Re: LuaBatteries.org ?

[Duncan Cross <duncan.cross@...>]

On Sat, Aug 28, 2010 at 5:25 PM, Patrick McCavery
<patrick@cakeandfile.com> wrote:
> I am glad that we have a Lua list again, thanks to all involved.
>
> It is my understanding that the Lua team is focused on the core language and
> generally tries to build mechanisms for programmers to work in their own
> way. I support this but there are also many calls for more libraries.
>
> I was wondering if other community members would like to start a new site
> specifically for creating libraries for the language.
>
> I know we already have LuaForge but I think that a new site with an explicit
> purpose might serve us all well. Perhaps we could do this in an open source
> way with Git repositories and such.
>
> --Patrick
>
>


Hi Patrick,

This is something that I've been thinking about a lot recently, the
problem of making "battery acquisition" less of a confusing chore in
the long run. The conclusion I've come to is that just trying to
encourage people to write more libraries isn't enough (though
obviously it is important) -- but what I think could have a strong
positive effect is if we had more of a culture of critical peer-review
around modules, and more of a sense that as module creators we are
working towards something collectively rather than all working in our
own silos.

Okay, so, I hope no-one misunderstands this. Just for fun I've written
a short mission statement for a fictional project that should
hopefully illustrate the kind of thing I have in mind. I have no
immediate plans to try to start this project, this is just for
discussion's sake:

(Also, I know the acronym is kind of silly.)


"The Lua Component Recommendation Network (LuaCoRN) is a web-based
project run by a volunteer group of enthusiastic Lua users who hope to
provide good quality guidance on Lua libraries/modules to anyone who
wants it without having to do a lot of their own research.

 "Our main goal is to provide an easy interface for comparing the
strengths, weaknesses and idiosynchrasies of the modules available for
solving a problem. This includes factual details like external
dependencies, platform constraints, and code license, as well as a
more subjective review of things like:
- Problem coverage. Are there any notable missing features, or signs
of unnecessary bloat/feature-creep?
- Style and quality of the API provided. Does using this module feel
very inconsistent with the official Lua modules?
- Developmental health of the project. Put bluntly, does it appear to be dead?

 "As well as information on individual modules, we also maintain lists
of libraries that the group believe are high quality and collectively
cover a good set of functionality, with a minimum of repetition. These
lists will have specific names like
'Recommended-Lua5.1-General-Win32-Permissive', meaning recommendations
for miscellaneous/general-purpose Lua 5.1 modules that are compatible
with the 32-bit Windows operating system and have a license at least
as permissive as Lua's own MIT License. (Hopefully, we will eventually
find a way to make these names a bit catchier.)

 "The LuaCoRN website's editorial policy may change in the future, but
for the moment, it does not offer module authors the ability to create
or maintain any information about their own projects. It also does not
offer Wikipedia-like open editing. Instead, we maintain all
information on the site ourselves. We feel collectively responsible
for all the site's content, and so are very careful and conservative
about what we put online. We will explain and defend our position on
the LuaCoRN mailing list, if necessary.

 "It is important to understand that LuaCoRN has no inherent
authority. Its recommendations are not officially backed by the core
Lua team, and in fact cannot be, due to aspects of the team's
philosophy about these things. Instead we hope to earn your confidence
by developing and maintaining a good reputation -- but we do not
assume it."


Again, I have *no* immediate plans to put this idea or something like
it into action. But I'd be interested to know what people think of it
in principle.

-Duncan