My predictions for 2008 ...

CPE focus for devices made for WiMAX and IPTV types of access. A multifunctional chip or SoC will probably be common for these types of network elements.

Nanometer technology will take off. We will be seeing chips developed around 65nm and 90nm for the communications elements. These chips will also be used mainly for CPE.

Pricing pressures will be significant due to increased market fragmentation and shrinking margins. Single chip solutions for ultra low cost handsets will be one of the fastest growth areas within chip segments

Chip companies will explore contextual-communication services/applications driven by the future Web 3.0 apps which for example could imply combination of GPS and georefrence optimization at chip level

Core network platforms will continue migration towards standard chips in order to reduce OPEX and consolidate data center operations. The collaborative nature of present internet services and open interfaces to share user info will catalyze the move towards standard chips

We will start seeing trials of common subscriber databases and common contact lists shared across PC, Phones (POTS and Mobile), and TV

In the US, rural telcos will continue to sustain mid tier and small IPTV vendors

Apart from telcos, MVNOs and Mobile VoIP providers will start exploring IP based TV services over mobiles

‘Pay Per Click’ advertising will catch up with IPTV

Microsoft will take lead in the middleware area

VoIP regulation in India requires the provider to be a licensed ISP in the country. So strictly speaking, SkypeOut type PC-to-Phone calls are actually illegal in India. Since technology started bypassing the regulation 1995 onwards, the ‘grey’ international call minutes have beefed up quite a bit in India. Those minutes could now be killed by the upcoming country-wide VoIP offering from BSNL. A lot of outgoing international grey traffic will come under the legal fold I suppose, just because of the scale of BSNL’s intended country wide offering. Grey minutes account for nearly 30% of India’s outgoing international voice traffic. A lot of BPO companies have recently been fined for VoIP usage.

BSNL will not confine VoIP to just the residential users. IP Centrex services are also going to be made available. STD Booths – few million small communications shops in the country that typically come with phone/fax/photocopying/bill-payment type services - will also be equipped with BSNL’s VoIP. The provider expects to sign up around 2.5 million VoIP customers which sounds a bit ambitious. Its target for broadband subscriber base was 3 million as of end December 2007. There is also a little hitch in the story. Domestic long distance calls will not be included fully in the offering. Domestic calls will be strictly PC-to-PC only or VoIP-user to VoIP-user.

BSNL is expecting 80% of the calls made by its future VoIP subscribers to be international.

I would like to confess today that I have been a little prejudiced about the small service provider segment. I thought the only thing they did was call termination (in case of VoIP). Having spoken to a few small service providers over the last couple of months, I think I have been wasting my time with the ‘Tier 1s’. The Tier 1s might have genuine concerns about scalability and robustness in new applications they test. However even an analyst needs some excitement once in a while.

Sometimes you want to taste and know about half baked stuff too. While the services offered and applications built by smaller providers/vendors may not offer 35 nines, many among us would still want to know the story. It is with this intention that we start coverage of small service providers and vendors from now on. In short that means greater coverage during 2008 for Voice 2.0 and small service provider segment.

If telecommunications is going to change for better, the change agents are likely to be smaller service providers and vendors going forward. Coverage of small service provider/vendor segment on iLocus will mostly come in form of company profiles. If you are a small telecom company and have a story to tell, just email us at info@ilocus.com and we will cover you. That simple!

Providing VoIP service since: June 2007. Pynk Global is part of Mellick Enterprises Pty Ltd that has been in the telecommunications business since 1996.

Markets served: Call Centers. The company also has its own Call Centers in Australia and Manila selling a range of telecom products. Built in-house call center product which the company later productized. Offers one-second billing to call center customers with no minimum charge. In November 2007 Pynk re-launched residential and business VoIP services.

Figures: Since launch Pynk has doubled its customer base each month, now routing over 2 million minutes per month.

Providing VoIP service and platforms since: Year 2000. ipaustria is one of the first fully licensed VoIP carriers in Austria to handle geographic phone numbers and emergency numbers/calls.

Target market: ISPs that are looking to offer private labelled VoBB services.

Figures: 37k retail residential subscribers; 65k retail business (provider portal) subscribers. Business and carrier customers in 21 European countries.

