Concerns with IP-PBXs

Concerns with IP-PBXs
We have a number of concerns with IP-PBXs that are specific to these systems and are not identified, as applicable to IP-Centrex. Some of the following considerations may be sorted out over time, but the last four in the list will probably stay with us for years to come:

- The capital cost of an IP-PBX is still likely to be higher than that of a TDM-based PBX with a similar capacity. In most price quotations that we saw in 2001, the IP-PBX vendor was asking an average of 30% more than for the traditional PBX. By the end of 2004 there will, probably, not be any difference in pricing between the two architectures.

- On the other hand, the manufacturer-supported life of a conventional PBX will probably be limited, as development funds are dedicated to the IP-based systems.

- Several IP-PBX vendors have launched their product and then changed their plans within a year or two. Additionally, every system includes some proprietary features that make the free interchange of components, such as phones and gateways, difficult.

- Many of the interesting applications that have been promised with IP telephony have not yet been written, or may exist only for one specific combination of telephone and computer systems.

- Several manufacturers, such as Alcatel and Mitel, have based their main IP-PBX on a dual-switching fabric architecture, retaining a TDM capability in the same box as an IP-oriented call controller. This complexity must add cost to the system and make fault diagnosis and maintenance more difficult.

- For these reasons, this approach may not be the best long-term investment for the customer. It may be a better strategy to acquire a pure, IP-only system and to support non-IP phones by outboard gateways, which could be moved to other locations as conversions proceed. Because of its complexity, a hybrid TDM/IP system is probably not the right vehicle on which to base an out-sourced IP telephony service.

- The integration of voice, video, and data signals over an IP network demands a special set of design and implementation skills, which take time and practical experience to acquire.

- One of the advantages claimed for IP telephony is the cost savings that results from using one system management package instead of two. However, telephony and video conferencing are not just two more applications that will now reside on the corporate LAN. There is a serious risk when management becomes convinced that anyone with some data expertise can plan, configure, and administer a feature-rich, multimedia system. Because organizations still depend on voice communications with their customers and prospects, they cannot afford to eliminate voice system administration expertise entirely.

Most of the early installations of IP-PBXs were treated as pilot trials, in conjunction with larger, legacy digital systems. The integration of IP phones into the existing voice mail subsystem and with the number display features of the TDM-based PBX proved difficult and remains insoluble in some cases.

Several small office-in-one-box products, including those from Cisco and Mitel, have come onto the market as part of the development of IP-PBXs. We should remember that some major manufacturers, including Nortel and Wang, launched integrated small office systems (at that time including a built-in 3270 controller for IBM mainframe terminals) in the mid-1980s, and that all of these products were dramatic failures. Most organizations have long since decided which manufacturer's data equipment is to be used in their networks and probably are not prepared to change at this stage.

Because Cisco controls some 80% of the enterprise-level data network equipment market, the company dominates its accounts, in the same way that IBM was all-powerful in the data processing business of the 1960s and 1970s. Some decisions to buy an IP-PBX were made by the CIO or another senior manager because of this influence, sometimes against the advice of the company's telecom manager.