Re: Type System as Design Tool [Was: We do not use C++ exceptions]
* Thant Tessman:
Alf P. Steinbach wrote:
* thant.tessman@gmail.com:
Anytime anyone uses 'new' or 'delete' in C++, it is because they are
not 'happy' with deterministic lifetimes.
You're confusing "nested lifetimes" and "deterministic lifetimes". [...]
Mathias Gaunard does indeed seem to use the term "deterministic
lifetime" as analogous to "nested lifetime" with the qualification that
it is acceptable to transfer ownership via some form of encapsulation
with value semantics such as a smart pointer (if I can paraphrase).
But it is exactly this "transfer of ownership" that I'm trying to draw
attention to. You can't pretend lifetimes are still 'deterministic'
merely because you packaged up the 'new' and 'delete' in a smart
pointer. Any criticism of true language-level garbage collection in the
non-deterministic sense applies just as equally to any program that
finds itself resorting to the use of smart pointers.
You're saying that any criticism of non-deterministic garbage collection applies
equally to use of smart pointers.
That's a statement that contains some technical sounding words.
And it is a statement that lacks any technical meaning.
It is completely meaningless mumbo jumbo.
Such meaningless statements, together with the refusal to exemplify claims (e.g.
in the article I responded to), together with the size of this thread, not to
mention your article's reference to something written by a well known Norwegian
Usenet provocateur (there are special collections of his rants), in response to
me who is a Norwegian, well, that makes me very very sure that you're trolling.
Cheers & bye (for now),
- Alf
--
Due to hosting requirements I need visits to [http://alfps.izfree.com/].
No ads, and there is some C++ stuff! :-) Just going there is good. Linking
to it is even better! Thanks in advance!
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]