I was at BarCamp in Dublin and I enjoyed Sean O'Sullivan's "look at the Voice 2.0 landscape" (talk slides here). It was a pitch for their MySay service and a very good pitch as I just wouldn't get it without Sean's overview. Sean said that the guys he works with in Rococo Software used to laugh at each new Voice 2.0 service mentioned in the talk until the ridiculous services started to get thousands of users. Then they were sorry they hadn't done it as most of the (worldwide) services have a low cost of entry. Basically it's about convenience. It doesn't matter if it's possible to do something - if it takes any effort at all it will slow market adoption.

  1. Plugging a headset into the back of a PC is what holds back Skype and other PC2PC telephony. This is a really big deal apparently and is why web-activated telephony systems that use normal phones like Jajah have higher margins. In my case I really only use Skype for instant messaging - so he has a point.
  2. txting is fine but it's easier to just say it to a voicemail-like system and let the voicemail-like system take care of delivering the message - especially if it's practically the same price as a txting anyway.
SayNow seems to be the breakthrough service where you can see voicemail-like systems replacing txting. It has the tagline "let your friends and fans hear from you!". Voice is much more personal to fans (fanatics) and it seems they love to hear "Hi I'm Robbie and I'm on the bus. Hello Dublin". VOIP technology makes the calls cheap or free. I can see why these services will flourish now.

One interesting point was that they used Rails, MySql, Linux and Asterisk (an Open Source PBX and telephony toolkit) . They had been using the technology for about a year and he was always expecting one of his senior guys to come and have "The Talk". The talk would say why the technology was fundamentally flawed in some way and they would have to start again from scratch again with Java or somewhat. Thankfully this talk hasn't happened yet - for what was a Java house it's great to see that they can take a leap into the unknown. The convenience of their tools makes it an easy to get to a point where scalability may be a problem - a problem he'd like to have according to Sean.

Getting an API to make phones ring is also a pretty big deal according to Sean. He said they use the Dublin-based Glantel to achieve this. Looking at the Glantel website you wouldn't think they provided this service but maybe it's because they know the Rococo guys that they can do it for them.

It's pretty cool stuff even if nobody providing this type of service has a clear idea of when they will break even.