On Wednesday, August 22, 2001, at 03:48 , Chris Hecker wrote:
>
> http://research.microsoft.com/vault/
>
> The only difference seems to be the "key tracking" pre/post-condition 
> thing (and no type inference).  I wonder how well that works in 
> practice.  I also wonder if their generics, which are ML-style 
> polymorphics, not C++ style templates (but that have C++ template 
> syntax), will confuse C++ programmers.

ref<Foo<int>> has the same problem in Vault that it does in C++.

In addition to type inference, Caml has other things that Vault seems to 
be missing:

	+ Garbage collection
	+ Exceptions
	+ Classes (and class types)
	+ Streams (and the parser keyword)
	+ Polymorphic variants
	+ Parameter labels

I'm not sure I want to trade away all of those things to get invariants, 
though... invariants would be Really Qool to have.

The C syntax is *not* a big selling point for me.  Hey, I just got used 
to the idea that I can use "l'haricot" as a name for something 
bean-like, and you want me to give that up so I can worry about whether 
some '['/']' pair is delimiting an invariant signature or an array 
index?  Hmmm.

(Oh and there is the depressing fact that Vault isn't available to the 
public yet.  And when it is, it probably won't run on my platform...)


--
j h woodyatt <jhw@wetware.com>
"You're standing on sacred ground.  Many strange and wonderful
things have transpired right where you're standing."
                                               --unattributable
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr