Boost logo

Boost :

From: Karl Nelson (kenelson_at_[hidden])
Date: 2000-11-01 14:08:58

> Using the tool itself is easy, it has a portable build system made in perl
> and is accessible from cvs site.
> Lamers could download precompiled binaries from web site.

Lamers would include all windows uses. Further, people such as myself
who doesn't want to see the monstrosity of a C++ library called Qt
grow would never use it. Call me a biased advocate but I am
boycotting all Qt using programs until they clean up their
use of keywords something which I have proven time and again is
quite possible. I don't like receiving mail from Qt library
users telling me that my library is broken because I choose to
use things they declared as keywords.

If you want others similar opinions read Nathan Myers post at

And believe me I have argued many times with KDE advocates
and been told that doing "#define emit" to make Qt code pretty
is good C++ practice and is not a keyword.

> I think the main reason of relative doxygen weakness in templates is that
> it is not commonly used with template libraries. I am sure boost usage of
> the tool will be a great "boost" of doxygen itself!

The reason it is bad at templates is that it is for documenting code
for a group which use a preprocessor rather than use the language
to get template functionality. See the MOC compiler which Qt

Why do we want to "boost" a project which uses a competing container
and algorithm library? Doxygen is written in Qt and does not use
STL. Qt is a commercial project and requires a license to compile
against for the Window architecture.

Shouldn't the documentation generator for boost be something which
uses C++ the way boost intends? If you are going to chose a doc
generator at least chose one which matches the type of C++ code we
wish to empower. What is the point of promoting a standard for
high quality C++ libraries when the documentation generator we
require doesn't even use those standards?



Boost list run by bdawes at, gregod at, cpdaniel at, john at