Challenges: Building service provider portal to handle phone numbers, billing, features

Areas of innovation: A self admin/quality-check tool to for VoIP connection.

If you would like to be covered on our site, please write to us.

Indoor coverage is a challenge and WiMAX is no exception. Even with WiMAX we are talking about using complementary indoor technologies like WiFi. 3G operators are looking to solve the issue with Femtocells indoors. Some of the WiMAX providers have been asked in the past to address the problem by increasing the density of macro and micro cells.

WiMAX vendor Redline’s pitch is that increasing cell density can be avoided for a while. They are pushing an interim solution, an unconventional approach. Redline believes that the optimal approach is to build a macro base station that delivers high power (4 to 10 Watts depending on frequency), then add on Matrix A and Matrix B MIMO support to provide statistical power enhancement of 6 – 10 dB for coverage and capacity and finally further enhance the system with beamforming to create an adaptive macro WiMAX base station solution.

What kind of businesses can potentially sustain free over-the-air, country-wide, Internet access over WIMAX enabled devices? I have been doing some math here. But let me touch upon the last part of the previous sentence. WIMAX enabled and WIMAX integrated devices are expected to become common by 2010. You would require WIMAX integrated phones/devices if you want to push ads down those devices. And you would have to wait till around 2010.

A country wide WIMAX network: Let us consider the United States. The operator would need around 40,000 base stations and it would cost around $300-400 dollars per site including installation. So the cost of laying down the network would be around $5 billion. Let us leave aside the OPEX for a while (likely to be slightly under a billion dollars per year, by the way). For such a network, one would need spectrum of around 50 – 70 MHz and getting that would cost anything in the range of $7-10 billion.

So the initial set up cost of such a network would be in the range of $12 – 15 billion. If we assume a realistic RoI of 10-20%, we are talking about revenue targets of $1.2 - $ 1.5 billion a year on the lower side and between $2.5 billion to $3 billion on the higher side.

It has become an accepted norm to have multiple remote controllers in your room. One for you TV, one for DVD player, one for set-top box (not all TV functionalities are available on set-top box controller), one for your music system etc. If we turned things like word processor into an application, an alarm into an application, watch into an application etc, surely we can turn these TV remote controllers into software applications that can be downloaded on to mobile phones.

You have infrared in nearly all these devices – DVDs, TVs, music systems, mobiles. I saw one demo made by UTStarcom in China last year that integrated phone with a TV remote. It is part of the IPTV offering so that when you receive a call, there is a caller ID flashing on TV and you can take call using the phone capability integrated into the TV remote controller. Not quite what I am talking about here. I am talking about the reverse engineering. Turn remote control into a software application and put it on mobile.

Providing VoIP service since: May 2006

Challenges: Managing VoIP over public Internet

Figures: 20k retail residential subscribers; 35-40 business customers; 30 active resellers worldwide

Advice for VoIP service providers: Solid business plan so that you don't waste time, but also some flexibility with respect to the local market conditions

Expected growth: The current goal is ~100K retail customers by June 2008.

Areas of innovation: embedded click-to-call capability for business websites; online callback functionality; softphone apart from handling VoIP will also integrate other online features; innovation in the area of solutions for callshop owners; networking subscribers of multiple VoIP providers.

Company URL: www.webacall.com

If you would like to be covered on our site, please write to us.

You offer turnkey IP Centrex and wholesale residential VoBB services to ISPs in EMEA. What in your experience are the main factors while scaling up the deployments?

Provisioning is becoming more and more important for our customers. The technology is obviously important but equally important is how to provision these services to the end customers. I guess security is the other main challenge while scaling up the deployments.

Your customers are telephony operators, ISPs and system integrators. So far which segment out of those customers have you seen more inclined towards VoIP services?

Broadband ISPs continue to be the segment where you see the largest uptake. VoIP is obviously the way to attract more customers and also maintain the market share besides providing a revenue stream on top of the broadband delivery.

The ISPs are obviously not going to stop at VoIP. They want to offer triple play and perhaps explore quad play as well. Are you contemplating some sort of wholesale offering for VoIP plus IPTV plus wireless?

We see IPTV coming in to allow triple play and we have a number of partnerships for that. Further when the customers will move into a quadplay scenario we have secured a number of partnerships to allow them to deliver a quadplay service. We have worked with a number of billing companies to ensure we can bring a single billing BSS/OSS platform to help them deliver triple play and quadplay.

