[Icecast] Some ideas
pem at levillage.org
Wed Dec 29 09:04:31 UTC 2004
I'll try to explain my suggestion with my poor english.
I've got the feeling that icecast is full of potential but not growing
as fast as it can or simply not having the support it could have.
The greatest lack are the "open doors" for new coders, webdesigners,
translators, etc. Icecast is open: the sourcecode is open, the mailing
list is open, etc. But a skill is not enough. You need some kind of
structure, some kind of simple tasks to do.
Look at KDE. You can find many "How to help" pages on many sites with
such things :
- You're a coder? Welcome, go here and ask or look at this list of
things to do and begin to code.
- You're a webdesigner/PHP wizard? Welcome, go here and do that.
- You're a PR/marketing/journalist pple? Welcome, we need to promote our
- You're a happy KDE user? Go here and fill our form, we're glad to ear
Then, the guy who wants to help exactly knows where to go and what to do.
Just look at me. Unable to code something correct, not good enough to be
a standalone PHP coder but I can read english and german and french is
my native language. So what? A week after having the idea to help, I was
rereading some docs and working with the translation team.
So my suggestion is: lets stop icecast. KDE is better.
I know it is hard to have a clear roadmap, to communicate and explain
all the things going around icecast. But that's also a way to open
icecast. When I know what's going on, I know what I can do.
The lack of open doors can also be known and not that bad if you want
icecast to remain the fact of few people and keep it more or less to
your own and be sure to grab most of the fama/pride/money.
Here are my suggestions:
- building up a simple webpage with a "bounties" system. A user can fill
a form and ask for some specific feature. When a feature request already
exists, a new user can simply vote for it. At the moment, I have to
subscribe to the list and ask it in english. Such a system is a closed
door to many users. The icecast developpers could have a "froze"
feature. When they know that it's dangerous to work on some feature
because it's going to be implemented or that the code is moving too
fast, they froze the feature. When a coder see a feature or a bug he can
solve, he registers, subscribe to the feature, get the code, work on it,
give the clue to an icecast core team and that's it. He's now and the
Hall Of Fame Of Nice People. Having such small things to do is a good
way to begin. Plug coders on the main trunk is too complicated.
Feature, bugs, etc., I'm sure that the main icecast coders are aware of
some bugs but do not have the time to solve them. They could also fill
And, why not, some company could give some money for specific bounties.
Again, that's an open door.
- have an internationalization team. First: translate the docs, second,
translate the website, etc. That can be made without any specific tool.
We just need a "start" and coordination.
- have a "happy user" page with a form : why are you using icecast? how
many users are you serving? etc. Then, when a new user discovers icecast
he can know how strong and powerful the server is. When I first
discovered icecast I could not know if the server could handle 100, 200
ou 500 users. How is icecast reacting when 200 streams @ 128kb/s are
sended? Etc. That's easy to do.
I'm ready to help :)
More information about the Icecast