DSP Group is one of very few companies out there that is promoting an alternative to VoIP-over-WiFi. It has significantly enhanced its processor to handle advanced VoIP functionalities. DSP Group DECT VoIP processor will, for instance, enable D-Link to develop features including PBX function, use of Cordless HS for reception, issue VoIP calls, having Wideband audio calls, having data over the cordless etc.
DECT is up against VoIP—over-WiFi in the residential market. However DECT admittedly provides better quality, longer stand-by, longer talk time and much better range than WiFi.
There are a few new applications where DECT is / will be integrated over the coming year. These include IAD (integration of DSLx with Home routing), eMTA, Infotainment Tablets/Handsets. These new applications will boost again the growth of DECT, which had otherwise seen a steady decline in Europe, its primary market.
DECT is also able to compete with WiFi in FMC, according to DSP Group. The company is cooperating with few OEMs and a big operator in a development of CDMA Mobile Handset that has DECT function. This Handset will use DECT in the home and CDMA outside of the home. This solution will surely have an edge in terms of voice quality at home.
Hosted VoIP provider, Alteva, recently started deployment of its in-house developed FMC solution. Most FMC applications hand the call off to a PSTN number such as cell phone. Once the call is answered on this device there is – in majority of the cases - no way to have the call transferred back to the employee corporate phone system. Alteva FMC solution makes a small improvement. A call can be transferred from cell phone to office phone and then back to another extension on the customer’s corporate phone system, enabling a better FMC experience.
There is however a little inconvenience. If the user drops desk phone and picks mobile phone, the transfer is done manually. But Alteva us looking to make this feature automatic one day.
According to Alteva, it is North America’s largest enterprise hosted VoIP provider, with clients in all 50 US states and 9 countries. The customers come from a variety of different verticals including retail, finance, accounting, broker dealers, private investigation, call centers and more.
According to the shipment numbers we have from vendors, there was an appreciable sequential decline from 4Q08 to 1Q09 in the volume of carrier VoIP equipment sales. Softswitch components, which comprise the bulk of carrier VoIP equipment, saw sales decline from $897 million in 4Q08 to $708 million in 1Q09. That represents a Q-o-Q decline of over 21 percent.
The worst hit among the softswitch components was the Class 4 softswitch which saw sales declining by over 40 percent.
Ever since Skype announced support for IP PBXs, we have seen some interest in how best to tie in Skype in the PBX environment. Integrating Skype can be tricky because Skype is a P2P application. XCast Labs is a hosted VoIP provider that recently announced $2.7 million funding. This is how they tie in Skype:
1. Skype user dials Skype-to-Skype call using Skype username of matrix.us.
2. Call goes to XCast hosted platform, where the company has a gateway that does Skype to SIP translation.
3. XCast then prompts the caller to enter the XCast phone number they wish to call. Thereafter there is an onnet transfer. Pretty much like calling a PBX auto-attendant and entering a long extension number.
4. XCast platform sees the Skype user name that is calling and assigns them permanently an internal XSID (XCast-Skype ID). XCast customers can dial that XSID and XCast platform translates the SIP call into an outbound Skype-to-Skype call to the Skype user name that is linked to the XSID. The XSID is a number that has no possible conflict with any PSTN number.
Dialing with a username other than your own is obviously not convenient. Gateway translation, which is how most developers have so far been handling Skype integration, sounds like old days of PC-to-Phone. Assigning unique number to Skype caller sounds like inefficient use of resources.
XCast has obviously made a bold attempt to integrate Skype into its platform. I am sure the funding is not for the Skype integration. We have not seen any platform companies that have been able to seamlessly integrate Skype due to its P2P nature. Ideally speaking Skype would have to build PBX features by itself and sell a virtual PBX services package.
XCast is a voice+video telephony service provider serving 50,000 voice lines. It carries about 50 million minutes of traffic per month. With the help of recent funding the service provider plans to triple its network capacity to be able to handle 10m minutes per day. The company also plans to target Apple iPhone users with video telephony apps.
Veraz will be offering a joint prepaid wholesale settlement/billing solution in partnership with Comarch. Prepaid wholesale must be gaining traction given the current economic recession. It becomes a requirement for carriers looking to offer service to those carriers that would not pass the necessary credit checks for a post-paid account. Pre-paid billing is also attractive to the wholesale carrier’s customers as a way to manage their interconnection costs.
According to Veraz, there has been a fairly limited deployment of carrier pre-paid billing systems, mainly because of the expense and complexity of traditional IN SCP based solutions. Those are some of the technical challenges that the vendor solves.
Veraz is also able to offer the prepaid wholesale billing solution as part of its IPX offering. Veraz’s SBC platform can be deployed as part of this solution.
The Veraz Service Broker provides standards compliant SIP and INAP interfaces for integration with VoIP softswitches and legacy PSTN switches. Between the Veraz Service Broker and the Comarch application the interface is DIAMETER. Comarch’s billing application supports relevant OSS/BSS standards and has been deployed in tier one carriers.
Agnity recently announced deployment of its nextgen SCP and applications such as LNP, Advanced Toll Free, Voice VPN, Account Code, Least Cost Routing, etc at Mexico’s leading telco, Alestra. Agnity has implemented its nextgen SCP solution at various telco networks worldwide.
There is certain overlap between nextgen SCP and SCIM. Agnity’s SCP works as an Application Server with SCIM functionality included. SCIM on the other hand provides only the service brokering functionality in IMS environment.
The business case for nextgen SCPs has been debated over the years. You have carriers who want to offer IP services across their TDM networks. But you also have those who want to utilize their existing IN applications within the TDM cloud and extend them over to IP networks. Nextgen SCPs and SCIMs are used to achieve that. If the service runs in network agnostic environment, this can result in huge CAPEX and OPEX reduction.
There are actually certain applications like LNP that mandate the use of nextgen SCP technology e.g. a carrier looking to offer LNP of a VoIP line/number over to some TDM line and vice versa. LNP is just an example of its kind. Nextgen SCPs are able to migrate most legacy applications to nextgen environment.
One SCP feature that we would like to see is to port a consumer wire line number to a wireless phone since it is all IP in 3G. In case of Agnity’s solution, the LNP feature provides the generic application logic and DB implementation. It is the carriers who own and manage the specific deployment aspects. As of now, the LNP DBs of wire line and wireless phone numbers are maintained separately.
Acme Packet has made a smart move acquiring enterprise-SBC company, Covergence. (Here is one of my writeups on Covergence). I think a smarter bet would be buying Nortel’s carrier VoIP assets, particularly the MCS platform. Acme Packet needs to become more diversified. It cannot conquer the market with a point solution.
The company is focussed on SBCs only. That is both its strength as well as its weakness. It is a weakness because some of the bigger guys have chosen to embed the SBC capability in routers and media gateways. The other bigger players that have not done so will surely come up with their own SBC product in due course. Acquiring a softswitch/feature-server could initially put Acme in competition with some of its main OEM channel partners like Alcatel-Lucent and NSN. This could perhaps lock Acme out of OEM deals. However, as mentioned, these vendors will eventually bring out their own SBC platforms. So Acme will be locked out of OEM deals anyway at some point. It therefore makes sense to be prepared for such a scenario. The company has to therefore diversify.
The best opportunity right now is to acquire Nortel’s VoIP assets (assuming Acme Packet can afford the price). Nortel’s VoIP platform is a market leader. It is certainly popular among the cable companies, which means Acme could get its foot inside the cable doors. I think with Nortel VoIP assets acquisition, we can (1) have another major non-legacy VoIP startup besides Sonus in the market, and such as deal will (2) secure the future of Acme Packet as a diversified telecom vendor, which the market badly needs right now.
NSN, Huawei and Genband are supposed to have considered acquiring Nortel’s VoIP assets. It does not make sense for Genband to acquire softswithes and feature servers since they have previously offloaded such assets. Huawei certainly won’t gain much in terms of solution capabilities. I think NSN might still be interested because it is aggressively pursuing cable business in North America. Besides that, Nortel is slightly ahead of NSN in terms of SIP capability in its platforms, which is something NSN might benefit from.
I am glad eBay has decided to spin off Skype. This was also my first choice a year ago. The logic is quite straightforward: there is no reason to beat the dead horse looking for some supernatural synergy with Googles and eBays. Let us give it a rest and recognize Skype in its own right. Besides, Skype should not remain one company’s asset anymore. It belongs to 400 million of us worldwide. It should be declared a World Heritage.
Imagine someone like Google or BT acquiring the asset and then killing it like they usually do! Skype is THE company that took VoIP to consumers. While tonnes of VoIP equipment are deployed in carrier networks worldwide, the predominant consumer VoIP interface continues to be Skype. There are no doubt certain weaknesses in the business model of the company. A year ago I would have pointed out the lack of SME offering and bad mobile VoIP strategy. But I think Skype is progressing well along both those fronts. And these offerings should help it generate better margin revenues and not rely on the PSTN termination alone. There also seems to be some lull in the developer activity. Perhaps web friendly APIs and better developer promotion strategies would come in handy.
I am copying below a link to one of my papers that contains a section on Skype arguing why Skype should be one of the 10 most influential VoIP companies ever:
FreeSwitch, the open source VoIP platform designed for carriers, has announced the latest version. The biggest addition is a telephony abstraction library that normalizes several existing voice protocols into a single API. The main reason FreesSwitch developed module was to get H323 support. According to FreeSwitch, it is also possible to expand the module to support other protocols such as IAX2 and SIP.
The open source project has also added the Least Cost Routing module in its latest version. Some of the service providers using FreeSwitch include Teliax, tollfreegateway.com, Truphone, Paetec and a few more that are migrating to FreeSwitch but have not announced it yet.
There are three main open source VoIP platforms in use. These include Asterisk, FreeSwitch, and OpenSER. Asterisk is mostly in use in the enterprise environment. FreeSwitch and OpenSER are used by carriers. OpenSER is a proxy that routes SIP packets. FreeSwitch is a b2bua that carries media and interconnects networks. Both work together to form a complete system.
FreeSwitch developer community seems to be growing on a regular basis. It’s IRC channel (irc.freenode.net #freeswitch) has close to 200 members in it on a daily basis. There are over a thousand members on the mailing list. Both those numbers have doubled in the last 6 months.
Sonus announced a GSMA-compliant IPX interconnect solution, Mobile SecurEdge. The IP-peering solution engineered for mobile operators addresses security issues related to international roaming and national peering traffic.
Sonus is up against SBC incumbent Acme Packet that sell standalone SBCs, which take care of security issues. In mobile networks, however, Sonus might have the edge having worked on large networks including AT&T Cingular. In these wireless network accounts Sonus has integrated IPX capabilities with MSC elements.
IPX services will launch this year. IPX services are supposed to drive down the mobile interconnection charges, thus enabling mobile operators to maintain margins. Carrier charges for roaming international calls is expected to decline significantly.
There are several GSMA IPX trials going on worldwide. Sonus has been involved in a number these. One of its customer, Tata Communications, recently concluded the IPX trials. Although Sonus did not reveal the remaining customers trialing its IPX solution, our guess is Belgacom ICS and COLT might be other Sonus customers running IPX trials.
ip.access, the developers of picocell and femtocell solution, last week announced that it had teamed up with SPIE, provider of professional services to mobile operators in France, to speed up the deployment of picocells at SFR. The existing deployment at SFR is the first IP-based picocell commercial offering in France.
The offering is part of SFR’s ‘One Office’ service which sets it apart from other indoor repeater solutions. The exact number of picocell users was not revealed by the companies. However they are supposed to include both large (> 10000 employees) and small enterprises. Apart from SFR, over 40 mobile network operators worldwide have deployed ip.access picocells.
SPIE has a broad presence in France with potential to reach SFR customers. Moreover, SPIE will provide assistance in deploying ADSL lines that connects picocells to SFR’s core network.
Repeaters and picocells are designed to solve similar problems: indoor coverage. So where lies the difference? A picocell has the ability to improve data rates and capacity while as repeaters only improve coverage. There is the possibility to offer in-building bespoke tariffs with picocells, which is not possible with repeaters.
Acme Packet, the incumbent SBC vendor, will be providing the security aspects for Verizon Wireless’ new VoIP offering. Multiple Net-Net 4000 series SBCs from Acme Packet are deployed in Verizon data centers at both the access and interconnect borders. The SBC vendor has seen quite a shift – in terms of deployment - from carrier-to-carrier SBC application over to the access application where it supports millions of VoIP end users globally.