What is your wireless VoIP strategy?

Certainly we have seen a lot of development going on in VoIP over WiMAX. We are watching that technology very closely. We have already secured one customer which is a wireless operator deploying VoIP successfully using this technology. We see it as a very interesting alternative where DSL network isn’t well established.

You had a quiet launch. Is the obvious name of the company - phone.com - compensating for the quiet launch?

Obviously when you hear the company name it stands out. That is what attracted me as well. If you have a quality product with such a name it lends added credibility and perception of a bigger company. All of that leads to a lower marketing cost direct and indirect. We are three months into our launch. We see steady growth in traffic and customer acquisition. There is direct typing traffic and search results traffic via Google where the word phone is one of the most searched word. People find us very little advertising.

Do you think some of these visitors could be people who want to purchase phone.com domain and their search leads them to your page?

I don’t think so. The history of this domain name is interesting. It used to belong to a company that was acquired a long time ago and it sat idle for few years. After it was re-purchased and I came in, we kept a sign-up page where we did a survey of visitors asking them why they came to the site and what kind of phone related services they are interested in. We got statistics. Among those answers, there were a few questions about purchasing the domain. But whoever is serious about purchasing a domain knows how to find the owner. A vast majority of the traffic we had pre launch was people looking for phone services.

If you are getting good amount of traffic you could monetize that by offering advertising space. Is that a possibility in the future?

When you hear the name phone.com everybody has suggestions what to do with it. We took a decision to offer services ourselves rather than being a referral site for telecom services of other companies. Looking at advertising models, the answer is yes this is definitely something we will be exploring.

As part of eBay strategy to make money from Skype, the VoIP provider’s softphone has found place on Sony PSP. Although it is widely believed the Sony is getting no revenue out of this, it is more likely to be the opposite. Skype icon on PSP is not going to drive PSP sales. That is almost for sure. Those who play games are not after cheap calls, or do not want to talk when they play. There is some use of voice avatars among the gamers but I am not aware of Skype offering such capability just yet.

Sony has since long been apprehensive about its business of consumer electronics items like cameras which are conveniently becoming part of mobile phones now. Since the likes of Nokia have been invading its territories, it is only fair that Sony has a go at telephonizing its gadgets. They will certainly have kids on their side. And who knows when those kids grow up with PSP as their companion, they might decide to stick to it rather than buy the mobile phone.

Providing IPTV service since: Early 2004

Most memorable challenges so far: Being an early adopter in the IPTV space, the majority of Surewest challenges have been technology related. The company had to work closely with all of its technology partners to ensure workable integration across all of the components that comprise the IPTV ecosystem.

Number of IPTV subscribers: Over 20,000

Sources IPTV equipment from:
Set-top box: Amino
Middleware: Minerva
Encoder: Variety of MPEG-2 encoders; uses BigBand BMR for IP encapsulation and rate clamping
VOD Server: Kasenna
Test and Monitoring: Ineoquest
Security: Irdeto Access, Widevine
Advertising Server: Outsourced through Viamedia

Lebanon based Ahwar, the instant messaging client in Arabic medium, has signed up over 200k subscribers during the two months since its launch. With the rate it is securing new subscriptions, the company is targeting 2 million users by the end of 2008.

Ahwar offers text chat as well as VoIP and video calls. VoIP comes in form of PC-to-PC as well as PC-to-Phone. The VoIP capability has been developed using open source SIP stacks and interoperates with Yahoo, MSN, Google and other major voice-enabled IM clients.

Providing IPTV service since: 4th quarter 2005

Challenges faced in deploying IPTV: The biggest challenge has been the expense of deploying IPTV. Being a small company and not having the deep pockets to afford leading edge technology has been the biggest challenge, as well as the buying power of ordering large quantities of set top boxes. At the same time, working through the uncharted territories, experimenting with unproven technology and trying to figure out what works with the company’s network has also been a major challenge related to IPTV.

Number of IPTV subscribers: 740

IPTV equipment sourced from:
Set-top box: Amino
Middleware: Minerva
Encoder: Optibase

Faroese Telecom is the incumbent operator of Faroe Islands, a small cluster of 18 islands with population of around 50,000. Its VoIP trial – which envisaged enterprise and residential VoIP offerings - prolonged unnecessarily for over 2 years. The company had thought a thousand flowers would blossom. So far they have not.

