Day one at Enterprise Connect had me moderating a panel titled "The Post PBX Platform: How Real, when and from whom?" At first glance, I wasn’t crazy about this topic and wondered how good a panel it would be, because the answer seemed a bit obvious to me--of course we’re in a post PBX era. However, after listening to the panelists, I've changed my mind and it's made me think a bit more about what the post PBX era would really look like.
The first question Fred Knight (he co-moderated) asked was "are we in the post PBX era?" and some of the panelists answered yes and others no. My answer is no and my answer is based more on the way this industry acts versus the technology. Typical deployment model for most organizations is to remove a PBX and replace everyone with an IP PBX. The company might use their WAN to do intra-company calling but, in all likelihood, the IP PBX will hand off to a gateway to connect to some traditional trunks, meaning the value of the VoIP network is internal only.
To move past this, vendors and buyers need to stop thinking about making "calls" and thinking about managing "sessions". As part of that, when the move to VoIP is made, the infrastructure needs to adapt as well. Having this model of putting call control locally in every location is an outdated deployment model and needs to change. Organizations should think of their next generation communications platforms as data center applications that can have the functionality distributed over the WAN. Additionally, if the value of VoIP is doing voice at the IP layer, then the IP connectivity should be extended all the way to the cloud and not terminate at the edge of the enterprise, as it is today. This would mean usage of SIP trunking (this is another term I really don’t like) instead of traditional POTS lines.
Of course, the architecture is just part of what the post PBX era would look like. Years from now, I think the architecture will look totally different. On the panel, I used the analogy of the shift from mainframe computing to client server computing. If we had run this panel 30 years ago and asked what the post mainframe era would look like, we would have had people deploying PCs and then running 3270 emulators to make their end points look like mainframe terminals. This is a method of bridging the old world to the new world.
We're doing some of that in telephony right now. Decades later in computing, the answer to the post mainframe era is that there is no analogy to the mainframe. Every large organization has multiple application and computing strategies. A company may choose to build some of their applications on WebSphere, some on .NET, use cloud based applications and a handful of niche ones. Generally the applications that need to pass data to one another are either built on the same development platform or some middleware is used as a higher abstraction layer to do the translation between applications.
Smaller companies will avoid many of these hassles by using a single vendor where everything is integrated or push everything to the cloud. So the answer to what does the post mainframe era look like is that it looks completely different than what we had before. We don't buy all our applications or compute infrastructure from one company, nor do we integrate EVERYTHING together. We integrate where we need to.
I think communications will evolve down a similar path. Decades from now organizations will have multiple communications platforms and build some functionality on each, use some cloud based applications and a handful of purpose built applications. Smaller companies will choose the suite approach and settle on a single vendor for ease of implementation or just push everything to the cloud. Where communications need to integrated to one another, "communications middleware" will exist to be that required higher abstraction layer. Avaya's ACE platform is actually a good example of communications middleware that could be used to tie communications functionality together. I think this functionality could live in the cloud as well and would be a great value add for a communications-as-a-service provider.
So, back to the original question, are we in the post PBX era? No, not even close and the move to this era is dependant on the UC vendors willing to be bold enough to push the industry towards it. One of the panelists (that I won’t name) said they agreed with my comments but keep one foot in the old world because everyone else does. Well if everyone else does, then how do we ever break out of where we are today? Agreed there is some risk but the upside is we’ll be able to do so much more with communications than we have in the past.
