VoIP represents a major paradigm shift in communications technology. It is changing business processes and consumer lifestyles worldwide. Not only is the change over to VoIP irreversible, VoIP has proven to be the main catalyst in convergence of services and networks. And yet, there is no documented recognition of the contribution made by the companies that established this industry. Although the contribution of varying nature and importance came from numerous organizations, a few among them stand out in the crowd. They have been the chief agents of this change, the paradigm shift. iLocus intends to write this important chapter in communications history which could otherwise remain unarticulated.

Starting next week, iLocus will write about the top 10 organizations that established VoIP industry. We will examine the VoIP market from 1996 onwards and list certain critical developments in the industry that helped establish it. We will then correlate those developments with 10 selected companies, and further discuss in detail the contribution made by each of those companies.

Product named Internet Phone, developed by VocalTec, was a pioneering product from two aspects: (1) It had the ability to do VoIP call over the bandwidth-starved public Internet of 1995, and (2) The developers of Internet Phone had to overcome the limitations of the hardware of that time which comprised of 486 based PCs and modems of 28.8 kbps speed.

Unique contribution here was to achieve real-time voice, including compression, with the standard PC hardware of 1995, and enable users to talk over public Internet in real-time. That was a major breakthrough that VocalTec achieved. Apart from Internet Phone, the first ever IP-PSTN breakout gateway was also introduced by VocalTec. The company demonstrated the first gateway at CT Expo in Los Angeles in the March of 1996.

There is no other company that can be associated with pioneering real-time voice over Internet. VocalTec was the first in introducing PC to PC VoIP client and hence is selected as one of the 10 organizations that established VoIP. There were other competing companies that were working on such projects who introduced similar solutions subsequently. These included Nuera, Net2phone, Cisco, Clarent, Vienna Systems etc.

VoIP at Level 3 started in 1997 with an internal project on inter-working of SS7 and IP. The project aimed to separate control and media elements for IP data traffic and was initially named as SIOS (SS7-IP Operating System) project. Sometime during 1998, SIOS was renamed as Softswitch. It was the Level 3 team that actually coined the term. Apart from the term Softswitch, other VoIP product names that are generally used today also have their origin in the early Level 3 VoIP project. Some of them include ‘Media Gateway’, ‘Trunking Gateway’, ‘SS7 Gateway’ etc etc.

If we are to believe what they say, then China is going to be the largest IPTV market soon. Microsoft seems to have delayed IP TV plans of many including AT&T and BT. France Telecom’s subscriber take-up seems to have slowed down a little while Telecom Italia is finding it hard to get the service launched. Even the trail blazer Fastweb admits to a modest subscriber number of 170k or so. But China also has certain challenges to overcome before it dominates the IP TV scene. An important issue is the proper structuring of AVS (China’s homegrown Audio Video Standard) eco-system there.

The Chinese government has understood that the MPEG-LA royalty scheme is a very expensive proposition for China. It has instead put its weight behind an alternative to MPEG-4 but the players will need to iron out the standards quickly because the more non-AVS solutions are deployed the more difficult it will be to backtrack. The IPTV vendors who do both MPEG-4 and AVS will do better simply because the customer has the choice. Envivio, for instance, has a working partnership going with all major Chinese integrators (Huawei, UTS and ZTE) with their AVS encoders and they are being trialed by all of the major Chinese telcos.

I have chosen ITXC among ‘The 10 That Established VoIP’ for the company’s pioneering contribution in the areas of International Long Distance (ILD) VoIP wholesale model, management of voice transportation over public Internet, commercial interoperability among varied VoIP platforms, and VoIP peering.

Working with a vast group of VoIP POP owners (ITXC itself owned a large numbers POPs) induced competition among termination partners which in turn resulted in more competitive rates than what companies like AT&T had access to. This kind of wholesale model was totally new.

I am carrying out a brief survey on Outsourcing of Communications Chip Manufacturing. It seems that more than half of the chip makers I have interacted with are either fully or partly outsourcing their manufacturing requirements. I am testing some of the main reasons why they outsource and also the main reasons for not outsourcing (if they do not outsource). The survey will be a short one. Hope to publish this paper within the next few days.

