what web programming language?

What I mean is that there is a lot of server overhead for streaming video to multiple users... if you start this on a single server you will need to move to multi-server if it takes off. So the question then becomes are you going to configure your SOFTWARE to do the work in figuring out what server a person should be connected to, or are you just going to do clustered servers and let the HARDWARE grow.

Either way will work, but it changes how you develop the software initially. If you want to just keep adding server power without changing your code, you'll need to go with a back-end (and a host) that can support clustering. If you don't want to bother with that, then your developer needs to write the application with a little bit of "future planning" in that it may one day be in a non-clustered multi-server environment. This is pretty easy to do as long as you do it initially, designing your database with a table for servers by name and IP address (even if there is only one initially) and then each server can access the same database. The application would simply connect to the central database, pull the name of a server to connect to from a list in the database, then connect all clients in that conference to that server.

Really all you need to ask your developer is how would future-proof it for a multi-server environment down the road. If he stares at you like you have five heads, he's not the developer for you. If he has an answer, he'll know how to set it up.

You could run into that problem half way through the project if the developer you hire flakes out too, then you have to pay someone else to pick up the ball or start over.

In this stage of the game, the likelihood of you failing first is greater. I don't mean that to sound rude or to come across as an insult, I mean you should approach the situation objectively:

1. They already have a working model.
2. They already have a hosting environment capable of supporting the business.
3. They already have customers.

You have none of those things yet. You'll have to spend your money just to get #1 and #2 and you won't even have a single customer after that expense. So what I'm saying is skip those steps, piggyback off of what they already built, and add your development budget to your marketing budget.

If you're unsuccessful, you'll know you gave it a good run and didn't waste your money developing a technology you can't monetize. And if you ARE successful, make your first priority getting something custom written so you aren't relying on someone else's platform for the long haul.

Walk before you run.

Look at it this way ... even if their licensing fees are structured in such a way where you make zero dollars per customer, you're still GETTING the customer - and you know what you'll be able to bill once you migrate them to a platform you own without those licensing fees. It also gives you a clear vision of what your budget needs to be, and how your platform will have to work to support the number of users you have, so when you do build out your own you're way ahead of where you are now and will have the benefit of "instant revenue" once you throw that switch.