Click here to Subscribe

VoiceCon Fall 2005
By John Latta, WAVE 0539 9/30/05

San Diego , CA
August 30 – September 1, 2005

The convergence of voice and data is coming. One unified network infrastructure will make the management and operations of the IT and communications backbone of the enterprise vastly easier. That is the promise. But here at VoiceCon Fall 2005, we came away with market dynamics much less formed and certain. At the center is Session Initiation Protocol (SIP) which is widely praised. Converged IT and communications is a promise unfilled. The vendors both fight convergence and hope for it. Yet, the enterprise buyers have yet to be convinced the ROI is there. Thus, convergence is a problematical process of replacing old with the newly converged, maybe. Buyer pull has yet to materialize.

This is the first Fall edition of VoiceCon. The conference has run many years and now is happening twice a year. The turn out is good and enterprise buyers are looking for answers. There is a robust but small exhibit area. The focus is on the convergence of enterprise communications.

The WAVE coverage of this event will differ from the usual. Rather, than provide a talk by talk summary we will build on key points and relate what was said by the speakers.

Convergence – The Promise

The logic, at least from vendor perspective, is strong: The enterprise needs only one network which is Ethernet and IP based, for both data and voice communications. Need a phone - just plug it into the data infrastructure. To enable mobility, walk around with VoWLAN phones which may be combined with a PDA or notebooks and eventually cell phones.

The logic, from a buyer perspective, is less strong. Enterprise applications communicate with employees on their converged devices. The ideal of one number for all individual communications with fully implemented presence makes contact trivial. Call centers are more efficient due to integration. The savvy buyer can get phones from one source, the call processor from another and the gateway from another. Buyers have great control of vendor selection due to high levels of interoperability, enabled by SIP. The phone system is like buying PCs from different companies.

The problem – this is 4 – 5 years away.

Convergence – The Reality

We will use a market forces approach to best characterize the market dynamics.

Open vs. Proprietary

This is really SIP vs. everything else. In spite of the fact that H.323 based products still dominate the market all vendors must have a SIP story. PBX vendors, in spite of the words, still want single provider lock-in. Avaya, Cisco and Nortel would much prefer to sell their story and equipment without interference from SIP. Openness has the lure of open source from Pingtel SIPxchange and SIPfoundry sipX.

Thus, the words of SIP increasingly occurring in RFPs came up in the discussions and as a result vendors cannot ignore it.

Voice is One Use of the IT Network

In blunt terms voice can become just an application on a data network. In spite of the fact that IP networking wants to trivialize this application, there  are significant barriers to overcome. These include:

Quality of which the impacts of latency is one of the most difficult to control;

.99999 reliability

Security

Organizational

User expectations

As one speaker stated: “In the enterprise, voice is not an island.” This implies that IT is the mainland and the telephone mindset might take just the opposite view.

Does it interoperate?

If it is open, it must interoperate. There is far less agreement here. Edwin Mier, Miercom, presented the results of a recent survey of SIP vendors. On a scale from 0 to 5, the response was 3.1 that interoperability is far from assured. On another scale from 10 to 0 with 10 being out of box plug and play interoperability, there was only a 6.5 assessment of interoperability today.

ROI ROI ROI

In corporate environments the ROI drives all major investments. Yet, in the first session of the day, Nissan North America and McKesson who both recently moved to a converged network, ROI was discussed far less than the replacement of old equipment that was increasingly hard to get support for. In fact, both speakers said that toll bypass was fiction. The prospect of lower long distance calling when calls are 2¢ a minute or less is vacuous.

Mobility Integration

Much of the discussion was about how convergence comes to devices within the enterprise such as VoWLAN (voice over IP over wireless) phones. Yet, only passing reference was made to IMS (IP Multimedia System) which is the service delivery environment for 3G. It supports SIP. The potential is significant that SIP could be the unifying protocol between fixed and cellular. The limitations of this forum and geography were quite evident. IMS is the purview of Europe and Asia not the US. US companies are not drivers in IMS. The mobility story at VoiceCon was limited and incomplete.

Convergence Use Cases

If convergence is so great, where is the meat on how it is being used outside of communications? One comment was if the industry only sells IP to replace TDM voice networks, the industry is going nowhere. The problem is that we have heard no examples of how converged IP networks have brought new capabilities to the enterprise that would not be readily accomplished in other ways and at less cost. Cisco stated that the next stage of convergence is where the productivity applications will come. Microsoft argues that presence is a major advantage. Collaboration is to be a big deal. But these are all promises and they must demonstrate significant ROI to justify the large capital investments required.

Organizational Integration

It is no small feat to merge voice and data and, equally as significant, to make this happen in  organizations. This is the Internet forces challenging a 100 year old industry. But the speaker from Nissan North America described how critical it was to cross train the technical staffs at their locations. If convergence is to happen, this is not only in the technical but organizational.

