Thanks! I have several motivations to work on Waverous, one of which
is to be
better at C/C++.
My initial motivation to get it to compile as C++ was I wanted to add
a new
type (a SQLite object). After reading the doc on how to add new types
in
the LambdaMOO source, I thought (perhaps naively), "wouldn't
subclassing be
easier?"
I started hanging out on the Waterpoint moo for a while, where the
LambdaMOO
server's maintainers hang; and to the question of C++ one of the
maintainers
told me, "Some of the server's internals could be expressed better in C
++."
I think this was about 2008... or 2009? I spent a few weekends
slogging through
the code casting returns left and right, renaming variables called
"this" and
"new," and finally one day got it to compile via g++.
I belong to a moo called Avalon, and they had some patches I wanted to
add;
that done, I then proceeded to do nothing for a very long time!
But there are some interesting problems to solve. I found out that GCC
tried
to compile it for 64 bit architectures on OS X, so it was crashing
when you
(the wizard) examined yourself ("@examine me"). So the code is not 64
bit
safe, and that would be neat to do.
I started cracking the Stevens networking book, and the thought of
IPv6 support
came to mind, so I'd like to try that as well.
Oh, and SSL and i18n support, but that's been done by many people
already so I
have, at least, some examples I can follow.
I come from years of doing LAMP stack web applications and I want to
do something
closer to the metal. I've been learning GDB, I built a new Linux box,
and I've
been teaching myself the autotools the past few days (being newly
unemployed,
I have time to do this stuff now :o)
In the long term my goal is to try to build a community around
Waverous. When
I founded the PhpWiki project back in 1999, I had three stated goals:
* It should be easy to use for end users
* It should be easy to administer by system administrators
* It should be easy to hack on by programmers
And for Waverous my plan is the same. These are, really, the biggest
goals of all. To have regular releases, to have a code base that is
perpetually moving towards something better, to have a community of
friendly developers, to have good documentation... these are things
that are really worthwhile.
~swain
p.s. This isn't the first time I've heard bad things about FUP, and
that's
something I'm willing to change!