I guess D really is becoming more popular. I'm hoping it'll eventually overtake C, much the same way C has overtaken COBOL. Guess I should dust off my "D Programming Language" book.

Regardless of the language used however, this is a great resource for learning about OS development.

Don't dust off your old one, read Andrei Alexandrescu's new one: http://www.amazon.com/exec/obidos/ASIN/0321635361/modecdesi-... (I believe that is Alexandrescu's affiliate link). It covers D2. As a C++ programmer who has exploited many corners of C++, I am pleased and impressed with the design of D2. (I can't say what was present in the first version.)

That was in fact the book I was referring to. I got it long enough ago that it's the version without an author on the front cover.

I think I'm coming from the same place as you. I've explored enough C++ to know its many pitfalls, and I'm impressed with some of the things they've done with D2.

Then that copy is 1 out of 1830 (http://erdani.com/). Hold on to it!

D is a very interesting language with a lot of good ideas in it... but that is also part of its problem.

no kidding. trying to make a better c++ is like trying to make a better cvs. just don't bother.

I think there is a need for a systems programming language with rich abstraction support. That is the niche C++ fills right now, despite its warts. I was skeptical of D, but reading through the book I linked in a sibling post, I think they have made some good design decisions. If you don't need to do systems programming with high-level abstractions, then you don't need D or C++.

My whimsical hope: that an OS distro based on the XOmB exokernel will be called "Exosystem, eXperimental Operating system-XOmB" or "Exs XO-XOmB" for short.

Not sure if that's worse or the "GNU Hurd" pair of pairs of mutually recursive acronyms. [1]

[1] http://www.gnu.org/software/hurd/hurd/what_is_the_gnu_hurd/o...

Actually, the point is to make a meta-reference to the Scott Pilgrim reference.