From: jstern@... (Josh Stern)
Date: 2001-01-18T18:10:02+09:00
Subject: [ruby-talk:9499] Re: 101 Misconceptions About Dynamic Languages

Christian <christians@syd.microforte.com.au> wrote:

>> Josh> The point of this part of the discussion was that there are
>practical
>> consequences to the lack of type safety in C++, for both "expert"
>> and novices.  Your claim that one can avoid that by using containers,
>> references, and no casts, is not true in theory or in practice.

>I don't understand what you mean by "lack of type safety in C++" (which was
>the statement that started this entire thread).

The terms 'type safe' and 'type safety' are commonly used in Computer
Science, but they don't have a formal meaning independent of context.
In the context of some particular language and syntax checking
sytem, they can be given formalization by listing a set of sytem
errors and then reasoning about whether or not the language and
syntax checking system has a soundness property so that any
piece of code which checks out can be proven not to be at risk
for triggering one of these errors during execution.
In the context of C++ it's understood that the syntax checking
system is the compiler and things like dereferencing an
invalid pointer are system errors - everything which is
described as "undefined" system behavior in the ISO standard
is a system error.  For a language like Ruby, the analog
is code that is parsed without complaint by the interpreter
and the system errors would presumably include things
like triggering "undefined method" exceptions.


>> In Ruby and other dynamic languages, there are not separate stages.
>> In C++, if one wants to use the full power of the C++ language to
>> do the meta-programming, then it would be a separate stage and it
>> would involve a system call to a compiler, which is not part of
>> the language.  Otherwise, one must work with unexpressive template
>> meta language.
>
>I take your point. I could argue that partial specialisation in C++ is not
>'unexpressive', except it would be a weak argument. The syntax can become
>obtuse to say the least.

Templates in C++ are very expressive and powerful for some purposes,
especially compared to things like cpp defined macros in C.
They are just very limited for meta-programming even if one
accepts the limitation that everything must be defined at
compile time.  There are various efforts to overcome this
with sophisticated pre-processing, for instance Open C++:

http://www.hlla.is.tsukuba.ac.jp/~chiba/openc++.html

-= Josh