Re: [RANT] What's up ???

[Matt Slot <fprefect@AmbrosiaSW.com>]

 > BUT DO NOT JUST STAY QUIET!!
 > I can't help it - If I don't get any feedback on this issue I have to
 > assume that you're either dead or not imterested anymore and thus I have
 > to continue PP as my personal little project!

Here's my feedback as a lurker:

While open source projects are based on teamwork of alot of folks, you'll
find that most of them are *driven* by one or two people who make the hard
design decisions, establish the framework, and implement the first pass at
functionality. Only then do they draw "lots" of helpers and users.

The mailing list has been very active lately, with discussion about coding
conventions and ppFile implementation details. This is the first time I've
seen this much work on this project since I began watching it over 2 years
ago, and I'm more and more impressed.

I think that both Christian and Bjarke are actually making good progress
cleaning up and giving the project good direction -- they have the time
and the vision. It doesn't seem that anyone else has the time or interest
yet to work on the project. This is unfortunate but par for the course.

There may be other people actively working on the other portions, but with
so little discussion, they sound like they are languishing. I think that
ppFile and ppNet, aka Zombie, are probably the most functional aspects of
the project.

The progress on ppFile is a good way to shake the bugs out of the design
and development process -- it's a relatively benign and straightforward
task. As ppFile stabilizes it's time to start focusing on other, more game-
oriented, libraries.

Decisions need to be made about which 3rd party tools (Crystal Space,
libGGI, SDL) the project will adopt and build upon -- so that ppFile can
be labeled "good enough" and people can start working on something else.

Personally, I think SDL <http://www.devolution.com/~slouken/projects/SDL/>
is a very compelling low-level layer for game and multimedia development.
It provides a foundation for graphics, sounds, threads, timers, and more.
It's primarily implemented for Linux, but Sam has done an excellent job at
making it portable enough that Win32 and even MacOS ports are progressing.

Since PPlay already implements high-level services like file and network,
I'd strongly consider leveraging the SDL services and building a high level
C++ framework on top of it. SDL already has a modest following, and has
shipped as part of several successful Linux ports.

In addition, there are plenty of opportunities for building on SDL: image
loaders, game engines, etc. Recently, several folks on the SDL list have
asked design questions about high-level services, like game engines, that
are appropriate to PPlay.

Basically, for the first time I feel that PPlay may be worth using at some
point. You've got alot of momentum, and you should run with it. Now that
ppFile is functional, move on and get something else working instead of
optimizing and rewriting.

People aren't likely to start volunteering until they have something basic
that works. Once you have a workable demo (graphics, sounds, file, and
network), then you'll see people start taking notice. Of course, using SDL
will earn you a captive audience from the start. It's not just a design
decision, but a political one as well.

Matt


/* Matt Slot, Bitwise Operator * One box, two box, yellow box, blue box. *
 *  <fprefect@ambrosiasw.com>  * <http://www.ambrosiasw.com/~fprefect/>  */