The delays caused in IMS enabled miracles has not discouraged Faroese Telecom. The company has found a ‘niche’ market to offer VoIP services. The main economic activity of the islanders is fishing and people remain onboard almost throughout the year to make a living. Monetizing the communications requirements of this segment, Faroese Telecom is offering telephony services to the ships and vessels which operate in the surrounding seas. It has a sister concern that gives these ships the satellite broadband connectivity. Faroese Telecom offers VoIP over these satellite broadband connections.

I have seen many hypotheses surrounding Cisco’s possible use of Tribe.net and Five Across assets. Some see Cisco enabling internal social networks for its corporate customers. Some see Cisco enabling its corporate customers to implement sites that gather customer feedback on their products e.g. a social networking site sponsored by Sony to bring together its customers in order to better organise the feedback and dialog with its end customers. The funniest one I have come across is that acquiring social networking software will drive more traffic to Cisco routers and switches. Cisco's official stance is that it is experimenting with some kind of an entertaiment operating system which it will be selling to media companies rather than telcos.

Anyway, I am going to add another hypothesis today. Cisco will (or should) add social networking software as an application server that inter-works with underlying telco/enterprise call management platform (There you go. We can’t have enough of telephony). Result: its telco customers looking for a more comprehensive Internet services strategy will have an integrated and off-the-shelf solution for offering social networking capability to their customers.

Telcos can potentially offer more meaningful social networking experience. They can optimize the experience around local communities to start with. Majority of Internet users don’t have an international friend circle. More than half of the Internet users worldwide use email addresses provided by their ISP rather than having a yahoo / gmail / hotmail account. What it means is that telcos/ISPs have potential to leverage their relationship with the customers to engage them with telco sponsored social networking sites.

Social networking services could also be places where a telco could offer/test its Voice 2.0 services. That way a telco could start small with a Voice 2.0 application and not require a monster 2 million BHCA Voice 2.0 platform inside its network.

Providing IPTV service since: September 2006

Service Details: Launched its Freewire IPTV service in September 2006 to UK University students living in on-campus accommodation - university halls of residence connected to JANET, the UK universities’ high speed broadband network -- and the initial launch was all free-to-air content. A pay service was launched in September 2007, the start of the new academic year, again to students in halls of residence. The next phase of development will see the Freewire being rolled out as an off-campus residential proposition using DSL in March 2008. Freewire also plans a white label IPTV solution.

Challenges: Content rights issues and regulatory environment as the service expands internationally. These aspects are often very different and provide a new set of challenges in each region that have a bearing on time to market. Currently Inuk Networks is launching Freewire in the Republic of Ireland and Canada.

Number of IPTV subscribers: Freewire is currently available in over 100,000 student bedrooms in halls of residence across the UK.

IPTV equipment vendors:
Set-top box: Telsey, plus Inuk Networks own developed Virtual Set Top Box application for PCs & Macs
Middleware: Zignal
Encoder: Tandberg
VOD Server: Edgeware
Test and Monitoring: Various
Security: Widevine
Advertising Server: PacketVision

(A) Cable companies have one of the best IP infrastructures in the carrier space. (B) Cable companies in the US are buying wireless spectrum and it seems wireless is very important to them going forward. Put A and B together and you will have Cable companies in the US utilizing their IP networks for wireless backhaul.

What has that got to do with their relationship with Sprint through the joint venture Pivot? Imagine if a joint Sprint-Comcast customer (Comcast effectively operating as MVNO here utilizing Sprint’s wireless network) used a dual mode phone within a WiFi zone and migrated the call over to IP using Comcast’s FMC server thus bypassing Sprint’s wireless network? That would leave Sprint high and dry. Sprint is apprehensive about the FMC related converged offerings being tested by Cable companies. Basically Sprint is not allowing for dual-mode single-number type of services within this joint venture.

The underlying theme in your products seems presence, and presence integrated with public networks. Are we ready for presence integration with something like PSTN and public cellular networks?

