...a waste of time. If you know C already, then just start using gcc - it's a perfectly good C compiler. And if you want to fart around with the C++ extensions, that'd be fine. But cross platform apps in C - much less C++ - just isn't worth the trouble for GUI intensive apps. Python, Ruby or Perl are far better ways to go. And no matter which way you turn, you'll probably have to work with embedded JavaScript.

Of course, that begs the question. What do you want to do with the language. Cross platform is a secondary, if not tertiary, concern. If it's your primary concern, then you're best off using a language that runs atop a VM, and let the language designers worry about porting it to umpteen different environments.

#1 Something I can use to create programs with that doesn't run in a VM that bogs it down or use an Interpriter or whatever. I want something that runs fast.

C programs can be tuned to run vewy vewy fast because plain old ansi-C sits just slightly above the hardware abstraction. Being that close, there's a wealth of techniques to tune code to take maximum advantage of the underlying hardware. That blazing low level speed eventually runs into a problem. Where's the libraries to write C code that allows the app to build applications that have the users have the same experience on the variuous platforms. How many C++ progs do you know that succeeded in writing applications that require an intensive graphical user interfaces and having that resulting cross-platform user interface be able to compete with native only applications.

And as for VM's, I'd suggest that VB code and forms was never really a compiled environment. Instead, the forms were data files and prior to VB6, the code was compiled to intermediate code sets. If you felt that VB could handle the requiired speed, I'd go out on a limb and suggest Python and Ruby can write applications that a competitive in user response times.

#2 Something I can adapt to. I already have C Language skills, and I am a bit rusty with them. I am trying to write simple C programs and am having difficulty, but I am relearning.

Best to get back on the C horse and get comfortable before the level of knowledge you have to know for C++ requires you to go directly into the heart of the dragon. In my opinion, the number of C & C++ jobs is a slow to no-growth market - unless you're in instrumentation. Means that there are a lot of smart C++ gurus out there that aren't gonna appreciate new faces walking in on their craft.

#3 Unlike Bryce, etc, I actually want to learn OOP.

Don't forget you'll have to get up to speed on templates and higher programming in general. It's possible to have the discipline to use C++ as an OO language. I just think that it is the most hostile environment for learning the principles of object oriented programming.

#4 I want something that can be ported to Linux, OpenBSD, Mac OSX, OS/2, AmigaOS, or whatever I want to port the app to. Sorry CRC, Delphi/Kylix won't do that.

Are you targeting the native API's for each of the platforms that you cite? Or to put it in more specific terms, show me some successful applications that are capable of running on Windows, Linux, Mac OSX, OS/2, AmigaOS. Most of the successful ones I've seen are written with unix pipes being the primary mode of input/output, with very little VB type GUI forms.

#5 I want something that open source developers are already using. C/C++ seems to be very popular using gcc and libraries developed for it.

C/C++ are very popular for writing Linux Apps. C/C++ is also very popular for writing Windows Apps. However, the two camps don't often meet, and you get into some major projects when you want to cross the platform gap.

#6 I want something with an IDE, so I can drop and drag controls on forms and have it generate and modifiy code for me. Then I can fill in the details to complete it.

Then you not only need to know the language, you need to know the graphic/windowing environment/library that you will be choosing. Lot's of choices (gtk, gnome, etc...). But that leaves you with additional problems in that you have to not only port your application across, but now you come to depend on libraries that were written where cross-platform was not necessarily a goal.

#7 It has to be virtually free, I cannot afford much. Even the $10USD Visual Studio.Net trial CD got my wife mad at me.

Python, Ruby, Perl, and JavaScript are about as free as free can be. I prefer python (not sure if admin still does), Jim Weirich prefers Ruby, and Tilly monks with Perl. All free, all expanding. And most important they are rapid prototyping languages - a feat of VB that will not be replaced by C++.

#8 Has to have a lot of free web site tutorials, howtos, faqs, etc written on it.

C++ is a mammoth language that requires you to be fully cognizant of the dangers. You will have to read a pile of books and trundle through some difficult projects prior to being able to do even some of the most basic things. Don't forget the "Effective" books as they are chalk full of problems which must be avoided.

Python, Ruby, and Perl have an equal amount of material avaiilable, with freely available libraries to do a lot of cool stuff.

#9 Has to be something I am comfortable using.

Having done a bit of C, with a tad of C++ trhown in for good measure, I can't say that the word comfortable and C++ are synonomous.

#10 Has to be a language that is marketable to companies that potentially will hire me. In other words, organizations have to use it a lot, and will offer me jobs if I know it.

C & C++ will continue to have a niche in the markets, but this will become dominated by engineering. Business applications are steering further and further from these languages.

Means that you can still make money with C & C++, but with the number of jobs not significantly expanding, I foresee that the C/C++ world will become dominated by C gurus that have been eating this stuff for lunch since they were in diapers. Entry level C/C++ is perhaps a thing of the past - unless you live in a foreign country.