Commodization

Vendors want to sell expensive razors and buyers want to buy cheap razors from any of many suppliers. Thus, the IP or PBX phones are expensive and frequently not interoperable. Conceptualize IP phones in the same way single line home phones are in the market today. Brand  has no value, price is everything and phones can be bought in many retail locations. In a panel   discussion, the vendors described how they were addressing phone interoperability but the end result is that they all want to sell phones in the 1,000s for each major enterprise sale. It is best to think of the phone system in terms of the PC where virtually everything is commoditized. In a fully converged network each part of the phone system can be an independent buy – in the same way that the components, both hardware and software, of the data center are independent buys. Thus, an element of the success of convergence is when Taiwan and China become major suppliers of the system components. This has not happened.

Is IP Communications Nirvana Pygal?

As the WAVE heard many times here at VoiceCon, the future of IP communications is a rich environment where one can be contacted 7 X 24 with many services which include: voice, video, messaging and e-mail based on presence which will allow for one to one and one to many communications and intense collaboration. All of this is enabled by rich clients which will populate desktops, PDAs and even phones which make ready access possible worldwide. We came away wondering – is this just technical distraction? For many individuals will not coping with rich functionality and omnipresent communications mean less work? Yes, some jobs can benefit but if this is the case it only implies a segmented and non-horizontal market. More important, will individuals want to be bothered by such distractions from the family as work makes it harder and harder to achieve such time.

UMA Technology Migrates to 3GPP

UMA Unlicensed Mobile Access was an independent standards effort to integrate unlicensed with licensed wireless. It is summarized below:

Unlicensed Mobile Access (UMA) technology provides access to GSM and GPRS mobile services over unlicensed spectrum technologies, including Bluetooth and 802.11. By deploying UMA technology, service providers can enable subscribers to roam and handover between cellular networks and public and private unlicensed wireless networks using dual-mode mobile handsets. With UMA, subscribers receive a consistent user experience for their mobile voice and data services as they transition between networks.

The participants included:

Alcatel
British Telecom
Cingular
Ericsson
Kineto Wireless
Motorola
Nokia
Nortel Networks
O2
Research in Motion
Rogers Wireless
Siemens
Sony Ericsson
T-Mobile US

As of June 19, 2005, with the completion of its specification on Generic Access to A/Bb, and its support in 3GPP Release 6, the efforts of the UMA are now folded into 3GPP. This validates the development of networks which support both 802.11/Bluetooth networks with GSM.

Motorola Implements Convergence with the Person

Although not citing UMA, Motorola has implemented something similar. The mantra for their efforts is what they describe as the only real convergence – at the person. That is, when wired and wireless converge it has meaning only with the individual. Consistent with this they have implemented the following:

Motorola CN620 Phone;
Motorola Wireless Service Manager
Proxim Wireless infrastructure;
Avaya Communications Manager; and
Cisco for networking.

Motorota Fills Mobility Hole

Motorola presented its CN620 phone and associated infrastructure. Motorola describes the CN620 as a mobile desktop phone – “move the phone to the location.” This is seen as a mobile desktop phone not a cellular phone. It seamlessly migrates between an internal enterprise 802.11 network and GSM. Partners include Avaya, Proxim and Cisco. No carrier offerings have been announced but SBC is running a trial. Motorola made the strongest business case yet for such a service and capability.

It is expected that the phone will be bought by the enterprise and not the carriers. The phone has push-to-talk capability. The phone only works over 802.11a networks and in conversations with Motorola they used this band because it better supports voice and more channels allow for better frequency reuse. Another key reason is better and more consistent voice quality. However, Motorola admits that this will require careful wireless engineering in the enterprise to assure seamless operation between access points.

The battery and power saving technology is impressive: 3 hours talk time on WLAN and 5 hours on GSM. The largest problem faced by Motorola is security. Supported are:

     802.11I
     802.11X
     EAP-TLS
     TKIP

within the enterprise and outside the enterprise VPN and IPSec.

Given that the CN620 is positioned as a mobile desktop phone, it has many of the PBX features. A simple one is hold which is not present on cellular. One of the impacts of this is that enterprise customers are asking the carriers – why cannot we have more PBX like features on cellular phones?

The economics are especially interesting. Most of the cellular phones in enterprises are reimbursed as an expense item. But 30% to 50% of the mobile calls occur when the phone is within the confines of the enterprise. Thus, there is a significant gain to better control the cell phone costs and to integrate within and external use. The CN620 is to address this. The more complex issue is – Why would a carrier dare offer a service that would support the CN620. Motorola makes the argument that here is a service that does not need a phone subsidy and has the carrier deal directly with the enterprise. Further, once the enterprise has adopted the phone, it is a large enterprise sale that cannot normally be achieved, due to the individual purchases of phones which are expensed.