This could be for the first time that Acme Packet has found its way inside the Verizon network. However as an incumbent SBC vendor, the company has been providing SBC enabled carrier interconnect solutions to most major telcos. Within the new VoIP offering, for instance, Verizon uses a wholesale e911 provider and Acme’s SBC manages the interconnect between Verizon and that provider, including routing the emergency calls.
Coiler Corporation, Taiwan based telecom equipment manufacturer, is positioning the latest version of its indoor repeater as a potential competitor to femtocell. The company introduced the new indoor repeater last week. Basically, both femtocell and Coiler type products are designed to solve similar problems. They are meant to solve coverage problem indoors, enabling uninterrupted mobile communication.
There is one difference though. While femtocell connects to the wireline broadband networks for data traffic and some voice offload, Coiler product transports all data via cellular networks. However Coiler is right in pointing out that the primary function of femtocell in the present market context is precisely the coverage issue – an issue Coiler claims to address better.
If you are a softswitch maker, you should be headed towards India right now. The carriers in India are expected to deploy softswitch equipment worth half a billion US dollars over the next five years. Here is a link to the report that I concluded today.
Carriers in India have already drawn up plans for deployment of more than 14.9 million Class 4 NGN ports and over 4.1 million Class 5 NGN lines, making it one of the top Nextgen Voice markets over the coming few years. In majority of the cases, vendors have already been finalized. The findings came as a major surprise while I was researching the subject. My impression was that VoIP in India was limited to certain small PC-to-Phone offerings only. I started my research some two months back with the intention of finding out why the carriers in India were not serious about NGN voice. I was proven wrong. There are tenders floating all around in the country.
According to this report on Telephony Online, Microsoft has shelved its SDP. My initial thoughts are that this is due to the resistance to SDP approach by carriers that have been pursuing IMS and IMS-like blueprints. IMS has its own service delivery function but this mainly supports SIP. SDP does make service delivery easier for telcos by bridging in non-SIP apps as well, but I guess the question carriers might have been asking is: where are the developers?
Microsoft has an army of developers behind it. But most of the 6k developers that BT/Ribbit has, for instance, put together were secured directly and not via Microsoft. Microsoft itself had no more than 900 developers registered with the Sandbox. That I think is the second reason whey SDP has been shelved.
In view of the architectural rigidity exhibited by telcos, Microsoft has been experimenting with the likes of eBay and Amazon for generic mashups (not just communications related mashups) delivery through its SDP. Those people have their own problems right now. I don’t expect them to challenge telcos any time soon. That is reason number 3 SDP has been shelved, I think.
We have seen companies like Level 3 trying to offer wholesale PBX. It did not really take off. What are the niches you have identified for your wholesale PBX offering?
We mostly serve small cable operators. Platform providers like Broadsoft and Sylantro do not have the reach and resources to serve all the small providers like these. They are focussed on larger carriers. On the other hand the small providers themselves do not have the resources to manage complex Class 5 feature set.
Some of the managed service providers now bring in their own proprietary BSS solution? You also have a BSS solution I guess.
BSS is critical to the success of our offering. Our BSS allows us to do all the backend stuff like number management, number porting, etc cost effectively. On the front end the system integrates with the cable operator provisioning systems including modem management modules.
In the cable space who do you compete with in the managed services offering? Net2phone used to be active in this space sometime back.
We are seeing some attrition of Net2phone customers. In fact two of the largest customers we have were former Net2Phone customers. There are a couple of competitors like Momentum Telecom. And of course you see Sprint to a certain degree. We also see NGT every now and then. Sprint obviously pushes termination side mostly.
Are there any elements of Packetcable in the offering you have, or is this SIP based?
We can do Packetcable but none of our operator customers want Packetcable. Our offering is SIP based and it gives a lot of flexibility.
Enterprise FMC vendor, Agito, has hired Amit Chawla as new CEO. Amit was previously with Veraz. I am not sure if he co-founded ipVerse, one of the original constituent elements in Veraz, but he was certainly the person that orchestrated much of the transition from ipVerse to Nexverse to Veraz. When Veraz was formed, the other constituent element came over from ECI.
ECI contributed the media gateway part of the business (and the DCME stuff too). A lot of those arrangements were made by Amit. Anyway, unless Agito has hired Amit to engineer a favourable exit, the selection raises a couple of questions about Agito strategy. Amit comes from a carrier background and Agito sells to enterprises (sometimes through OEM partnerships with the likes of Avaya). Maybe Agito is looking to sell to carriers after all. FMC vendors have had a tough time deciding who to sell to. Do they sell direct to enterprises (DiVitas) or sell via PBX partners (Firsthand/Counterpath) to enterprises, or sell via carriers to enterprises (OnRelay), or try to sell to both enterprises and carriers (Tango Networks).
Of course there are some more combinations possible there. Having said that, I believe in FMC somewhat. Perhaps we are in the early stages of the FMC and something more simpler and robust needs to evolve to bridge the future fiber and 4g networks seamlessly.
All the VoIP and IM apps developed for social networks have been freebies so far. I think Nimbuzz is the first one that will be making money out of such an arrangement. The mobile VoIP startup had teamed with StudiVZ, Germany’s largest social network with over 12 million members. StudiVZ will exclusively market Nimbuzz. Not just that. Nimbuzz is the first and the only third party application that StudiVZ has integrated.
Nimbuzz should be able to drive increased user interaction and activity on the StudiVZ site thus driving more advertising revenues. Those revenues will not be shared with Nimbuzz just yet and the sharing arrangement applies from 2009 onwards when Nimbuzz stands to get 50% of the advertising revenues resulting from the usage of its application.
Breaking the silence of 40 long days today, a silence protesting the ugly 8 long years of telcom recession. That is one reason. The other reason is that I have been working on some custom research projects while you guys were watching the markets throw those familiar tantrums. Anyway, the blogging will be light for the next few months, but we will have a full-time blogger for you sometime in December. I have not recruited for over a year now. In India, that means attenuation to the tune of 50%. Almost an organized attrition. And that is a killer especially if you have to do the baby sitting while they stay!!!