I think the bigger question is integration of presence with applications and not the networks. When we started out we used to go out there and say we have an application for your networks when really we should have been saying we have solution that applies presence to your applications. So for instance we have built conference calling application for Facebook that uses some presence features. That has generated a substantial amount of interest and is acquiring users at a fairly quick rate. It is much more successful than anything we have done in past. So the lesson learnt was that presence is applicable to lots of places and it is just the matter of bringing it to those places.

Your Talk-Now for Blackberry and Relevance Engine product lines are two different products in the sense that one is for consumers and other for operators or OEM vendors. How do you plan to handle both consumers and operators as customers?

We are not focussed on targeting operator customers at this point of time. It is long ritual to sign up an operator. In the end we decided to provide the service to end users directly.

Any figures you can share with us about your Talk-Now downloads so far?

We haven’t announced publicly about Talk-Now downloads but I can tell you our Facebook downloads has exceeded 90,000 users.

How is Broadvox positioned in the next-generation telecom market?

Our vision is to become the leading provider of SIP Trunking services to SMBs and enterprises in North America. Currently, we have an excellent business as a carrier's carrier but we seen the rapid adoption of IP telephony as an opportunity to evolve our business. We can provide the same cost savings and reliability of our traditional business to this rapidly growing business customer base.

What specific markets does Broadvox SIP Trunking serve?

Any business wanting to leverage their new investment in IP Telephony equipment such as an IP PBX or SIP Phones and any business using an IAD to connect their legacy gear to an IP Server.

How does Broadvox tailor their solution to the various markets it serves?

Our SIP Trunking package can be customized based upon the desired number of simultaneous calls, long distance volumes, in-bound and out-bound toll-free volumes, number of DIDs, local number portability requirements, and other capabilities. Each business defines its service package with Broadvox, not the other way around. Moreover, as we are using an indirect channel of Value Added Resellers to sell the product, this gives then greater flexibility in address the unique demands of their customers.

XO’s new pricing plans for integrated data+voice “eliminate pricing based on the number of voice lines.” The first thing that came to my mind was how the number of voice lines can be irrelevant when XO is paying incrementally for each of those lines it sources from its vendor. So let me build up the momentum a bit more. If I am an enterprise customer and I have 3-liner-8-extension hosted PBX arrangement with XO, I will either pay for the 8 extensions or I will pay for 3 lines (or maximum pre-set simultaneous calls limit for an enterprise). Now if I make 10 simultaneous calls, that means I am engaging 10 lines on XO platform. Surely I can’t get 10 lines for the price of 3 I previously subscribed to. So what is the catch?

655k VoIP customers as of end 4Q07

£157.2m (VoIP plus free broadband bundled in) revenue for the year 2007. £72m loss

Customer service and provisioning issues with VoIP

Total broadband base over 2.6m through various offerings. Total on-net base 1.6m on its own unbundled network

No increase in churn levels despite significant numbers of VoIP customers coming out of their initial 18 month contract

2.7m voice customers which includes TDM voice customers

1533 exchanges enabled with VoIP capability (see this earlier post)

We need a web server utility that recognizes the origin of the visitor (like in ad servers) or scans the browser language of the visitor and then automatically displays the content in relevant language. That would be more productive than offering manual selection of alternative languages for your international website audience. I suspect these things might be available out there but my browsing experience has not thrown up such instances.

Extending the idea further, when web based telephony rules, you could be greeted by your local language by the IVR system when you click-and-call an overseas company. “No one speaks English here. Please hang up.” You could probably do that right now even with telephony 0.0 utilizing the Caller ID header.

I very much fancy the Ribbit idea. Their focus on bringing together Voice 2.0 developers and effectively offering an aggregated new telephony platform sounds exciting. The Flash APIs for developers jazz up the offering a bit more. Ribbit however goes direct to the consumer. And that is probably more practical because telcos are horrendously complacent when it comes to deploying new applications.

I have not come across a vendor that aggregates Voice 2.0 applications and offer an integrated platform that can be licensed out to telcos. There are companies that offer their Voice 2.0 application both to consumers as well as license them out to an operator. However the aggregated platform in the latter category cannot be found so far. Most vendors think that having an underlying Class 5 type feature server (and may be some call control element as well) is a pre-requisite for offering such an aggregated platform. I have to disagree with that.

