[anon4lol]

I've been developing C/C++ for over 20 years on everything from large scale high-performance clusters to 8 bit embedded devices. Most of my career I have been consulting and I've had the opportunity to work in almost every domain, and I've had to work with a lot of other developers.

My thoughts are this guy is type of the developer I hate to work with most. The style and naming Windows and Borland C++ specific, so this probably grew organically from Win3.1/NT. This probably pre-dates STL, but holy cow. Given how he misuses some terminology, he is probably a self-taught programmer from days of old.

Generally, when I see that a developer has decided to use their own typdefs to represent intrinsic data types, it's a bad sign. However, what I see that this guy created his OWN class as a replacement for bool, I know I'm going to be rewriting a lot of code.

There is no way I would use this library, regardless of the terms.

[DeanRoddey]

It's designed to be portable. There needs to be well defined types that each per-platform driver can define such that they always are the same.

It would have been ugly and asymmetric to have some done that way and some not. So all fundamental types are defined as part of the library.

[sytelus]

There is a huge difference in actually being portable and “designed to be portable”. I think library’s usefulness would increase many fold if former was the case.

[DeanCIDLib]

Well, that's one of the reasons for open sourcing it. I'm no a Linux guy. But I know the issues very well, having worked on a lot of portable code before. And there's an existing platform driver for Linux already there, it's just very old and needs to be caught up.