Here is the backgrounder to the topic: The trend of outsourcing of manufacturing communication chips has grown into a $25 billion market with about 74% of the total communication chips being manufactured through outsourcing to Fabs and Foundries across globe, mainly to Asia-Pac. About 50% of the outsourcing market is held by large chip manufacturing facilities like TSMC, UMC, SMIC and Chartered Semiconductors located mainly in Taiwan, China, Singapore and Korea.

Minerva has been involved in video-over-IP space since late 90’s. The company is one of the pioneers in the world of IPTV. The main product Minerva sells is iTVManager. iTVManager has two components: middleware and applications. Applications include EPG delivery, pay per view, video-on-demand, PVR etc. The middleware itself has two components: a back office server component and a client component that resides on the subscriber’s set-top box.

A bit about Minerva’s background?

We have been traditionally engaged on two major fronts: video processing (i.e. video encoding and transcoding systems) and service delivery (i.e. middleware and applications). Over the past couple of years we have been morphing into a pure software infrastructure vendor.

Which geographic region are you mainly focused on?

Our initial focus was in the United States, targeting small and medium size independent operators. We are now gaining significant traction with large operators in Canada, the United States, Latin America and Europe.

Can you share some numbers with us related to your deployments?

We have over 120 installations worldwide. Surewest, one of our customers, has the largest IPTV deployment in the US. Large operators in both North and South America will launch services by the end of the year, significantly boosting the number of subscribers enjoying television services powered by Minerva software.

Over the past few months, I have been wondering why the perceived value of our email newsletter has dwindled. Along with our research reports, we have been updating over 8,000 newsletter subscribers with market analysis (mainly VoIP) on a weekly basis. We never doubted the quality of content in the newsletter. Yet the number of subscriptions has remained almost stagnant over the last 18 months.

We have come to two main conclusions: (1) Email subscriptions (although reader solicited in our case) are giving way to syndicated feeds, and (2) Market research companies such as iLocus should not interfere in the domain that is better served by the media and qualified bloggers i.e. readers do not expect us to report on an announcement made by companies, or even provide comment/analysis on a press announcement. I think media sources such as Light Reading and blog sites such as GigaOm do that job exceptionally well.

Texas Instruments is expecting to ship about 5 million video surveillance and security chip devices in 2007. These will mainly come from TI’s DaVinci product family. Video surveillance and security solutions have seen a boom since the 9/11 incident.

But the issue is that TI has higher end programmable DSPs on offer and these are mainly used in IP cameras, DVRs and other storage devices. Reaching 5 million figure this year seems an uphill task. Out of the total 15 million estimated cameras video surveillance cameras shipped per year, only about 2 million are IP cameras where TI solutions could go in. If TI estimate is to believed IP cameras market must see an exponential growth for the remaining months of 2007.

I am getting started on the theme ‘Public Mobile VoIP Clients’. Let me define the segment: these would be the likes of Fring, Barablu, Nimbuzz, Mig33, Jajah, Skype etc i.e. public mobile softclients that have VoIP capability. First off, a question pops up whether these services will impact cell operator revenues?

It seems logical to assume that a subscriber can potentially substitute calls of his/her cell operator with the help of these clients. However, right now, there are two issues. The bandwidth available for wireless internet connections is not adequate just yet. The other issue is the overall context in which these softclients reside on a mobile handset.

NewPhone is a triple play service provider offering services in Norway and Sweden using the incumbent service providers' broadband and mobile Internet. The company has grown from a turnover of SEK 8 million in 2004 to SEK 51 million in 2006.

You are a mobile service provider in Sweden. To what extent have you utilized VoIP so far?

We integrate the IP PBX with mobile phones. If you have a fixed line phone, we can take the fixed line numbers, put them in the PBX and when somebody calls that number the PBX will divert the call to the mobile phones. Between the PBX and the mobile phones we send calls through a SIP tunnel and that is where we use VoIP. Being IP based, subscribers can take calls on their handsets anywhere in the world provided they have IP connection and our wireless IP Phone.

Which VoIP vendor do you work with?

We have deployed Broadsoft solution.

How would you compare fixed line VoIP with the mobile VoIP in Sweden?

All the research we have done so far tells us that the market here in Sweden is going mobile and that makes it even more relevant to just go mobile with VoIP from the beginning.

Flash could become part of set top boxes as it is a very good substitute to HTML or other user interface presentation layer. In France, Free makes YouTube-like channel available to their IP TV subscriber that allows subscriber to upload their home-made videos to a specific Free channel, and it is all Flash based.

Flash has a video codec that can be used in IP TV offering. However, Flash was not originally designed for video so it requires some work to do this properly (Bluestreak does a good job at it). And Flash is proprietary so that there are interoperability and scalability issues if used in a video environment that is very non-heterogeneous such as the mobile video world.

Home entertainment solutions, such as the PlayStation, Xbox and Wii, are enabling services that could bring them in direct competition with IP set-top boxes. Today most IP STBs do not offer gaming at the level proposed by any of the Home Entertainment Solutions (HESs) like Xbox or Wii. Going forward there are two ways to look at it.

Either the HESs are going to become STBs or the STBs are also going to have HES features. It is expected that Sony and Microsoft will have their HESs also become reasonably priced STBs. Microsoft has already announced this type of plan and their announcement of Mediaroom (merger of their IPTV middleware and the Mediacenter) is certainly taking us in that direction.

Entropic, established in 2001, has been focused on Home Entertainment networking. It provides chipsets along with software in this segment. C-Link, its flagship product allows the sharing of multimedia content over COAX infrastructure. The company recently acquired RFIC providers RF Magic and Israel based Arabella Softwares.

Your market focus is home entertainment networking and broadband access. What changes have you witnessed in these markets over the past six years, and what have those changes meant for you as a chipset vendor?

The home networking space historically has centered around data and voice. We are seeing an emerging trend toward digital entertainment including streaming video. The vision of Entropic is to capitalize on this trend. The other trend we are capitalizing upon is the need for higher speed which out chipset facilitates.

c-LINK is your flagship product. Can you share with us its current market status in terms of shipments and other data?

Our growth has been spectacular. We have shipped cumulatively more than five million c.LINK chipsets.

By 1999, the first phase of VoIP miracle had been achieved. People could make international calls for a fraction of the cost they previously paid. Companies like Vocaltec and ITXC had made significant contributions in making that happen. The next big thing in VoIP was pushing packetization to the trunk level. But the vision varied in the industry.

Service providers at the time had two choices: either continue using public Internet to transport national long distance voice traffic, or somehow re-engineer their own networks to become compatible with the data-voice convergence trend. The first option was simply not good enough because there was not enough IP bandwidth to transport high volumes of voice traffic. VoIP technology also was not mature and scalable enough to handle high volume voice traffic. Service providers therefore began a long march toward a re-engineered PSTN.

Companies like Alcatel-Lucent are working on system-on-chip design for set-top box (STB) whereby the STB functionality will reside inside the TV. There are both advantages as well as disadvantages in incorporating STB functionalities directly into televisions. For example, on one hand this can help operators reduce costs because they typically subsidize the STB. On the other hand, the STB technology will change much faster than the frequency with which consumers change their TVs. The consumers could therefore not be able to benefit from all the new enhancements.

There are certain new requirements in the home. There is need for a home gateway. That will be a separate item and not embedded necessarily inside the TV. The decoding capability is likely to move to the TV set. The trend we see is that certain bits of the set top box functionality will move inside the TV. The other bits of functionality will become richer and more powerful and become part of a home gateway. That could be good for IPTV because the cost for installation of IP TV will come down overtime. At the same time, functionality of the home gateway will enable new applications including wireless capability.

A little under 15 million ports of Power over Ethernet (PoE) equipment are being shipped a year. That number does not match the hype created a few years back. The issues seem to go beyond standards (IEEE 802.3af, IEEE 802.3at). The main problem is that network owners are concerned with speed of their networks right now rather than the utilization of that network. Utilization aspect has taken a back seat for a while. Another major problem is the orientation of PoE market. It is an application driven market.

PoE is used only in very specific applications where its benefits are really seen and tangible to the market … for instance VoIP. It is the largest driver of PoE. Vendors have been able to push their IP Phones with the help of PoEs. Similarly WiFi and WiMAX are the other areas where PoE is getting traction.

But vendors see plenty of opportunities ahead such as the ability to power laptops. Segment leader Microsemi which holds over two-thirds of the market shipping about 10 million ports in 2006 expects its shipments to grow 50% annually. According to Microsemi, 80% to 90% of ethernet switch ports should be using PoE in years to come.

In order to justify our choice for selecting 3GPP among the 10 most influential organizations in VoIP history, we have to start with answering two straightforward questions: (1) How far is 3GPP responsible for drafting IMS blueprint, and (2) What role has IMS played in promoting VoIP?

Question 1 is easy to answer. 3GPP as a forum is wholly and solely responsible for drafting the IMS blueprint. There was no particular member (out of the usual suspects there such as Nokia, Ericsson, Motorola, but also companies like Huawei, LG, and RIM)

Click here to read Part 1 of the interview.

As a consumer, I can share one broadband connection on several home PCs. Why I am not allowed to share entertainment like that over TVs or other output devices?

Not sure if that is really the case. In the U.S., the consumer can get several STBs for multiple home connections. The satellite or cable TV provider is not charging two or three times per TV. They are charging an incremental amount though you do have to pay for the entire service.

Yes but when you share a broadband connection at home you do not pay incremental charges.

Yes, that is right. There is that difference.

What if the regulators allow multiple TV users within a home without incremental charges? How would that affect your business?

We are not really providing additional links to the WAN per se. What we are doing is allowing the distribution of content inside the home. So it is a totally different model than what you are thinking about.

Apparently Broadsoft ended 2006 with a strong y-o-y growth rate. The company continues to expand its business internationally with an increasing amount of its revenue coming from outside the US. Europe, in particular is experiencing strong growth with a number of new customer wins focused on IMS, including KPN and Telenor. During the first two quarters of 2007, there has also been an increased level of interest in mobile and FMC.

While BroadSoft has been successful in penetrating competitive carriers and ISPs, the focus is still the fixed-line and mobile incumbents. The total number of IMS wins has approached 30 now, and lots of them are through Ericsson, which is proving to be the biggest channel of IMS deployments for Broadsoft. The majority of the IMS wins have come from outside the US. The scale of deployments within IMS contracts is still modest however the potential is there.

Voice enabled WiFi or VoWiFi market has not been doing well for the last several quarters now. The reason we had heard in the past is the FMC. With the advent of FMC, single mode VoWiFi is being seen as addressing myopic opportunities.

The VoWiFi market could be of interest within two broad categories. One is the establishment of city wide WiFi network and the other would be under roofs whether in offices or at homes. But FMC is giving better alternative in both these areas and as a result there has been a considerable reduction in the revenues of chipset products for VoWiFi segment. iLocus 2007 VoIP annual report sees a reduction of about quarter of revenues on yoy basis.

Could we start with a brief background to ViewQwest?

ViewQwest was founded in 2001. It was originally an ISP focused on data connectivity in the central business district of Singapore. From there on the company started adding value added services such as managed VPN services, back up services, disaster recovery etc. Viewqwest made a niche market for itself in the financial services industry where we provide a lot of outsourced IP management services to merchant banks.

Tell us about your VoIP service.

Over the last two years we have been testing various VoIP platforms. In February last year a service called One Voice was launched. The One Voice service which is essentially targeted at the corporate market with people who have offices in Singapore as well as globally, the primary motivation is to replace existing PABX systems. We are using a hosted Centrex model.

In order for the telcos in North America to compete against the cable operators, the IPTV technology has to be really mature and be feature rich. You have in North America to offer HD and PVR if you want to compete against cable. Only now with the nextgen set top boxes is it possible for an operator in the US to offer HD and PVR and deliver the services over their bandwidth constrained ADSL networks. On the other hand, in rural America, cable is not as strong as the cable operators have not invested as much.

In Europe you do not have a strong cable presence. IP TV in Europe would sell even if it offered just SD and without VOD because cable operators are not so strong there.

Part of my interactions with companies enabling public mobile VoIP clients ….

SoonR is about mobilizing desktops …. Remote Desktop Connection … a concept that is gaining traction among the business VoIP users as well. There are some companies that are offering solutions to roaming users providing access to their desktops when they are remote. The concept is that a user's handset with a data connection - wherever the user travels - can have access to all the applications that are on his desktop (including Skype).

Click here to read Part 1 of the interview.

Which vendors have you been working with?

All our gateways are from Cisco. The call switching platform is from Tekelec. For the end user devices we use Cisco phones.

What made you choose Tekelec and Cisco?

With most other hosted solutions you have a call switching platform and then there is the voice firewall to be sourced from SBC vendors. With Tekelec both the voice firewalls and the call switching platform are made by the same company. That reduces the finger pointing when there are technical issues. And we essentially like the feature set that was available from Tekelec and the price point was right. (iLocus note: The products that are mentioned here are now owned by GENBAND).

A follow up to my post yesterday ….

Single mode voice enabled WiFi chip makers like Conexant and CSR have their versions of dual mode handset chips. And so have others. Not many seem to admit that single mode VoWiFi is a declining application. Conexant sees both markets remaining intact.

It seems like the main markets for single mode VoWiFi will be the enterprise segment: enterprises with large campuses having mostly the On-Net calling requirements. Universities, Factories, Hospitals and Fire & Emergency departments are also the likely candidates. These entities all have either large campuses and/or their calling pattern is limited within the group of their colleagues, peers, etc.

This is what I have been hearing from vendors recently:

Eastern Europe : will draw a great deal of attention in the coming years since the region now has a good IP infrastructure, and a number of digital TV providers already exist in the market.

Japan: will grow very fast due to favorable environment for deregulation, consumer behavior, competition, and access to capital.

US: will continue to be stymied until carriers switch from non-scalable Microsoft platform!!!!

Western Europe: already well addressed by vendors and service providers

Other new regions: China and US will become hot markets with the predictable migration of cablecos to hybrid cable / IPTV offerings.

The industry is not witnessing any acceleration in the end-office replacement. There is a lull in this regard worldwide. Service providers are going for VoIP overlaying more: either as completely separate service offering or in certain cases part of other IP offerings as in the triple play service offerings. That is not to say that legacy switches are not being replaced. They are being replaced but the speed of that replacement has not accelerated.

Replacement has slowed down for a couple of reasons. One of the reasons is that the service providers have decided to extend their broadband program by offering high bit rate DSL based services. Accordingly their plan is to upgrade their access gear. So they are replacing their older access gear or just copper pairs with FTTX.

If we go through a typical supply chain in communications chip industry we have Intellectual Property (IP) developers, Reference Designers (RD), chip manufacturers, and fabless chip vendors. In practice, IP developers and/or Reference Designers exercise little influence although they form the basis of a chip.

Fabless vendors in comm chip market are the ones that dominate the scene. They take the maximum chunk of revenues. Even the start ups here have been able to either break-even or are nearing break-even as their RoI is high. Obviously by outsourcing things, they are able to allocate more resources toward R&D and understanding customer needs.

On the other hand the royalties that make the basis of revenues of IP sellers are still in the range of 15% - 45% of the total revenues. And if the customers are not shipping the new chips yet, these companies do not see any immediate revenue realization for their work.

Over the last couple of years, there has been a move from customized solution (phase 1) to open platform solution (phase 2) in the IP TV market. In the first phase the challenge that the market had was to really make television work over the broadband infrastructure. T