Some comments on the major developments over the past month or so:
US elections: Easier to say things when you are out of power. The moment you assume it, you get a microscopic view of things you were previously seeing with your naked eye. Anyway, as long as the US produces innovation, there should not be a major problem. In the context of communications industry, the problem is to monetize that innovation. In ICT, some new and equally competitive pockets of innovation are opening up - areas like Eastern Europe and Ireland. I would probe the BRIC countries next.
The financial crisis: not quite Correction 2.0. However, I wish the VC model would implement some longer term correction. There is at times excessive optimism among the young entrepreneurs about their business plans (in the ICT industry). The VC model of nineties has also cultivated the culture of ‘let us work ridiculously hard for 18 months and then market ourselves as an acquisition target’. That sort of culture just defeats sustainable entrepreneurship. I wish that could go away. Other than that I think the financial crisis of Oct 08 could precipitate something that has been cooking over the last few years. A couple of telecom giants will fall. Their places will be taken by Huawei and ZTE. Open Source could gain importance in communications. I wish the landline could make some sort of a comeback with the fiber deployments. I can’t stand Internet over wireless, especially time sensitive apps. The crisis has also been conveniently used by some companies to do layoffs.
There is currently no forum out there that promotes partnerships among the three main parties in Voice 2.0: the operators, developers, and platform providers. While events like eComm have focussed on direct-to-consumer offerings, other event organisers have paid little attention toward this new ‘evolved voice communication’.
My personal belief is that Voice 2.0 startups alone will not re-invent voice communication. Viral marketing alone will not suffice, no matter how innovative the services. For an efficient distribution of new services and adequate monetization of those services, the industry requires partnerships among all three parties (mentioned above).
Forging those partnerships is the main purpose of our Voice 2.0 Summit. We welcome all progressive telecom professionals to the event. Details can be found here.
Our mission at Jaxtr is to enable new conversations by letting people link their phones to the web.
So the value you bring in is the integration of PSTN and web. What happens when people start to bypass PSTN more and more using broadband? You would then have to re-invent your business dramatically.
That scenario is way off. The cell phone companies are not going to go out of business anytime soon. But as the voice communication moves to IP more and more we will terminate sessions on IP device. That does not matter much to us. It is in fact cheaper for us.
But when most voice communication is IP to IP, it is likely to be Skypes and Gtalks and MSN Messengers of the world that would be on the minds of consumers. It will be difficult situation to compete in.
Not necessarily. Jaxtr offers a unified way of communications whether it is SMS or voicemail or real-time voice. Consumers prefer having one consistent communications identity.
Is that one consistent identity going to be the mirror number that you offer whereby users are able to keep their numbers private?
The consistent identity is the Jaxtr link. Numbers that get generated depend on the device you are calling from. So it is not the number that stays consistent. It is the Jaxtr link that stays the same. A mirror number is not so flexible when you are travelling. A link is much more flexible.
ENUM look up system is also part of the IMS blueprint. Are the IMS projects implementing ENUM in practice? Where are ENUM servers being deployed today?
Some of the IMS projects are becoming smaller due the present economic conditions. Public ENUM has gone by the wayside. But there are a host of new applications for ENUM server products. There can be wholesale applications like our deployment at Telus where our product is being used for arbitrage route plans. Having said that, we have a number of IMS wins. France Telecom is one of our larger ENUM deployments where they are using our solution along with NSN’s IMS solution. We see a lot of inside carrier federation projects where you might have multiple or heterogeneous softswitch environment. ENUM servers there are being used as aggregation point for route plan data. That data is then made available to softswitches and SBCs. So you provide one provisioning and aggregation point for route plan data and push that out to networking elements. When you do those type of projects you naturally look into LNP kind of possibilities. So the scope is vast in theory as well as in practice.
Could there be some overlap in functionalities i.e. an SBC or a softswitch subsuming some of the ENUM routing functionalities?
We could have some overlap in deciding where in the network the call routing decision is made, for instance. Other than that we don’t expect much overlap as different network elements evolve and scale as per particular functions. IMS could live or die. But one of the things it is going to teach us is how to architect new generation networks. IMS defines the specific roles of various different network elements and I think that is the trend we are likely to see.
I guess the prospects for your solutions depend upon the extent to which a carrier in the future wants to keep the control at the core of the network. When you have more and more intelligence migrating towards the edge such as a peer-to-peer scenario like Skype, what kind of role would your product be playing?
Your instincts are spot on. When I talk to carriers I tell them this is the platform that delivers control. What carriers realize is that the external services, such as the peering fabrics, are essentially there to provide layers of reliability, disaster recovery, backup and storage – things of that nature. There is a paradigm shift going on inside carriers. Only when the burden of managing the data that they want to manage internally becomes huge, only at that point will they go out to an outside service provider. At the same time, with centralized infrastructure, there are certain costs that can be reduced. For instance you can centralize LNP and centralize provisioning of multiple softswitches.
So … Sonus has for a long time bet on hosted VoBB i.e. its vision of Class 5 VoIP has been that of an over-the-top voice over a converged broadband pipe. That was until the BT deal came along. BT vision says broadband and narrowband voice will co-exist for a long time. In terms of architecture that translates into a bridging technology, the one that bridges narrowband and broadband. As such you will find the Access Gateway + DSLAM kind of technology deployed in BT exchanges. If the customer is an over the top VoBB user, the voice hits the DSLAM direct. If the user is a narrowband customer, the voice is packetized by the Access Gateway in the same box before it moves over to DSLAM for IP dumping. The combined Access Gateway + DSLAM box technology is what they call MSAN.
So why am I pretending to be technical? Just to tell you that Sonus now has the hots for these MSAN type deployments. BT might not be alone in implementing this architecture. I think even France Telecom - that has over one-third of its landline users now subscribing to over-the-top VoIP - has realized that not all its customers will move away from POTS. And the regulation will never allow FT to ignore those that keep using POTS. Conclusion: use the BT model.
And therefore ….. what good is Sycamore as a merging partner for Sonus? Well, Sonus has the Access Gateway product and the Access Server. Sycamore brings in the access equipment for data. That means Sonus gets to keep all the money for the access POP that handles both data and voice access. Going forward, this access POP might be the only POP a telco deploys as Class 4 and Class 5 functions get integrated into a single server. So the merger makes sense. Besides, Sonus has been too conservative in terms of acquisitions. It needs to diversify in order to move ahead.
