Jim Van Meggelen’s Blog

People hate talking to Bionic Betty. The IVR has become something that is universally disliked.

Still, it has a future, and here’s why.

When IVR systems first came into being, the costs to implement such a thing were astronomical. In order to justify the massive capital outlay such projects required, the ROI was often calculated in part by how many empolyees would be replaced. While this made it easier to justify the price, what it failed to take into account was the annoyance that callers would experience having to use the 12 buttons on their telephone to interact with a computer. People hated it. Terms like “voice mail jail” became part of the vernacular.

The power of IVR, however, has yet to be fully realized. Why? Because up until now, it has never been cost effective to deploy it in simple situations, solving simple problems. IVR should never have been used to replace humans, but rather to create value for customers. Due to the prohibitive cost, that didn’t happen. Now, it can.

Joel Sisko and I were having a conversation about this earlier today, and he sold me on the fact that the new world of open source telecom means that IVR systems can now be deployed in ways that people might actually enjoy using. An example of this is: when you call your favourite pizza joint to order some pie, the system will say “if you would like the exact same order as last time, press 1 …”. Does this bother you? I’ll bet it does not. If they gave you 15 choices to navigate though, it would be frustrating. If they keep it simple, it is pleasant, useful, and saves you time. Previously only the large pizza purveyors could offer such a service. Now? Much smaller enterprises can afford to access the same technology.

The point here is that becuase an IVR can now be deployed very inexpensively, it becomes feasible to design simple systems that have only one simple job to do, rather than a massively complicated system that has to justify it’s enormous price tag by attempting to be all things to all people, which only ends up annoying everyone.

The ultimate IVR system would only have one choice. “press any key to …”, and that’s it.

VoIP technology makes it easy to interconnect between systems. It is likely that many VoIP enabled companies don’t even realize that all the technology already exists to route calls across internet connections using the same numbering plan the PSTN uses.

Initiatives such as ENUM have already developed the protocols to handle this, and if you have a next-generation open-source PBX, you already have this technology.

What I’m not sure is when the tipping point will be. For any company to take advantage of what amounts to free long distance calling anywhere in the world, two things have to happen: 1) you have to be referenced in the ENUM databases, and 2) any company you wish to call has to similarly be referenced.

The technology is ready. Next we have to figure out how to get everybody on board.

Randy Resnick and the fine folks of the VoIP Users Conference kindly invited me to join them for a chat.

We ended up having a very interesting conversation about building small, inexpensive systems that would compete on price with a traditional key system, but then also started talking about high availability. Two subjects that would not generally be part of the same conversation, which is just another example of why open source telecom is so important.

Check it out:

http://www.voipusersconference.org/2009/05/jmv-build-system/

Those of us head-down in this business of open source and emerging telecom will sometimes lose sight of the big picture. So busy with the day-to-day routine of running our respective businesses, we need to raise our eyes from the grindstone every now and then and have a look at the landscape.

A study recently done by the Eastern Management Group looked to quantify open source PBX adoption. The results show that open source telecom has been wildly successful.

This is personally satisfying, because I predicted in the first edition of Asterisk: The Future of Telephony that open source PBXs were going to change the telecom industry. I was not alone in this belief; it was something that was very obvious to those of us immersed in this space. Telecom had lost it’s soul, and we had found it back.

Two weeks ago I was at the Amoocon conference in Rostock, Germany. Conferences are important to someone like me. Our industry is still very grass-roots, and if we don’t hang out with our peers every now and then, we run the risk of losing sight of the big picture.

Open source telecom is alive and well, and still taking over the world.

The recent announcement that Nortel purchased Pingtel and SipFoundry has left me with mixed emotions.

On the one hand, it is good to see that the big boys in the telecom industry are taking open source telecom seriously. Congratulations to the SipFoundry team for having produced something of such obvious quality that they were the first of the open source telecom projects to be scooped up by one of the big boys.

On the other hand, it is hard to contemplate the magnitude of cultural shift that would have to happen at a place like Nortel, in order for something like this to avoid being squashed by corporate politics and entrenched ways of thinking.

Hopefully, Nortel will assume a hands-off policy with respect to SipFoundry, but how many big corporations can resist imposing policy on any new aquisition?

At the risk of adding nothing to the discussion other than negativity, I must be honest and state that I fear for the future of SipFoundry.

I hope that I am wrong, and that good things come from this.

I just finished getting YATE up and running as a conference server. Despite the lean documentation, I was impressed by how little I had to do to get it working.

I would expect that implementing more powerful features such as admin control and scheduling is going to require a bit more work, but it was amazing how fast it came together.

I’m writing a white paper for Sangoma on this, so look for that in the next few months on Sangoma’s website

I just got home from ClueCon 2008 last night. It’s the second time for me to be at that conference, and I was just as pleased as I was the first time.

This is the conference where all the best brains in open source telecom go, so if you want to meet as many of them as you can in the same place, ClueCon is one of the best conferences to do that.

It’s not a big conference, and there’s no trade show floor; it’s just development-minded people who want to meet and focus on the development of open source telecom.

I expect ClueCon to grow, and yet there is a selfish part of me that hopes it stays small; big conferences seem to lose their personality.

It seems that recently a lot of open source projects have emerged with excellent pedigree. In a nutshell, Mighty Asterisk has got some competition.

FreeSwitch has recently released their first non-beta version (which is already 1.0.1), and their community is exploding. FreeSwitch has performance capabilities that are very impressive, and if you are a developer of complex or carrier-grade voice applications, you owe it to yourself to give FreeSwitch a careful look.

YATE has been around for almost 5 years, and is well-regarded for it’s rock-solid stability. It has fantastic performance capabilities as well, and is solidly carrier-grade. The YATE community is a bit lean, and the documentation is a bit too light, but this powerful engine should not be ignored.

sipXecs evolved out of Pingtel, who recognized the importance of open source telecom and started SIPFoundry. These folks are serious about SIP, and cannot be ignored. Many of the SIPFoundry folks contribute to the IETF SIP standard, so they are well-respected by the community. Recently, Nortel has created a product that uses this technology, which demonstrates the professionalism and technical accomplishment of this product. Also, Amazon uses this for their PBX, so you know it scales and is reliable.

Open source telecom has grown up, and it’s going to be exciting to see how these projects mature and evolve. One thing is certain, all this innovation in the open source telecom space is certain to benefit both solution providers, and–more importantly–customers.

As someone who only knows one programming language (REXX), I was always paralyzed by trying to figure out what one to learn next. Perl, PHP, Ruby, Python; I’ve tried them all, and while I basically get what’s going on, I find that the time required to actually do anything useful is simply more than I’m willing to invest.

Freeswitch has been a project that has interested me since the day it was started, but I always figured it’d be a tough thing to get into, simply because it is geared towards people with a developer bent, (and I don’t see them writing a REXX module any time soon

But then along comes Lua. Hey! This looks kinda like REXX. I have started playing around with it, and it looks like I can actually get it to do useful things very quickly. FreeSwitch has a Lua interpreter built right in.

This means that I can grab a copy of FreeSwitch, and start playing with simple application development right out of the gate!

I’ll be playing . . .

I’ve been keeping a half an eye on the FreeSwitch project since it started, and a big change I have noticed in the last few weeks is that activity on the mailing list has really begun to heat up.

It should be interesting to see what sorts of things people are doing with this new bit of open source telephony software.

FreeSwitch is something to keep an eye on.