Tony Rybczynski – affectionately known within Nortel as Tony Ryb is one of the top thought leaders in the communications space and has been with Nortel as long as I can remember. He has also been a TMC columnist and speaker at Internet Telephony Conference & Expo for a decade.
He recently sent me this e-mail about open-source communications and Asterisk and I thought is was worth passing along. Perhaps his ideas will spur a dialogue on this “guest blog” from Tony Rybczynski.
Tom Malone of MIT has observed that the most important innovation of the 21st century will be new ways of organizing work, this enabled through the low cost global Internet and collaboration technologies. Open source software is one good example, though there is an open source motorcycle under development in China!
The open source software movement appeals to most everyone. For technologists and the vendor community, it represents a new way of developing basic technology building blocks, with the promise of lower development costs and accelerated solution development. For enterprises, it promises more open more agile systems, less dependence on the telecom vendor and ultimately lower cost. Open source is not a panacea and there are multiple non-technology reasons why some open sources have experienced limited success: take open source routers for example.
Growth of the IP Telephony market has created opportunities for IP Telephony open source software, Asterisk being the best known. Asterisk alone does not make an IP Telephony system, which has to include end points devices (like SIP desktop phones, WiFi handsets and various forms of soft clients), media gateways to the public network, to legacy PBXs and to fax machines, and ancillary areas such as voicemail, conference bridges and music-on-hold and announcement.
There are generally, three ways that open source software can find its way into enterprise and SMB environments.
1. Some enterprises may embark on being their own system integrator and roll their own implementations, assembling parts from wherever. While a small percentage of tech savvy enterprises may go this way, this is not a core competency of the vast majority of enterprises, nor an area with adequate business value to warrant investment. UPenn has embarked on this approach, and stated (at a session at the spring Voicecon) that this was driven more by the need to integrate with back office applications. Cost reduction (on a TCO basis) over their existing Centrex service was roughly the same for a number of IP Telephony systems they evaluated. At the same conference, the Southern Company presented their application of IP Telephony open source to build a paging system for 200 users.
2. Some may embark on working with ‘open source IP Telephony suppliers’ who undertake the SI function and wrap technology and services into a product solution. These can potentially meet individual customer needs, but only if they meet the feature/functionality, reliability, scalability, scalability, regulatory and support requirements of the buyer. However, enterprises need to ask the question of what’s the path to unified communications, an area which is not addressed by the open source community in an integrated fashion. In fact, with IBM and Microsoft entering the UC market, IP Telephony RFQs are increasingly being replaced by UC RFQs (at least directionally). We believe that this raises the stakes for these open source suppliers and will diminish the appeal of IP-Telephony-only open source solutions.
3. Some will buy open source unknowingly from an established telecom supplier who have integrated open source software into an IP Telephony system. For all intents and purposes, this is the ‘traditional IP PBX/Unified Communications market place.
What’s our view on open source? Nortel is focused on delivering solutions meeting enterprise and SMB requirements and always looking for ways to lower the development cost while accelerating development time. In general, we view open source as an interesting software technology that is useful for many applications. In fact, Nortel already leverages open source technologies such as Linux and SIP protocol stacks in a number of our solutions.
Nortel is committed to transforming telephony into a Unified Communications software application, and we are actively investing R&D towards this objective both individually and working with partners such as Microsoft and IBM. Openness is a key design principle, leveraging SIP, open APIs, Web Services and SOA, and will continue to be one of the focal points of our solution development strategy.
Do you see this differently?