For T-Mobile, it is ok to use WiFi in the US to offload voice traffic (sometimes without informing the customers), but in the UK it refuses to interconnect with Truphone and in Germany it altogether seeks a ban on VoIP over iPhone, whether the call is made over 3g broadband or WiFi. That sounds like more than double standards.
The German court has unfortunately sided with T-Mobile this time. In the UK, it did not have much luck though.
Mobile VoIP is actually not allowed fully in the US. You cannot use 3g mobile broadband connection to make a VoIP call. The wireless carriers are allowed to block such use. Only VoIP-over-WiFi is allowed. Indeed Apple has also sided with the wireless carriers in the US. It is allowing VoIP on its iPhone handset only as long as the call goes over WiFi.
7.7 million VoIP access lines shipped during the quarter, down from 7.9 million lines in 1Q08. This represents second sequentially quarterly decline from 4Q07. Of the 7.7 million lines shipped during the quarter, estimated 6.8 million went towards residential VoBB. The remaining were deployed as IP Centrex lines.
6.8 million Class 4 VoIP softswitch licenses shipped during the quarter, down from 7.1 million in 1Q08.
8.9 million service provider media gateways ports shipped during 2Q08, up significantly from 8.1 million ports in the previous quarter
5.2 million SBC sessions capacity shipped in 2Q08, down from 8.9 million in 1Q08. Revenue, however, did now show such a depression Q-o-Q. Revenue decreased 1.3% sequentially.
Marginal improvement in IP media service business Q-o-Q
385.4 billion VoIP minutes handled by carriers during 2Q08. Breakdown: 23.3 billion ILD, 270.2 billion NLD, 91.9 billion Local.
(Source: iLocus 2Q08 quarterly reports)
Please note that iLocus does not track IP upgrades to TDM ports. We only track pure VoIP deployments which in the context of Class 5 NGN includes VoIP hosted telephony implementations (such as hosted PBX and VoBB), new Greenfield VoIP deployments, complete replacement of legacy switches with VoIP, and extension of existing legacy networks with VoIP equipment in new geographies.
When I first looked at VoIP I thought it was a great technology to reduce your ISD bill. What was the first thought that crossed your mind when you first stumbled upon VoIP technology?
Back in 1996, there were a handful of VoIP products in the market. It wasn’t really an industry. So you could not envision all the other pieces in the puzzle. But you got the sense that it was a new way of communications. By 1997 we had decided to start a magazine in this space called Internet Telephony that has been dedicated to VoIP only. So we thought this was going to be big.
Did you work with the first generation VoIP vendors?
Absolutely. We worked with companies like Vocaltec, Dialogic, Natural Microsystems and others. Strangely enough, Microrsoft also entered the market early with their Netmeeting. You would not expect them to enter the market that early. That further cemented the feeling that VoIP was going to change communications.
And then nothing happened in the area of Netmeeting for a long long time. In fact nothing in the area of VoIP from Microsoft up until the recent introduction of OCS.
The people that I spoke within Microsoft tell me that they shifted a lot of people from VoIP projects over to Internet services side. And they have become a leader on that front. Some three years back they realised that there was an opportunity in communications on the enterprise side. So they re-invested in the communications.
VoIP has been reasonably disruptive. But VoIP architecture itself has evolved quite a bit over the years: from monolithic boxes to softswitches to IMS and further onward to Voice 2.0. What do you envisage to be the impact of VoIP from here on? How can something evolve and yet be disruptive at the same time?
Vivox is planning to open up its platform to third party developers. Vivox provides a VoIP platform for developers of online games and virtual worlds. The first round of apps that the company expects from its developer partners include wide ranging features from lip sync to natural language speech recognition and response. There is also interest among developers in porting the code to different mobile platforms to enable users to connect to their communities from multiple devices.
Vivox provides voice services to a number of games and virtual worlds including CCP Games (Eve Online), Linden Lab, Sony Online Entertainment (Everquest, Everquest II & Star wars Galaxies), Wizards of the Coast, Multiverse, NCsoft and K2 Network. As the network provider to these companies, Vivox serves over 5 million users. Currently the company offers premium services that are often sold directly to end users and include things like voice fonts which give players the ability to morph their voice to match their avatar, i.e. an elf.
• 800k IPTV subscribers as of end 2Q08. Added 100k subscribers during 2Q08.
• hanaTV IPTV revenue was KRW 82 billion in 2Q08.
• Hanarotelecom has achieved a market share of 9% through strengthened VoIP.
• Wireline operators launched TV Portal services to prepare for the commercialization of IPTV.
• 508k IPTV subscribers. Net additions during the quarter: 73k. Number of subscribers increased 29% q-o-q
• “2.3% Local revenue decrease and the 5.6% Domestic Long Distance revenue decrease, mainly due to mobile and VoIP substitution”.
• 4.3m broadband subscribers (including ADSL and FTTB) at the end of the 2Q08. 3.5m ADSL subscribers, 782.4 thousand FTTX subscribers.
Flash phones were supposed to be hot this year. The market has instead seen only four players so far: Ribbit, TringMe, flashphone.ru, and this startup I came across yesterday - Indafon. I guess Ribbit’s successful exit might speed up the Flash side of VoIP. Anyway, Hungary based Indafon has had about 50,000 users trying its product in Hungary. The company is now targeting a global audience for the next stage of its trial.
As far as VoIP development within Flash environment is concerned, one of the biggest challenges is to maintain call quality when connecting a Flash player with PSTN networks. There are problems related to codecs, for instance. Flash is not open to some of the codecs. Adobe is beta testing Flash Player 10, the new Flash media server. This one is expected to support real-time voice and video transfer more adequately. Adobe is also sorting out the codec issue.
Talking about creative ideas, here is a third idea of the day: A hardware add-on that can be plugged into an IAD (through an Ethernet port) serving as a base station for a DECT phone. That is what I thought SiTel had announced recently. However SiTel is a chip company and I should have known better. The Netherlands based company announced a DECT chip that can be placed in existing IAD applications with telephone connections.
Modem manufacturers have been shipping IADs and cable modems with DECT base stations for quite some time now. Thomson has shipped more than 1 million DSL+DECT to BT for its residential VoIP offering. Starting later this year, BT will be sourcing its DECT enabled IAD from a Germany based vendor in addition. The Germany based vendor uses SiTel chip. SiTel expects about a million of their DECT VoIP chips to be integrated inside IADs by the end of this year. Majority of these IADs will land inside residential customers in Germany.
I am putting up one of my papers for free download. I had hoped to make it available using a wiki pluggin for Movable Type so that the readers would be able to edit it. I have been looking for such a pluggin but have not found it.
The reason such a paper is an ideal wiki material is because the selection of companies inside this paper is subjective. The selection and the justification for that selection represents purely my own personal opinion and judgement. I have tried to weed out the 10 most influential companies in VoIP over the last 13 years.
By the way, if you are a web developer, it could be a good idea to write a wiki pluggin for MT or Wordpress, which enables a blogger to select some of the blogs ... opening them up for editing by readers.
It must be hard trying to sell hosted VoIP to SMEs in Europe. Which verticals have you found traction in?
Certainly some of the verticals are adopting it much more quickly. Some of the traditional business segments such as legal and accountancy firms are slower due to almost the definition of the style of their business. We have found traction among IT and media companies in particular. Recruitment consultants and travel agencies are also keen on trying VoIP. The overall response has been better than what we had expected.
What has been your unique selling point in trying to win business away from the incumbent Eircom?
We have a fiber based network. For business customers we offer both fiber and DSL access. In the residential segment we have laid fiber to the home, and we have just started offering 50mbps service. That is winning a lot of business away from the incumbents that are trying to sell their traditional leased line and premise-based PBX services. We also have our own transatlantic fiber infrastructure to support our growth in IP services.
What is the plan for the residential market?
We are focusing on the housing development projects and have over 80% penetration there. We are planning to offer triple play services there including IPTV.
Which vendors are you working with for IPTV and VoIP?
The tremendously independent telecom policy maker of India, TRAI, has sent recommendations to DoT asking the admin guys to allow VoIP in its full form. It is likely therefore that you will finally be able to save some lawyer fees that you must have paid so far in order to understand the mind boggling complexities of VoIP regulation in India. For the market, though, I reckon the policy decision – if taken – will be at least 10 years late.
In the UK when International Simple Resale (ISR) hit the market, Ofcom put a price cap on BT. BT could not lower down its price by more than X% per year on its ISD rates. Let us do a PhD to understand why. Assuming you are regulating a market driven segment of an economy, you would ideally protect the consumer interest by strengthening the competition in the segment. That is why resellers were free to price their ISD services in the UK while BT pricing was regulated. In India, however, DoT let its former government department (BSNL, which is essentially DoT) to drop prices before the competitors could come into the market. That took the life out of several competitive measures including network migration towards VoIP.
Dutch Cable MSOs have advanced the capabilities of their VoIP peering federation by hybridizing new (ENUM based) routing with the traditional (IN/C7 based) call management. The new capability, developed jointly by NSN and XConnect, essentially keeps more voice traffic within the IP cloud thus (1) eliminating PSTN interconnection fees, and (2) increasing chances of ‘HD voice quality’.
Conversions from IP to TDM/PSTN can result in call quality deterioration. If the call remains on the IP cloud end-to-end, and provided there is adequate bandwidth channel, the quality can be dramatically improved by using wideband codecs.
NSN is the primary equipment vendor to the Dutch MSOs. However their equipment has not been able to simultaneously support ENUM and IN/C7. The capability has been brought in by XConnect, which is now officially in the latest software version of NSN platform. This enables the Dutch Cable peering project to move to the next phases towards production, and also potentially allows ENUM registry based routing to spread into other NSN based networks.
Developers have never been so hot in telecoms. They must be spoilt for choice given the numerous programs out there enticing them with high level voice APIs. On the other hand these high level APIs across multiple vendor platforms could give developers an opportunity to easily scale across those platforms. However I do think that too many companies are getting on the bandwagon. Particularly when the monetization of the apps developed is still an issue.
This particular announcement from Sangoma, a VoIP hardware company, is indicative of how vendors that have nothing much to do with developer programs, want to drive some marketing from this industry trend. Sangoma is trying to set up a developer network in order to ….. as I understand it ….. drive some qualified web traffic to its site.
Enterprise media gateway business is dominated by Cisco. My understanding is that it is a tough business to be in unless your shipments scale significantly. Would that be a fair assessment?
Scale is important and we are looking for channels to scale our business. But I think there are a number of ways you can compete. When we looked at this product category we decided to go after the SME market. Cisco has products that serve this market segment but this is where they tend to be more vulnerable. Most of the competing media gateway vendors have not done a good job targeting the SME segment. So that is where we see an opportunity.
On the other hand, the fact that there is a big player in the market validates the potential of the segment.
Yes, Cisco’s presence validates the market potential. That is the good news. But the bad news also is that they in there. They are a 900 pound gorilla that can dominate the market. We have had to put in a lot of attention into how we compete.
What is your main focus of your product differentiation?
We believe that SMEs are really looking at UC and that is an important element for us in terms of differentiation. Our gateway is engineered to efficiently handle the UC environment providing features like routing, security, QoS, and data integration.
Why do you think Cisco has not had a competitor of their strength in enterprise media gateways? They surely do not come from voice background. I think Nortel or Alcatel-Lucent might have been able to do a better job given their voice expertise. Why is Cisco dominating the market?
Arcep, the French communications regulator, has published its 1Q08 report on French telecoms market. The document reveals that majority of the VoIP subscribers in the country use VoIP as the primary line. In fact a fifth of all French fixed line connections at the end of 1Q08 were VoIP-only.
Some more figures from the report: 30% of the 40 million fixed lines in the country are VoIP lines. These 30% lines account for over 40% of the fixed line minutes. Sound like a few major indicators of confidence that consumers are showing in VoIP. It is safe to conclude that at least in France the migration to VoIP is now irreversible.
Over $50 million worth telecom equipment that belonged to Starvox, will be sold via online auction next week on the 12th of this month. The equipment includes over 40,000 Veraz gateway ports and around 5,000 Cisco VoIP gateway ports. Cisco VoIP sales had an impact due to the second hand market way back in 2000 and 2001.
40,000 ports sounds like a lot for someone like Veraz. Looking at the average port shipment numbers of Veraz, this could represent around 10% of their quarterly VoIP sales. So the vendor could also see an impact on its revenues over the next couple of quarters.
