From: David Abrahams (dave_at_[hidden])
Date: 2005-06-06 07:06:24
"Brian Braatz" <brianb_at_[hidden]> writes:
> boost-bounces_at_[hidden] wrote:
>> On Behalf Of David Abrahams
>> > I have read the MPL book now FIVE TIMES cover to cover. On the fith
>> > pass, I "saw it". (the mpl::for_each<> is what kicked it in for me)
>> Wow, I applaud your fortitude and apologize for writing something that
>> took you so long to "see."
> [Brian Braatz Writes:]
> I believe the MPL book is an excellent body of work.
> The difficulty is
> 1- I am either VERY smart about a subject or it is a black box. there
> seems to be little middle ground (this is learning style I think)
> 2- Metaprogramming is "training your brain" to think differently.
> Honestly with the subject matter being so DIFFERENT from what us normal
> blokes are used to, I am not sure if I could even suggest a better way
> to present it. Being able to write compile time code is a skill that
> belongs to people who have the commitment to struggle through and learn
Well, it shouldn't be, really. If you could process types and other
compile-time data with regular procedural C++ code it would not seem
so different, I'm sure. The problem is that you not only have to get
used to the idea of writing code that processes the program during its
compilation phase, but you also have to adjust to this strange
computer language that uses class template instantiations to implement
> the TWISTED thing, is as I am getting more and more, it feels natural.
Then the book is doing its job.
> Also I think I am wearing out my copy of the book- I continually go back
> and re-read sections and chapters.
> THANKS to both you and Aleksey for birthing that sucker. Lots of good
> information in there.
> TO ANYONE READING THIS BOOK:
> DON'T GIVE UP. KEEP AT IT. YOU WILL GET IT. You just have to read it
> enough that you begin to bleed through your eyes. When that happens, you
> know you are close......
Aww, Brian, I'm not sure that's helping. Add a smiley at least!
-- Dave Abrahams Boost Consulting www.boost-consulting.com
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk