We have been looking for ONE BILL for over two decades now, but we still put up with multiple bills. If there really has to be just one bill, why limit it to just telecom services? While we are at it, why not have really one master billing platform that bills for everything from telco services, TV, gas, electricity etc – basically bill for all those things where a centralized billing infrastructure is feasible.

Consumers had to put up with separate billing for shopping. And then credit cards came along and we outsourced the billing part to the banks. I think banks are best equipped to bill for everything. They have historically been the best collectors. If the existing banking software could be extended to interface with the future IMS billing, for instance, the banks would be able to host a billing network for telcos.

If your memory serves you well, you remember where you left your nail cutter last time you used it. If are not a homely person like me, and you frantically look for things when you need them, and you nearly drop dead looking for it, then I have an idea that we should explore.

People like me demand a software utility (loaded on some device such a mobile) in which we can select the name of a household item (added sometime in the past to the list), click Enter, and get the current location of the item within the house.

Applications market was supposed to be huge in NGN networks. What we have seen instead is a simple PSTN emulation led by softswitches and not application servers. What are the reasons for this?

There are several key barriers depending upon who you speak to. New comers have to at latest match the existing core offerings and that is their first priority. For established players there are integration issues. The applications technology exists but the business models are still in emerging stages. Obviously user adaptation is also a delay factor.

There are not many generic application server vendors in NGN market right now. Dynamicsoft and Ubiquity were acquired some time back. Why has the market not sustained many application server vendors in NGN, when all NGN is about is applications?

Revenue for Sequans in 2Q'07 was in the range of $4-5 million. The company crossed a total of 25 customers in this quarter. This excludes the design wins.

Unit wise, it shipped around 63,000 chips for base stations and client devices. Majority of them were for client side. These chips were both for IEEE 802.16-2004(d) and IEEE 802.16e.

Worldwide cellular chip market revenues for 2Q 2007 touched an estimated $1,910 million. An estimated 130.6 million chips were shipped during the quarter.

The market was led by Texas Instruments which had a share of around 24.2%. That lead was followed by Qualcomm, Freescale, Broadcom and STMicrosystems in that order. Together these five generated around 68.8% of the total revenues generated by cellular chip sales.

Cellular chips 2Q07 update tracks the shipment of chips that are mainly used in handsets and core network elements in cellular technologies that include GSM, CDMA, and 3G.

When hosted VoIP deployment was not so hip and switch replacement was in, vendors like Taqua and MetaSwitch were in the driving seat. Taqua, the leader in the pack, had revenue rate of around $8 million per quarter. That was around year 2003. Then companies like Softbank BB and Vonage lit the match in the other quarter and service providers started running blind towards hosted VoIP.

When in 2004 Tekelec acquired Taqua, the nextgen Class 5 product startup was managing turnover of around $6 million per quarter. By late 2006 that had gone down to around $2 million per quarter under Tekelec’s ownership.

• China Unicom: 3.2 billion National Long Distance (NLD) VoIP minutes, 26 million International Long Distance (ILD) VoIP minutes

• China Mobile: Estimated 9.8 billion VoIP minutes, mostly NLD

• China Telecom: Estimated 12.5 billion VoIP minutes of NLD, 29.4 billion minutes of VoIP local call volume (through iPAS)

• China Netcom: LD VoIP minutes info: to be updated later, 19.6 billion minutes of VoIP local call volume (through iPAS)

• 924k VoBB subscribers (in Italy) at the end of 2Q07, near about 70% increase over 1Q07 subscriber figures

• VoBB subscribers represent about 15% of broadband subscriber base

• VoIP International Long Distance (ILD) minutes carried during 2Q07: 920 million

• Ended 2Q07 with 1.7 million VoBB subscribers, a 23% increase over previous quarter

• Capex spent of over GBP 900 million (approx $1.8 billion); nearly 30% spent on 21CN and related projects within BT Global Services and BT Wholesale divisions.

• Commercial deployment of VoIP within 21CN going well.

• Estimated VoIP traffic carried during the quarter: 1.8 billion, over half of which is local call volume

• 3.017 million VoBB subscribers in France as of end 2Q07

• VoBB subscriber base represents about 46% ADSL customer accounts

• 623k VoBB customers outside France (mainly in the UK)

• $114 million revenues from carrier voice, which represents about $46 million decrease on same quarter last year. Decline due to reduced emphasis on carrier voice wholesale and increased focus on enterprise VoIP services.

• VoIP traffic: estimated between 4 to 5 billion minutes during 2Q07, about 8-to-10% of which is international long distance

• Has decommissioned 8 legacy TDM switches so far

Swisscom is pursuing one of the easiest roadmaps to IMS. Ericsson has deployed the core IMS components (CSCF) and will be integrating three separate network components of Swisscom.

Three different network components or service offerings that Swisscom will integrate to achieve one of the first widescale IMS compliant networks are: fixed line Voice-over-Broadband, residential mobile service, and a converged component. The converged component consists of a single phone number that services both the mobile phone as well as fixed phone.

Business IP Trunking has been one of the fastest selling products of those service providers that have been selling VoIP access to enterprises. Business IP trunking is a low risk proposition for both service providers as well as customers. Throwing a broadband access pipe and not having to change operations inside a customer premises is a much easier and a quicker sale for a service provider.

On the equipment side there are two types of vendors who could potentially compete with each other in this area. Type A would be Class 4 softswitch vendors like Sonus and Veraz (apart from legacy Class 4 softswitch vendors). Type B would be vendors like Broadsoft and Sylantro. Type A typically bundle their media gateways with the Class 4 softswitch product. Type B, Broadsoft and Sylantro, on the other hand bring in a disaggregated application server for the purpose.

How would you sum up the competition you face in UMTS and Femtocell areas?

In UMTS, there is competition on the big base-station level from Nokia, Ericsson, etc. But innovation comes from the smaller side of base stations. We are going from macro base stations to pico and femto base stations. This is where we can offer cost effective chip solution.

On the femtocell we don’t yet see competition. But I am pretty sure there will soon.be some: companies like ST, TI, ADI and others are all exploring this space.

What about competition in WiMAX area?

WiMAX is a new market and we entered WiMAX three years ago as one of the early entrants. It is more data centric market. Data centric market has traditionally been open to new comers like PicoChip. On the telecom side, change in suppliers is made reluctantly because there are many standard conformations, etc. WiMAX is a much open market and the smaller ones have jumped in. On the big base station for WiMAX it’s mainly TI and Xilinx.

It can also be complicated sometimes. Freescale is sometimes a competitor as well as partner for WiMAX. They have a way to implement PHY which is part of the base station we specialize upon and at the same time we use their processors. ADI competes with DSP but we use other components from them. You have some smaller companies in WiMAX which are focusing on terminal end but may provide solution for base station also.

If you believe that startups in chip are able to get in only through outsourcing you may think again. It has become inevitable to outsource chip manufacturing irrespective of how long the vendor has been in the business, or how big a vendor is.

I have just completed a short survey of about 21 chip makers worldwide. Out of the 21 I interacted with on the subject, 14 outsource. Out of those 14 vendors, 8 are established large chip makers. It is not just the upfront investment in manufacturing facility (which can cost up to $3 billion to set up, and), which outsourcing saves but there are many other factors. I will be sharing some info shortly.

• Voice revenues of $292 million during 2Q07

• VoIP revenues in the range of $100 million to $120 million. VoIP revenue generated from domestic and international voice termination services, and VoIP local inbound, enhanced local and toll free services

• Revenue includes $6 million reciprocal compensation

• NTT VoBB customers: 3.8 million. NTT East 2.1m, NTT West 1.7m

• NTT VoIP/VoBB ARPU: 4800 yen

• Softbank BB: VoIP subscribers around 5m

Voicemail as an application has been mostly passive. I would like to see it becoming an intelligent agent analyzing a customer’s usage patterns, for instance.

For instance, when I call someone and he does not answer, I get anxious as to why the call is not being answered. There are Presence integrated functions that I can implement e.g. you get the message that the person is in a meeting or in some do-not-disturb mode etc. But that is not enough. What if my next door senior citizen neighbor has been lying dead for days and I am thinking the old man is becoming arrogant. If I hear something like ‘…This phone has not been answered for the last 8 days …’ I would want to probe further rather than just ignore it.

There must be several YouTube equivalents of audio. But they require you to be Internet users. How about those overwhelmingly huge non-Net users who want their views, and content in general, to get across?

There is a simple way to add that dimension to audio citizen journalism. Set up a premium rate phone line number and let people leave their messages. Buy a solution (and there are some readily available such solutions in the market) that convert those messages into computer audio files (WAV and maybe later converted to MP3). Just post them online. You can even have the IVR of the phone line ask the speaker to summarize the post for you.

I am going to post the TOC of the report tomorrow. But here are the main findings:

• Service providers worldwide recorded an estimated traffic volume of 302.7 billion VoIP minutes during 2Q07

• Split: 64.6 billion local, 214 billion national long distance (nld), 24.1 billion international long distance (ild)

• Local split: 61.9 bn is retail VoIP, 2.7 bn is wholesale local VoIP

The chart above lists the factors that are driving outsourcing in communication chips industry. Out of the 14 that outsource, 7 of them ranked 'Focus' as the main factor. One would have thought that cost saving is the main factor. The sample size is small (21 responses) no doubt. But then the communications chip industry is not that large. cost saving justifies outsourcing as management can see quick results in their account statements. But to really make a strategic difference, chip vendors want to outsource mainly in order to concentrate on areas where they feel they are adding value.

802.11x and Bluetooth chip market worldwide revenues during 2Q 2007 touched over a billion at $1,073 million. Nearly 39.75 million chips were shipped during the period. These chips were used for both core and CPE devices in various wireless technologies like WiFi, WiMAX, Bluetooth, Zigbee and WiBro. However, chip usage in core devices fetched more revenue compared to CPE where the usage was more in terms of units.