• Comcast ended 2Q08 with a total of 5.6 million VoIP subscribers.
• 5.6 million VoIP customer base represents 12.5% penetration of the total Comcast addressable market.
• VoIP ARPU: $39.48.
• Revenue from VoIP service increased 50% from $425 million to $640 million in the 2Q08.
• Circuit-switched phone revenue declined 96% in the 2Q08, as Comcast exits that product offering. As of 2Q08, Comcast served 10,000 circuit-switched customers.
It has been almost two years since the Acme Packet IPO. VoIP industry could do with another little boost. I have my eyes set on Asterisk/Digum next. I think there could either be an acquisition of the company or an IPO within the next 12 months. The usage of Open Source VoIP has hit that critical level where it becomes a force to reckon with.
Digium serves as the corporate identity of the Asterisk project, an Open Source VoIP platform which on average is downloaded 3,000 times per day. The company makes money from (1) selling hardware that lets Asterisk platform interwork with PSTN, (2) licensed version of Asterisk IP PBX system, and (3) IP PBX license revenue selling the recently acquired Switchvox product.
Switchvox product is one of several licensed versions of Asterisk in the market. Digium has in fact also acquired a company that organised Astricon trade show that brings under a thousand developers together from various countries. Astricon owners also ran an Asterisk training program.
So those are the sources of revenue. The company claims to have had 26th consecutive profitable growth quarter. It has shipped over 4 million gateway ports so far. Approximate geographic split of ports/revenues is as follows: 52% North America, 25% EMEA, 23% APAC and CALA. Europe is ramping up fast. Nearly 45% of all Asterisk downloads these days come from Europe.
I guess the first major success GIPS had was Skype. You have been behind Skype which became a big phenomenon. You also have new startups like Nimbuzz as your customers. Who in your opinion has the right strategy in place to emerge as the next major player in consumer VoIP?
I think the competition now is not about emergence of another Skype. VoIP is being added as a capability to almost all the major messaging softclients out there. Most of them are GIPS customers. You also have VoIP integrated into the mobile handsets. If Skype is to see competition, it is likely to come from mobile VoIP companies.
How far has GIPS been able to leverage the success it had with Skype?
Skype was certainly the customer that put GIPS on the roadmap. That opened the doors for many of the deals that we did subsequently. Fortunately the reverse is not true. When we lost Skype as a customer we did not lose other deals. Over the last few quarters we have had some great wins such as Baidu.
I think media gateway wins would have been a bigger source of revenue for GIPS. Why is GIPS not targeting media gateway vendors? Is it because they tend to have their own voice processing solutions?
You have probably not come across our customer list. Out customers include companies like Cisco, Nortel and Avaya. You should assume that we will in the future move from pure P2P client implementations to gateway deployments. That is a logical conclusion to have and you can assume that we are going in that direction.
This is Vonage’s second attempt at mobile VoIP. We saw the first iteration some three years back when Vonage teamed up with UTStarcom to offer single mode Voice-over-WiFi. UTStarcom supplied the handset.
Having a handset dedicated for just mobile VoIP might have turned out to be a bad idea in retrospect. You would have restricted yourself to just those customers who would be willing to buy a UTStarcom handset. UTStarcom recently restructured the ownership of its handset business.
• Added 42,210 new IPTV subscribers during 2Q08. At the end of June 2008, the total Belgacom IPTV subscriber base was 391,460
• IPTV ARPU: EUR 17.6 compared to EUR 15.7 a year ago
• Added 8k new VoIP subscribers during 2Q08. At the end of June 2008, the total VoIP subscriber base 24k.
• During the quarter, Belgacom started offering Video-on-Demand and HD channels
• Belgacom customers that want only TV service running over their broadband connection can also avail of the service. Customers are not required to subscribe to Internet services.
Over the last two years you seem to have shifted focus from PC centric communications to mobile centric communications. What opportunities do you see on the mobile side?
Many of our customers are mobile operators as well. So unifying the user experience across desktop and mobile is a big opportunity for us. Apart from the service providers we also have our OEM partners like Nortel collaborating with us across our desktop and mobile product offerings.
You have invested in FMC and handover technologies through the Firsthand and Bridgeport acquisitions. What are the prospects of handover technologies once wireless broadband becomes ubiquitous?
The FMC and handover technologies turn the Internet into a large roaming network. That is very valuable for a service provider. It also enables a service provider to offload a lot of voice traffic over to IP. So from a service provider standpoint there are several benefits at this stage. This kind of technology also facilitates a lot of new wireless data services.
How far have the Firsthand and Bridgeport acquisitions been assimilated into your other product lines? And are there going to be further acquisitions?
We have had two new deployments that involve 3 of the 4 solutions acquired by us. So the integration of the acquired solutions has been satisfactory. There are more things we can enable through a more tighter integration: things like mobile centrex and widgets for various features. We are working on those bits. We were partners with these companies before we acquired them. We validated the prospects of joint products with our existing customers prior to the acquisitions.
On the acquisitions front, making three acquisitions in nine months is a huge task. We grew from 50 people to 150 within a year. So we have to be careful. But if they make sense and they have interesting solutions around mobile, identity, presence, and they add value to our existing customer base and strategy, we are prepared to consider. But it is not going to be our big focus for the next two quarters.
There is one common source of revenue for mobile VoIP/callback companies: cheap long distance calling. Some of them arrange the termination into the public telephone networks themselves. Some do it via Skype. While mobile telephony arbitrage is rather new (except for the calling card market), the barriers to entry into mobile VoIP/callback are not exactly tough. You pay a few thousand dollars and you can be up and running in a matter of days.
In markets with low entry barriers, you have to bet big money on the venture. Mobile VoIP companies do not have that kind of money. I have been exploring the revenue models of these companies for some time. Outside the arbitrage business, you find models such as Nimbuzz that are based on advertising. Nimbuzz will be dealing with text based ads. There are also certain audio based ad models in the market. There is some whitelabelling going on as well whereby a mobile VoIP company enables a cell operator to offer the service.
Natural Convergence has announced the latest version of its VoIP platform. It has been almost a year since the vendor made any product related announcement. Natural Convergence Inc (NCI) develops hosted VoIP platform meant for providers serving smaller businesses. Just to give you an idea of the size of the company's deployments, its customers support in excess of 49,000 small business users.
