The JavaPolis Birds-Of-a-Feather presentation of Mobicents was quite fun. The room was full and there were a few standing people. The audience consisted mostly of developers with explicit interest in VoIP.
The discussion started with a quick poll. It turned out that 100% of the attendees were familiar with VoIP but none of them has written a VoIP application. This was no surprise to me. The first few slides of my presentation explained why I think most developers are not writing their own VoIP code. Several people in the audience shared their own thoughts on this paradox. It boiled down to two fundamental problems - 1) Until recently there was no widely available open source platform for developers to play with; 2) Unlike the Web, there is no truly public VoIP network, such that developers can deploy a new VoIP app and have anyone try it easily. Instead there are a number of VoIP islands such as Skype, Google Talk, and Yahoo! Messenger.
Then went into several examples of new kinds of communications services. Every had a chance to voice an opinion and add new ideas to the examples. It turned into a lively brainstorming session that made everyone realize that there are a lot of interestign applications that can be implemented when voice, video, mobility, location and web are mixed together.
The conversation was so involved that we went over time and had to leave the room because the next presenter walked in. The offline discussions we also pretty interesting. A number of folks wanted to know about the Mobicents roadmap and where they could help.
We also talked about SLEE best practices and architecture blue prints for complex applications such as call centers.
Take a look at the presentation and feel free to post your comment on this blog.
Mobicents-JavaPolis-2005.pdf
A good experience overall. The JSLEE community is growing out of the Early Adopters phase.
Ivelin
No comments:
Post a Comment