Here is NSN view on why FMC has not cemented itself just yet. This in reply to a question posed by an analyst during a recent briefing.

Most of the operators are still in their early phase with convergence and there haven't been many commercial FMC services offered to the subscribers. FMC has two main aspects for the operators; cost savings through network simplification and new revenues through converged end-user services. So far the focus has been more on planning and implementing the network simplification, bringing in IP backbone networks into use for all the access networks. In network simplification there are lots of early phase activities going on.

For the end-users the voice convergence has not yet really started. Operators are modernizing the PSTN networks and VoIP is being used more and more for the fixed voice. But in the mobile networks circuit switched voice is still the most efficient way of providing the voice services. Converged services have been based more on non-voice services like email and browsing, where we have started to see increasing usage amongst the smartphone users.

VoIP made its mark via PC in the mid nineties. Then it was applied to phone-to-phone environment and that took the glory. Packetization of international gateways. As we passed through the telecom downturn, VoIP effected a quiet revolution at the core of telecom networks. Packetization of the core. During those latter two stages end user VoIP took a back seat. When broadband helped VoIP push itself to the edge, we had the phone+ata take the glory. In parallel however Skype kept the PC interface alive and kicking.

With Voice 2.0, the PC is back. Most Voice 2.0 applications require interaction through web, which is best accessed over a PC. This is just a little bit of history repeating. Expect PC ergonomics to drive VoIP for quite some time till Voice over Wireless Broadband (VoWB) passes the VoIP torch to the new computing platform, the mobile phone.

Technology developed by Blinkx, the video search company, can perhaps be an answer to the growing concerns over unwanted video content on the Internet. That is because Blinkx crawler virtually sees the content of the video that it is trying to tag and categorize for search purposes. The crawler technology visualizes and analyses the clip – like a human eye would do, according to the company - and accordingly stores the information in an index that is accessed in the retrieval mechanism to obtain the results.

There are several among us that are concerned about the unwanted video content on the Internet. Adult video content is obviously at the top of the list. May be the level of concern varies across cultures but these issues have to be dealt with in a sensitive manner across societies. Unfortunately we have not come up with a mechanism to really filter things properly. At most the adult websites will carry a statutory warning that you have to be over 18 in order to access the content. Neither that nor the parental control software has proven to carry out filtering adequately.

Great to see one of my predictions for 2008 coming true. AT&T is re-packaging and re-launching its VoIP offering. Having met less than exciting demand for its VoIP over DSL, the telco is re-launching the same offering over its much celebrated U-Verse fiber network. And that makes a lot of sense. The voice quality would certainly be much better. However pricing is rather depressing: $40 per month just for the voice plan!!

T-Mobile USA has sold over 250k FMC integrated handsets. The company does not reveal how many of those have actually taken up the FMC service. Sources suggest that about half the people who buy the phone sign up for the service.

If half those who bought the FMC pre-integrated handsets subscribed, that would still make T-Mobile the second biggest consumer FMC provider after Orange. There are not many commercial FMC offers out there. The ones offered by the fixed operators did not bother iLocus number crunchers too much. BT is still going at around 50k subscribers. The last reported figure that came from Orange was 573k FMC subscribers at the end of October 2007.

Coming back to T-Mobile dual mode FMC service Hotspot@Home, which is offered across the US, it is possible to buy a dual mode FMC integrated phone and not subscribe to the FMC service. There is a subtle point however. If you buy a dual-mode handset, it will work over WiFi anyway (clever T-Mobile …. we will tell you in a minute why). So you effectively get to “try” FMC. For example, you can buy Samsung dual mode phone through T-Mobile and operate it over WiFi at home. You don’t get any additional calling benefits. The calls that you make over WiFi are just like the calls you make over GSM network. They still come out of your bucket of minutes. This is T-Mobile conveniently offloading calls to WiFi thanks to the fantastic FMC technology.

Let us start with numbers. What is the number of customers you have, number of POPs, amount of traffic you are handling etc.

These are numbers that we are comfortable to release: We have over 500 million numbers in the registry. That consists of a combination of blocks of numbers allocated to service providers and individual numbers. We have now just over 400 service providers on a worldwide basis utilizing our services from pure ENUM registry exchange all the way up to a complete federation model whereby we are delivering not just registry services but also signalling services, security services and other wide range functions. We have millions of minutes flowing through our systems and that continues to grow.

