[Date Prev][Date Next][Thread Prev][Thread Next]
- Subject: Re: [ANN] Bou - a Lua-based Build System
- From: KHMan <keinhong@...>
- Date: Sat, 09 Feb 2008 17:44:57 +0800
Brandon Van Every wrote:
> On Feb 9, 2008 4:07 AM, Wesley Smith <firstname.lastname@example.org> wrote:
>> > I take your point about NIH (one of my new year resolutions is not to
>>>> reinvent wheels unnecessarily) but there is a serious purpose in
>>>> experimentation - it's called 'research'.
>>> Real researchers study prior art and justify why they're doing better
>>> than it. They don't just hack because it's there and it's fun.
>> I do. Nothing like kicking some tires to see how it really runs.
> So what does that mean for a build system? Do you just start writing
> one in Lua, or do you actually try writing non-trivial stuff in other
> build systems first? Do you actually read and ask questions about
> other people's case uses? Try to learn from others instead of just
> doing it all yourself?
This thread I am reading is interesting, but a bit 'noisy'. I like
the Lua list because it is casual and accommodating with
exploratory stuff, even as the core developers are conservative,
careful and deliberate.
I prefer a diversity of views. Yes, there are some work or
projects done with much rigor and excellence. There are also
casual projects, and there are not-so-serious projects. We should
respect and give space to all of this. Sure, there are a lot of
build systems around. Let them sprout. It's like the adage that
almost every programmer (or C programmer?) would have tried to
write a text editor at least once in their lives. Ditto with Lua
and build systems, just look at LuaForge.
So CMake is great. Great. Let others sprout and let the market
decide. Many will stay as casual projects obviously. The market
can't really support too many ubiquitous build systems anyway. So
let us all allow for a diversity of views and keep things cool.
Kein-Hong Man (esq.)
Kuala Lumpur, Malaysia