally
>>> wrong.
>> 
>> Not of access of a null pointer raises an exceptions. Exceptions are
>> considered well defined. See the talk page.
> 
> I gather you are trying to say some people plan to have their programs
> access via a null pointer and then handle an exception.  That seems
> incredibly sloppy.  It is also likely to be quite inefficient on some
> Ada implementations.

In praxis it is indeed sloppy. However: the program does not get "stuck" and
has a well defined path of execution - and that is the important part for
the theory.
 
> I have no idea what you mean by "the talk page".

Any Wikimedia page has a talk page associated where the authors discuss the
page progress:

Page: http://en.wikipedia.org/wiki/Type_safety
Talk: http://en.wikipedia.org/wiki/Talk:Type_safety

There is a link at the top of the page to the talk page. Using the
English/Monobook user interface the link is called [_Discussion_].

For the page in question the discussion is about thee time the main article
- a very controversial toppic indeed.

Martin
-- 
mailto://krischik@users.sourceforge.net
Ada programming at: http://ada.krischik.com