How are new C++ standards designed before they exist?  Are they purely
vaporware or does someone implement a quick a dirty compiler hack (say
on g++)?  If they are vaporware, how do people know if they will
actually work?

What compiler was used to implement the Standard Template Library?  Its
seems to to me that, until recently, there were no compilers that could
actually compile it (certainly not HP's cfront compiler).

This gets particularly interesting when one is reading an article in C++
Report which displays source code using C++ features which, according to
the author, no compiler yet implements.

Does this sound like a chicken before the egg question?

MikeB
---
[ comp.std.c++ is moderated.  To submit articles: Try just posting with your
                newsreader.  If that fails, use mailto:std-c++@ncar.ucar.edu
  comp.std.c++ FAQ: http://reality.sgi.com/austern/std-c++/faq.html
  Moderation policy: http://reality.sgi.com/austern/std-c++/policy.html
  Comments? mailto:std-c++-request@ncar.ucar.edu
]

In article <335E6ADF.B8E@mmm.com> mbresnahan1@mmm.com writes:
>How are new C++ standards designed before they exist?  Are they purely
>vaporware or does someone implement a quick a dirty compiler hack (say
>on g++)?

I'd rather say mentalware than vaporware.  Ditto for "lots of hard work"
over hack.  To say otherwise, is immensely trivializing what did or
didn't happen.  On your questions:
1) Did somebody implement everything first: no.
2) Were those features that were implementated totally implementated: no.
3) Was anything implemented first: yes, lots.
4) Did any of this go to far: probably.

>If they are vaporware, how do people know if they will actually work?

For the most part, lots of it was not things people picked out of hats
and just proposed live.  Most of it was actually discussed for _years_
first.  And not just discussed but proposed in depth and debated.

Was this picture perfect?  Nope.  But then, what is?
Even implementating if first _is not possible_!

>What compiler was used to implement the Standard Template Library?  Its
>seems to to me that, until recently, there were no compilers that could
>actually compile it (certainly not HP's cfront compiler).

"STL" is a work in progress spanning at least a decade.  So, on the
one hand, the answer is many.  On the other, given the literal absolute
latest specification: none.  Has this led to any problems?  Yes.
Would having done it some other way not have led to no problems?
I doubt it.

>This gets particularly interesting when one is reading an article in C++
>Report which displays source code using C++ features which, according to
>the author, no compiler yet implements.

Yep, it does seem a cautious ground to tred.
But it always is, even with compilers that might accept it.

>Does this sound like a chicken before the egg question?

Yes, but for instance, isn't that something you do when you
write your own programs??  Don't you iterate?  Do you always know
the answer to every question?  How do you design your app before
your app exists?

I'm left wondering what the real concern or implications of your
questions are.

- Greg
--
       Comeau Computing, 91-34 120th Street, Richmond Hill, NY, 11418-3214
               Producers of Comeau C++ 4.0 front-end pre-release
****WEB: http://www.comeaucomputing.com / Voice:718-945-0009 / Fax:718-441-2310
 Here:comeau@comeaucomputing.com / BIX:comeau or comeau@bix.com / CIS:72331,3421
---
[ comp.std.c++ is moderated.  To submit articles: Try just posting with your
                newsreader.  If that fails, use mailto:std-c++@ncar.ucar.edu
  comp.std.c++ FAQ: http://reality.sgi.com/austern/std-c++/faq.html
  Moderation policy: http://reality.sgi.com/austern/std-c++/policy.html
  Comments? mailto:std-c++-request@ncar.ucar.edu
]

Been there, doing that.

Implementing the standard is an interesting task. Our library provider
has had to write some code using the latest features that the compiler
they used did not yet support. Not an enviable position.

Our front-end supplier has kept us fairly close to the "current" C++
document.  We end up spending a bit of time debugging the library
implementation and the C++ document itself. We've found at least one
place where the library sections specify one thing and the language
sections forbids that construct.

Over the next 12 months, you should see more and more compilers move
closer to the "current" document. For one thing, the moving target is
not moving quite as fast as it had in the past. Compilers now have a
chance to catch up. 8^)

--
#include <std/disclaimer.h>              Kuck and Associates
David Nelson (david@kai.com)             1906 Fox Drive
http://www.kai.com/C_plus_plus           Champaign, IL   61820
KAI C++ - Cross Platform C++ Compiler    (217) 356-2288
---
[ comp.std.c++ is moderated.  To submit articles: Try just posting with your
                newsreader.  If that fails, use mailto:std-c++@ncar.ucar.edu
  comp.std.c++ FAQ: http://reality.sgi.com/austern/std-c++/faq.html
  Moderation policy: http://reality.sgi.com/austern/std-c++/policy.html
  Comments? mailto:std-c++-request@ncar.ucar.edu
]