How many ENUM queries are we talking about per month?

We have a few hundred million ENUM queries occurring on a monthly basis.

These few hundred million ENUM queries you handle exclude the in-country federations. Right?

Yes that is correct.

Out of the 400 service providers that you serve, approximately how many utilize your capabilities other than registry queries, stuff like call signalling and security?

Probably around 100+.

If the US does indeed catch recession I imagine the winners in telecom are once again going to be the bigger players in the market. Let us take the case of enterprise VoIP, a segment that has been nicely growing since 2002. Obviously with recession like situation enterprises will cut their communication costs. IT people will see some job cuts too. If I was hired to devise a defensive strategy for the IT staff evaluating VoIP proposals, even I would suggest they go for solutions perceived as safe if they seek security of their jobs.

‘Safe’ solutions for enterprise VoIP come from Cisco and Avaya. If you procure solutions from these vendors and the system malfunctions, you don’t get the blame. If on the other hand you procure solutions from a smaller vendor and the system malfunctions – well you might lose the job. I hear that even the small businesses which rarely went to these large vendors for a VoIP solution have started looking towards them. This is affecting the business of small VoIP vendors.

636k VoBB subscribers at end 4Q07. These now include business customers as well

62.5k new additions during 4Q07

Business VoIP is a year old service. It is a 1-4 lines service.

Previous quarter revenues of $52.0 million from VoIP. Previous quarter household penetration rate of VoIP was 23%

Verizon’s FiOS TV had 943k subscribers at the end of 2007, making it the third largest IPTV provider after France Telecom and PCCW. IPTV has been a part of Verizon FiOS TV service since its launch in 2005. FiOS TV utilizes a hybrid QAM / IP platform that uses QAM RF to deliver broadcast channel programming and IPTV to deliver video on-demand content. The middleware is largely proprietary with certain elements inherited from Microsoft. The set top box is sourced from Motorola.

An innovate aspect of Verizon’s FiOS TV service is the two-part realization of the whole-home media concept, something that no other (major) IPTV provider or MSO currently offers. Part one is “Home Media DVR” service, a multi-room DVR solution that is delivered by the IP-based FiOS in-home LAN and its IP-enabled STBs. It allows a customer to view different DVR-stored programs on up to three STB-connected TVs (DVR STB + 2 remote terminal STBs) simultaneously in their home with trick-play control and the ability to pause a program in one room and resume watching it in another room. Currently this service delivers SD streams to the two remote terminal STBs but Verizon has disclosed its near-term plans to add HD streaming to Home Media DVR in 1H08.

Providing IPTV service since: Launched first of eight phases in June 2004. Have IPTV deployed in 72 towns in Central and Western Oklahoma.

Major challenges so far: Integration of IPTV solutions, keeping subscriber acquisition costs down in order to compete with incumbent Cable providers and the DBS providers.

Number of IPTV subscribers: 8500

IPTV equipment vendors:
Set-top box: Amino, Entone and Scientific Atlanta
Middleware: Myrio (NSN)
Encoder: Tut/Motorola
VOD Server: Harmonic
Test and Monitoring: In house
Security: Verimatrix
Advertising Server: ETI

Etisalat is one of the very few operators who still manage to mint money from international calls. And it manages to do so because almost 80 percent among the country’s population of 3 million are expatriates who helped generate 4 billion minutes of international calls during 2007. Any form of arbitrage would threaten that healthy revenue stream. That is why VoIP is banned in the UAE. The official ban applies to VoIP as a competitive international service offered to UAE residents in form of offerings such as Skype. Other forms of VoIP, such as the local loop voice over broadband, do not have to be kept at bay however.

Etisalat has therefore kept itself VoIP-ready from technology standpoint. It conveyed to the regulator yesterday that its network is ready for VoIP based services. The regulator however is not entirely comfortable yet in terms of a regulatory framework that tackles VoIP based emergency services, legal intercept, and interconnection issues. The UAE incumbent had been trialing VoBB on top of Marconi’s softswitch a few years back before the acquisition of the latter by Ericsson. It has lately been field-testing various VoIP services for enterprises including IP Centrex. Etisalat’s competitor du also uses VoIP internally in form of softstwithes to transit/refile the regional voice traffic of the Middle East carriers.

