In article KtM@world.std.com, tob@world.std.com (Tom O Breton) writes:
> You apparently base your argument on the ~a priori~ assumption that I
> must surely have miscalculated how complexity scales. I repeat, I've
> worked with more complex things. I don't think there's any further
> response neccessary.

Actually, I have based my argument on the experience of the majority of C++
implementors and many of the would-be implementors.  Until you join that
crew, your opinions on the complexity of C++ compilers aren't very well
grounded.

I suppose it's like having kids.  Parents try to tell you what it's like, but
you never really understand until you have them.

Sigh------

Mike Ball

In article 89M@world.std.com, tob@world.std.com (Tom O Breton) writes:
>
> I'm sorry but that is just obscurantism.

Not having a clue what you mean by that, I will drop this obviously unprofitable
discussion.

-Mike-

In article BIw@world.std.com, tob@world.std.com (Tom O Breton) writes:
> Sure, "C++ is bigger and more complex than your dinky little compiler",
> you may argue. I know that. I'm scaling to account for that, and it's
> still my distinct impression that I'm hearing enormous exaggeration of
> the difficulties. I add tougher stuff to bigger designs, and frankly I
> get impatient hearing how difficult it's supposed to be.

Unfortunately, you are talking about an order of magnitude difference in complexity,
and unless you have tried it, you really DON'T know how hard it is.  Many very
good compiler developers have tried and failed.  They too thought that they
could scale things, and they were wrong.  Some have tried and succeeded, all
that I have talked to greatly underestimated the difficulty.

It's not to say that it's impossible, but it's much closer to an Ada compiler
and environment than to, say a C or Pascal compiler operating by itself.

To discuss ways in which the language has become more complex, you will find much
of it in small details changed in the language as it was "clarified".  The explicit
extensions haven't been too bad, but they include RTTI, a variety of casts,
expansions in templates and exceptions, a bool type, namespaces, mutable, and
more.

There have also been ways in which the language was simplified, but somehow these
don't get the same press.

Then there's the large and almost entirely new library that was added.  Most of
it was clearly necessary in one form or another, though there are some items
that seem very specialized.

We won't know the value of all these changes for years.  The costs we can see
looming in front of us now.  This tends to color viewpoints.  My experience on
the committee indicates that implementors are trying to minimize changes that
would break their customers code, while also trying to avoid features that will have
costs for everyone but benefits only for a few.

Personally, I don't think that's bad.

Mike Ball
SunSoft Developer Products.

ball@cygany.Eng.Sun.COM (Mike Ball) writes:
> Unfortunately, you are talking about an order of magnitude difference in
> complexity, and unless you have tried it, you really DON'T know how hard
> it is.
> Many very good compiler developers have tried and failed.  They too
                                                             ^^^^^^^^
> thought that they could scale things, and they were wrong.
                    ^^^^^^^^^^^^^^^^^^ [ Please don't read things in.
  Just because you saw the word "scaling" does not justify an
  imputation of linearity.]
> Some have
> tried and succeeded, all that I have talked to greatly underestimated
> the difficulty.

You apparently base your argument on the ~a priori~ assumption that I
must surely have miscalculated how complexity scales. I repeat, I've
worked with more complex things. I don't think there's any further
response neccessary.

        Tom

ball@cygany.Eng.Sun.COM (Mike Ball) writes:
> Actually, I have based my argument on the experience of the majority of C++
> implementors and many of the would-be implementors.  Until you join that
> crew, your opinions on the complexity of C++ compilers aren't very well
> grounded.

I'm sorry but that is just obscurantism.

        Tom

--
finger me for how Tehomega is coming along (at tob@world.std.com)
Author of The Burning Tower (from TomBreton@delphi.com) (weekly in
rec.games.frp.archives)