TT>
  > (I keep hearing about how bad the resource workshop is)
TT>

  Well, you hear wrong.  It's better than DLGEDIT, as most people who
  have tried both (as I have) will attest.  It's only problems are that
  syntactically incorrect resource files can confuse it sometimes.

  Since you only get them when you edit by hand ....

TT>
  > IBM C++ was one of the first to support templates and other advanced C++
  > constructs, on any platform.
TT>

  Wrong.  The honour of being first went (and still goes) to Metaware
  High C++, which is ahead of all of the other four.

  Incidentally, CSet++ 2.1 doesn't support Run-Time Type Information,
  whereas Borland C++ 1.5 for OS/2 does, so I suspect severe mis-use of
  the word "first" here.

TT>
  > Also, their class library is excellent.
TT>

  Yup.  I don't like it myself, but it's excellent despite that.  (-:

TT>
  >                                            I feel that an investment in
  > IBM C++ is protected.  I don't feel the same about anything from Borland.
TT>

  I don't think that Borland C++ for OS/2 is going away anytime soon. It
  also happens to be an excellent "first compiler" for OS/2, since it's
  cheap, familiar, and integrated. It was my first compiler for OS/2.

  Maybe it's not the compiler of choice for the hardened professional
  who wants an industrial strength optimiser and stricter ISO (but not
  stricter POSIX.1) compliance, but it is still a compiler suitable for
  a lot of people.

  Of course, the truly hardened OS/2 professional will be wanting to use
  a C++ compiler with DirectToSOM, and so will use Metaware High C++.  (-:

  As I keep saying, it's all a matter of pros and cons ...

  > JdeBP <
___
 X MegaMail 2.10 #0:
--- Maximus/2 2.01wb