Is there a need for a dedicated video search engine? People on Internet usually prefer to use just one search engine?

Search is all down to relevance and accuracy. The existing search engines were developed for text search and are optimized for textual web pages. Video search engines are optimized for analyzing all the components – moving pictures and sounds. This is where the technology from Blinkx comes into play. Our technology literally watches and listens to the video. We are able to understand frame by frame what is inside that video. That understanding allows us to deliver higher accuracy level and position on the video searching.

But Google and MSN type companies have also added video search capabilities.

If you look at Google for example, its focus on video search is around searching their own video content from YouTube and a little bit of external content. Yahoo has a video search engine which they acquired 8 years ago. That has never pushed on the front page because the accuracy is low. There have been products in the video space but not truly video search engines.

The only big player having a true video search engine is AOL. AOL got it by acquiring Truveo 3 years ago. But even AOL has not really pushed video search capabilities because Truveo is also dependent on reading and analyzing text rather than looking at the video content itself. For all these reasons the average user using text search engine isn’t really finding the videos they want.

How does your search engine really crawl the videos? I mean how is the visual analysis carried out?

We have a web spider which basically crawls the web. Every time it finds a web page it analyses the page for a video and if it finds video files on that page it goes further looking at how that video is positioned and analyses it from a visual point of view. Based on that it extracts words or text around that, which seem relevant to the video: things like titles, descriptions, tags, etc.

Once that is done, our software plays back that video. It is similar to a human being watching that video. We literally sit back and press the play button. When the video plays, the Blinkx system analyses the visual elements. It looks at things like text on the screen, it can also do some limited facial recognition as well that can spot the faces of famous people. Other than this we also analyze the speech track of the video. That helps in knowing the words spoken in the video.

We check all this information - text, visuals, speech - and combine them into a single conceptual record about the video. It checks what the video seems to be about, what the main context is that is all indexed into a search engine index. So when you come along with your search, you can do exact search to all these conceptual records we have.

Our next webinar will discuss FMC Security. This is our second webinar in the series focused on Asia Pacific service provider community. For a while we will stick to just one webinar a month. It seems like a lot of work. Lot of telemarketing for sure. During the webinar I will be providing 2007 update on FMC services worldwide while Acme Packet will share its experience with FMC security issues that it has stumbled upon.

If you would like to co-sponsor the upcoming webinar along with Acme Packet, let us know. We have several webinar topics lined up that range from VoIP to IPTV to Mobile Data. Webinars on demand are also possible. Interested parties may get in touch with us. Again the focus for a while will be Asia Pacific.

FMC during 2007 has mostly been about the femtocell trials. UMA based consumer offerings did gain a little ground. But no way near what the industry had expected. SIP/VCC based FMC offerings have altogether been a disappointment. It is the latter category of FMC that NewStep enables. New funding for a VCC based vendor is therefore a bit of a surprise for me.

NewStep however sees things differently. According to the vendor, the market is moving in terms of FMC adoption. In particular it sees encouraging signs in Europe. And that is where most the new funding will be spent. Last time NewStep announced finding was in October 2006 when the company closed its Series B round. The $9.5 million Series C round it announced yesterday takes the total funding to $36.5 million. The company is also bringing in the new CEO.

Here are a few data points related to NewStep:

Yearly revenues of £32.3m, Y/Y increase of 27%. Gross margins 35.1%. Profit before tax £1.4m

598k IPTV Set-Top Boxes shipped during the year. Nearly all Mpeg-2 based products. Y/Y increase in shipment was 35%

On a cumulative basis, Amino has delivered 1.5 million IPTV STBs to over 1800 telcos (mostly tier 2 and tier 3)

Telecom Italia has separated the handover part from the FMC equation with its revamped UNICA offering what was re-launched in October 2007.

The new service is basically a Voice-over-WiFi offering for fixed Voice-over-Broadband customers. UNICA is therefore effectively a way to put on the same handset a mobile number and a fixed number. It is not a single number service as one would typically expect from an FMC offering. There is no handover involved between GSM and WiFi. So if you start the conversation with WiFi phone you initiate it using your VoBB number. When you move over to the GSM cloud you disconnect and redial using the mobile number.