It remains to be seen if the carriers accept the prospect of a phone that they do not control and especially one which may be perceived as lowering the minutes billed on the network. However, the WAVE came away impressed that Motorola has used IP telephony and SIP to address a missing hole of individual mobility in the enterprise.

BridgePort Networks Shows How to Integrate VoIP with Cellular, Home and Enterprises

Using a more carrier centric approach BridgePort Networks described its Network Convergence Gateway. This is purchased by carriers and operates as a bridge between external SIP VoIP and the mobile network. This box acts just like another switch on the mobile network. On the Internet side is acts as a SIP Register and Proxy. The box operates on both GSM and CDMA networks. It is claimed that the gateway will support the migration to 3G/IMS. One of the advantages of this gateway is that it will work with phones in the home, where there is a wireless router, the enterprise and in public access spots. BridgePort also cites that it is possible to extend Enterprise call features into the mobile network. The only operator cited using this gateway, IMC in Iceland.

To support building the ecosystem BridgePort Networks cites the Mobile IGNITE ((Mobile Integrated Go-to-Market Network IP Telephony Experience) Alliance. This is to allow for interoperable networks that support SIP and IMS.

Peer-to-Peer Enterprise VoIP – Not Ready

As VoiceCon ended there was one session on Peer-to-Peer VoIP. Bottom line – not ready for prime time. Too many issues at this stage of development.

Peer-to-Peer VoIP

The two presenters were: Jeff Fried, CTO Empirix and John Appler, VP, Popular Telephony.

John focused on the advancements in the technology. In particular, he discussed the use of Distributed Hash Tables (DHT) as a means to implement a distributed communications architecture. This avoids detrimental effects of P2P telephony on network performance. It also allows for dispersed unicast communications rather than multicast. The other component is uniform peer ID systems to allow for scalable growth. But when it comes to E911 and CALEA, John was uncertain how compliance could be accomplished.

The architecture which Popular Telephony advocates is a serverless network. Files on the network support voice mail, user administration, call detail and more. These are actually distributed over the network. The network nodes are PCs, PDA, cell phones and more. The argument was made that this does not create any additional card cost. But the details on the architecture were limited.

When it came to a recommendation on Peer-to-Peer VoIP, John only stated that beta sites are running in the US and Europe. Claims were made that an “order of magnitude reduction in initial and ongoing costs” is possible.

Jeff was more direct when he stated that VoIP is not ready for “prime time” but that it is coming anyway. There are many quality implications of which security threats are real. His final caution is that the latency demands of VoIP and the need to manage network usage make VoIP a challenge.

Empirix did a survey of companies and found that 63% would be concerned about putting a VoIP phone on the CEO’s desk. When it came to applications, Empirix found that 150 out of 177 respondents viewed the value of VoIP support of “general telephony” important. 102 saw messaging as an important application and 82 toll bypass. The major quality concerns are in voice (47%), security (25%) and interoperability (8%).

Testing at Empirix found issues in network micro-congestion, route change delays, and losses from route changes.

Overall this assessment was the most cautious of any presented at the conference.

WAVE Comments

The convergence of IT and voice communications is a slowly emerging market. It blends two very different technical, organizational and business cultures. Think of making everything, from products to services, a commodity. The line between voice and IT no longer exists. The problem is that no one has a clear idea what the benefit is to the end user, be it a consumer or enterprise. Yes, we hear all the arguments about more features but this fragments the ROI proposition across enterprises. What is good for one enterprise may not be good for another. Thus, at VoiceCon Fall 2005 we find the forces typical of emerging markets that seek to tear apart existing markets interesting. But this alone is not sufficient to make a sustainable market. Critically missing are compelling uses and integration with mobile.

We came away from the sessions asking – What is missing? Given the enterprise focus of this event, the one area left open is mobility. Today something like 80% of the mobile phones used in enterprises are reimbursed and the TCO is $130/month and rising. Some CIOs see this as just another device to support in the enterprise and one whose cost is not under control. The cell phone implements mobility ad hoc and largely not related to the enterprise. This is about to change. Motorola and BridgePort showed how it can be done. Clearly Europe is well ahead of the US in terms of standards when UMA was folded into 3GPP.

The WAVE assessment is that:

     A coherent approach to enterprise mobility is needed and solutions are appearing;

     Technology is not the issue;

     Business models are undefined with the carriers being either at the center of enablement or lack of support.

Extending the enterprise with mobility, as a part of the IT infrastructure, has a close parallel with what the cell phone did for consumers.

Comments?
E-mail webmaster
Page updated 1/24/07
Copyright 4th Wave Inc, 2007