Broadcom leads the market with around 22% of revenue share. Freescale, Texas Instruments, CSR and STMicrosystems followed that lead in that order. The top 5 vendors in this segment made up 71% of the total revenues for the period. Consumption-wise they supplied 69% of the total chips used.

Our estimates suggest that manufacturing of about 74% of all communication chips are outsourced. The retail value of that is around $25 billion per annum. If we assume that the manufacturer gets about half of that, we are looking at a market size of $12.5 billion for chip outsourcing within communications industry alone.

TSMC, UMC, SMIC and Chartered Semiconductors are some of the leading manufacturers in communication chips and make up over 50% of the total outsourcing business. TSMC leads the market followed by UMC and SMIC in that order.

Once upon a time, data was doing the catch up and voice was the king. Then data took over. Further down the line data and voice started to merge over a common network. Today, you won't find any major fixed line service provider in the world with voice-only offering. However voice still rules in the wireless world. It remains to be seen which generic path wireless operators take with regard to data services.

Asian and North American operators are more inclined towards WiMAX and that is why the regulators in these countries are opening spectrum for WiMAX. Compared to this the European regulators are not so sure just yet. The result is that WiMAX is still weak in Europe. There have not been any major deployments in this region for WiMAX.

Total number of trials at the end of 2Q07: six active trials, one active femtocell trail, total eleven commercial UMA offerings where Kineto is involved. New service providers signed up during the recent quarter (2Q07): one

Kineto has been pushing the Femtocell FMC option lately. The company has been exploring how UMA could be applied to Femtocell. It claims that UMA is a natural fit for Femtocell. (You can run a UMA client in a femtocell back to the UMA controller connected to the public Internet. From the Femtocell to the handset there is a standard 3G protocol. So you can use a standard 3G handset. The WiFi-GSM type of FMC requires a special dual mode handset.)

Out of the 214 billion National Long Distance VoIP minutes during 2Q07 that we reported a few days back, 58.9 billion were packetized wireless trunking minutes carried by Voice-over-IP/Packet networks implemented by cellular operators worldwide. Leading carriers include AT&T, China Mobile, T-Mobile, Telefonica, and China Unicom.

But there are some other forward looking carriers in this list too. And they include Saudi Telecom, True Move (Thailand), and VimpleCom (Russia).

How many mobile WiMAX users would there be at present and what percentage of that is served by SOMA Networks platform?

Many analyst firms have estimated the size of the Mobile WiMAX market and according to analyst firm Visant Strategies, there will be expected 12 million subscribers using 802.16e powered services in 2010, with well over 50% of these being primarily "fixed" wireless customers.

As Mobile WiMAX is not yet a certified standard, there are no subscribers on certified Mobile WiMAX systems. Estimates by another analyst firm, Maravedis, count approximately one million current broadband wireless subscribers using WiMAX-ready technology. Many of these will have an upgrade path to Mobile WiMAX, but such statistics are not well understood at the moment. The industry is currently at an inflection point and is poised to take off. We anticipate this to be a multi-billion dollar market within a few years, and SOMA Networks will command a sizeable share of the market.

Notwithstanding the advantages related to chip outsourcing, there are numerous challenges. Quality of the chip delivered by the contract manufacturers remains the prime concern. Manufacturers offer virtual inspections and pre-production inspections to ensure quality remains no issue.

Another issue is reliability of the manufacturer even though in communications chips, over 50% of the production is carried out by large well known fabs. Other reasons that chip vendors highlight include the risk of their business models being easily replicated, cross cultural mix and flexibility. But these are largely concerns of startup fabless chip suppliers and/or of those who outsource to small manufacturers.

FirstHand Technologies has OEM go-to-market business model which allows it to pre-integrate with PBX systems. Integration and provisioning is a big issue in enterprise FMC. Another unique feature about Firsthand’s approach is the fact that it is not that excited about the dual mode handover type FMC. It is more focused on extending PBX features to cell phones over cellular networks. Extending PBX features over WiFi phone is an easier task in comparison. If on the other hand you have to extend them over the cellular cloud you have to deal with slow speed packet data of cellular. The technology piece is explained towards the end of this post.

In terms of the business update Firsthand has so far signed up three main OEM partnerships: NEC, Nortel, and 3Com. The three of them ship about 25% of the PBX lines worldwide. Firsthand supports both IP PBX as well as traditional PBXs, the latter being carried out through CTI interface. These OEMs are in various stages of product introduction including beta testing. Firsthand expects full product launches in Q3 and Q4 this year.

Some six years back I heard John Chambers of Cisco on CNBC telling a reporter how Internet would change our lives. The first example he gave was how one would be able to listen to his song of choice on his car stereo instantly downloaded from the internet. That got me thinking how awkward it would be to search for a song while driving, then selecting out of the search results, and then some more clicks here and there. I am sure such an application would be banned by the traffic police everywhere.

• Last blog related to 2Q07 VoIP update

• Enterprise VoIP gateways: 3.8m ports generating revenues of $127 million

• Enterprise ATAs: 1.03 million ATAs generating revenues of $55.6 million

• Enterprise IP Phones: 2.56 million phones generating revenues of $225.4 million

• Market leader is Cisco in all three categories