i created a chat roulette , should it go open source ?

[paranoio]

Hello my name is Luis

i have been a red5 enthusiast for a long time , i have created several systems using red5 and as3,
a year ago  i created a video conference app like chatroulette.com just for fun and forgot about it because i was working in other projects
so i would like to see this app to be published and actively used thats why im asking for the help of this community with
advice and maybe contributors.

Never took the time to create a good video but you can see some test here :
http://www.youtube.com/watch?v=4QSolluxx-A
http://www.youtube.com/watch?v=6PZ31IUE4ys    (first test)

First of all i havent decided if i should share this project so im asking for your advice.

 i could open  the source the code (have no problem with that) but dont know if that would
make this project work,  i was thinking in distributing only the red5 code and make it communicate with several red5 servers somehow
like a cluster (dont know if its posible) with the idea of always having users in the system no matter where the swf is installed (web servers , desktop apps or mobile devices).

So for this app to work it will need several red5 servers ,development and  maybe a way to monetize or get a sponsor so it can grow ,
the way i see it is more a bussiness/collaboration challenge than a development challenge.

If we can get enougth contributors and a good plan to monetize the system this could work , otherwise it will be a waste of time, so
im asking for your advice since i've never started a project like this and have no idea if its a good idea,
if this work it could be a cool toy to meet people over the world :)

By the way some of you may remember chatroulette had a problem of having a lot of weird users making a gross experience so to solve this i integrated the system with facebook
so users need to use their facebook account to access , this is somehow a way of moderation (since others will know who they are talking to).

Let me know what you think .

Cheers from méxico.

[Julien Devouassoud]

In my opinion if your code is just resting in the dust at the moment, going open source cannot do any bad to it.

It won't naturaly boost your project though.

 Only you and eventual close collaborators can boost it to the point people get interrested in it an even then, unless you put some sort of API in place or advanced a lot on the interresting parts like the cluster (look into Ehcache for that by the way, it's a fairly simple way to do it) people won't collaborate to the point that your project get "massCoded". You'll get help on testing for sure: plenty of people reporting bugs, some will fix them, a few might even improve the system, one good soul could add a feature.

Well this is a pretty pessimistic view I'm giving you here, and (i should have said at the start) I have zero experience with putting an app opensource. 

But from what i can see around and even with this great tool that is red5, the application is maintained mainly by the committers (correct me if i'm wrong committers).

I'm not revising this message i might delete it (it's such a vague assumption)

good luck in any case.

--
 
 
 

[paranoio]

Hello

sorry to take so long in answer back, i was waiting for more replies.

I guess you are right the only person maintaining the code will be me and some people testing it , 

the problem is i dont have too much time or resouces to start this project 

its not that there are too much to develop or maintain more like finding people sharing red5 server bandwidth.

And get enough users to make the project attractive to advertisers then give it some more work in the code 

and that should do the trick but i willl wait for more opinions  :)

[Dominick Accattato]

Paranoio,

Sometimes it's difficult to know whether a project will succeed in the open source world or not. Sometimes a niche market surfaces where there's money to be made and sometimes there are no downloads and just crickets. I think the projects that do extremely well hit the road with plenty of great documentation and good use cases... which excludes Red5 since there's not really any great documentation out there, just a smart community that has a good background in media server development. Red5 was lucky cause there was a need for the server, however most projects won't get touched if there isn't good documentation from the outset. The other thing you'll have to think about is how much you plan on supporting your community. At first it will be quite intense and you'll be giving a lot of your spare time in order to get others up and running. If you don't plan on helping others get up and running, the project might just sit there. 

The other thing I'd say is that chatroullette was a short phenomena and people would probably use your project for something different, but you probably already know that :). 

I do however think that you'll have people who are looking for something close or similar and you could find the best use of the code by open sourcing it. 

Anyways, good luck with your decision.

--
 
 
 

--
Dominick Accattato

Red5 Co-Project Manager & Engineer

---

http://www.infrared5.com/
http://code.google.com/p/red5/ 
http://code.google.com/p/jedai/ 

[blue °]

I really apreciate your time and advices on this subject,

i agree about the short phenomena and also about inverting time on the project otherwise it could end in a ghost project ,

any way just wanted to tell you i got some proposals after posting my question in this mail list and decided to not open source for now since we are going to invest money and development and see we if we can get this project to a commercial solution.

thanks again  :)