> Did you really write your whole server code in Erlang?
Yes. I started with Lisp but realized I didn't know how to scale it. Then I tried Erlang, thought it scales automatically and finally realized that it's still a lot of work and careful analysis of your architecture.
I wrote another version of my poker server in Lisp in the past few weeks and it's almost complete. I know how to scale it now but see no point in pursuing it. I wrote it as a way of relieving my frustration with Erlang, a way of moving forward, even by taking a step back.
> Your intelligence scares me a little bit.
Thanks! If was as smart as you think, I would have been rich by the age of 30 and I'm not.
> First off, can you elaborate on what these "social network" features are?
I don't know, actually, I haven't given it much thought. I did google for poker social network and found that existing sites are forums without any ability to play poker.
I see my site as a friendly place to play poker, one you can customize and tailor to your needs and wants.
> 1. Why will your poker program be unique?
Beats me. It's not unique at all. It didn't have a poker client for a couple of years, I'm just finishing up tournaments and there's only Texas Hold'em at the moment.
Once upon a time, a sportsbook and casino in the Caribbean commissioned poker software. I was running an offshore development shop from my apartment at the time (NYC and then Atlanta). I outsourced development to Russia. The sportsbook fired their top management because their software could not handle the load during football season.
The sportsbook then dropped custom developed software entirely and went with off-the-shelf solutions. That's how I ended up with buggy PC-based poker software written in Delphi. I'm still kicking myself not not forcing development in C++. Devs said it would cost much more (Delphi specialists them) and I capitulated.
I thought I could do better myself and wrote OpenPoker.
> 2. How will you achieve critical mass (what's your distribution channel/marketing)?
Word of mouth and slow growth.
> 3. How will you stop your competitors from stealing your idea and implementing them with their existing user base?
There aren't that many scalable poker servers to go around. Also, most of the competitors are huge real money poker servers. I looked at gPokr but I think I can do better.
> 1. There are plenty of people trying to solve the multiplayer problem (my company for example)
I'm available for consulting.
> and we would love to buy some sort of license to custom code written to handle generic server requests.
What are generic server requests? Are they binary? JSON? UDP? TCP?
> Then you can package your Poker software on top of that as a "demo". But if you go this route, you really need to redo your server in a more traditional language, Erlang is fun for show and tell, but not very practical for the masses.
What is a more traditional language? Python? Java? C? C++?
> I would recommend making a Ning for Poker Sites. There are so many poker sites out there right now that it doesn't make sense to try and compete with all of them.
Great idea, thanks! Hosted white-label poker.