Hi Makarius,

|    * Standard ML with its no-sense semantics

This reminds me that last time I looked at SML (which was admittedly a
long time ago) the formal semantics left integer arithmetic almost
completely undefined, and at least one implementation used machine
arithmetic. Has that situation now changed?

I agree that mutable strings are indeed a gross misfeature of OCaml. I
think even Xavier might concede as much; at least, last time I whinged
to him about it he didn't actually demur but just explained how the
design came about. I have sometimes thought about using my own pet
string type in HOL Light, as Mark has done in HOL Zero, but resisted
the slight uglification of the code, an example of exactly the kind of
dilemma between simplicity and security that you mentioned.

Note that F# has immutable strings and is otherwise pretty close to
core OCaml, and it seems to be getting quite popular. I'd use it more
myself if it were a bit more Linux-friendly, and I have the idea of
porting HOL Light to it there in the back of my mind somewhere. I
think the code for my theorem proving book would be an easier first
step, though.

John.

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
hol-info mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/hol-info

Reply via email to