The licenses sold is doubling every 8 months, according to the vendor. NCI has customers in North and South America and the EU.
It has been over 13 years since Vocaltec pioneered VoIP. And for most of those 13 years it has been unable to keep its head above the water. The company has just announced the departure of its existing CEO. Pioneers ought to be preserved like the British Royalty. I think the best way for Vocaltec to preserve itself is to follow the open source market trend. Few random thoughts below …
- Vocaltec product is the earliest VoIP product around. Vocaltec was first with several things: first PC-to-PC calling software, first PC-to-phone client, first media gateway, first gatekeeper (h.323). The company has plenty of VoIP experience. Developers will benefit a great deal from this platform.
- Most new generation VoIP companies offer services off the Web in a hosted fashion. Vocaltec product is supposed to be very good in terms of managing voice over public internet.
Added 29 million users during 2Q08. Total subscriber base now stands at 338.2 million registered users worldwide. This includes subscribers sign ups through partners such as iSkoot, Fring etc.
$136 million in revenue for the quarter, representing 51% year-over-year growth.
Skype-to-Skype (PC-to-PC) minutes are estimated 14.8 billion and representing 38% year-over-year growth.
Skype Out Minutes (PC-to-Phone) are estimated 1.9 billion and representing 42% year-over-year growth.
Skype launched 4.0 in Beta, the largest redesign of the Skype interface since the company’s inception.
My impression is that your revenue model is currently based on advertising and whitelabeling? How soon do you expect those sources to generate meaningful revenues for you?
We don’t do whitelabelling exactly. Although apart from consumer solutions we also have industry solutions, those industry solutions are social networks for mobile operators and device manufacturers. These are outsourced IM and VoIP platforms that we operate for a mobile operator. We also bring mobility to web based social networks.
Is that not whitelabelling?
No. In whitelabelling you would typically charge a certain license fee per subscriber. We do not charge the operator anything. Our solution is used for free. What we do is share the ad revenues.
How many social networks and operators have you partnered with? What is the usage among the operators right now?
We have signed up 10 social networks and 3 mobile operators. One of the operators is using the full Nimbuzz solution. The other two, for now, are promoting our web client to their mobile users. With regard to social networks we will go live with our joint offering with a social network in Germany in August.
You will also have PSTN termination business in place. Which part of your business do you expect to be bigger: ad revenue sharing or the PSTN termination?
DiVitas, the FMC vendor, has quietly raised another round of funding. I hear through sources that the company raised $12 million last month. The round was led by SVB Capital. Total funding raised till date is $32 million.
Usually the early stage companies are keen to announce fundings in order to attract new talent. No announcement from DiVitas yet. Is this the sign, therefore, that the company has matured? Unlike most FMC startup vendors, DiVitas has been able to gain some traction in the enterprise segment. It mainly runs into Avaya and Cisco where the latter two are unable to meet dual-mode solution requirements adequately.
Few thoughts on the BT/Ribbit rumour. I see only one attraction for BT to acquire Ribbit: the 4,000 strong developer community that Ribbit claims to have cultivated. But then BT is supposed to have signed up over 10,000 developers for its Web21C program.
How far these developers will be able to extend their Ribbit app over to BT’s Web21C platform, I am not so sure about. Ribbit uses former Syndeo platform as the underlying feature server. BT has a different underlying feature server. And then there is all the OSS/BSS integration work that developer programs need to sort out prior to making their APIs public. For BT Web21C, Microsoft takes care of the OSS/BSS integration. Suffice it to say, the apps may not be easily transferable.
There are tonnes of VoIP gear in the enterprise networks now. If these enterprises start deploying SBC solutions, companies like Covergence could make it big. The company just announced new CEO, James Moran, who happens to have a strong background selling to enterprises (Moran was co-founder at edocs). Covergence has been supplying SBC solutions to both the enterprises as well as the service providers. However the vendor plans to increase its focus on the enterprise market.
The deployment mix Covergence has right now (service provider versus enterprise) is in the region of 80:20 revenue-wise. However, based on some of the large, global enterprises Covergence has struck deals with, it will probably be closer to 50:50 mix by the end of the year. Covergence website says 5 of the Fortune 25 use its SBC platform. One of the largest e-commerce companies also uses Covergence. My guess is amazon.com.
Covergence has about 150 customers, and about 30 of them are enterprises. The service provider customers are largely tier 2 and tier 3 companies that sell to business customers – generally hosted IP PBX.
VoIP lines were up 57.7k during the quarter to 549.9k. Subscriber numbers include pending installs.
The service is now available to over 90% of homes passed. VoIP line penetration stands at 28% of Basic customers who have the service available to them.
VoIP service margins are strengthening.
Announced limited-features, basic VoIP package, third product offering in order to appeal to a larger customer base.
Agito, the enterprise FMC vendor selling direct to enterprises, has raised $13 million in Series B. The startup has focused on verticals such as Higher Education, Manufacturing/Retail, Healthcare, and Energy/Utilities. Over 12 customers have implemented Agito solution so far.
PBX integration has been a major issue resulting in delays in enterprise FMC deployments. PBX makers take a long time in the certification process. Agito has so far achieved integration of its product with IPPBXs from Cisco, Avaya, Nortel, Asterisk, and Microsoft OCS. There is tighter integration with Cisco products. To date the company has not announced any OEM relationship.
While it is all wonderful to sign up a million subscribers and get a second round of funding, I thought it might be a better idea to explore with Nimbuzz what their revenue model really is. I get the impression that for a while, it is based on advertising revenues. And I say that is probably wise!
Starting September this year, Nimbuzz expects to generate meaningful advertising revenues through its revenue sharing arrangement with a couple of large social networks. One of them happens to be the largest social network in Germany (not hard to guess which one). Users of these social networks will be able to communicate across their communities and various IM clouds using Nimbuzz’s web based client. Abril, a media company, will be supplying most of the ads.
Mirror numbers are a nice way to circumvent high calling rates. But they are best applied to apps like Jangl etc (provided you have luck monetizing the service). So when Jaxtr announced ‘offnet’ calling to any phone number for low price, I was a bit disappointed. After all there are over a million companies out there offering convenient low cost long distance calling service.
A 2.0 company such as Jaxtr leads you to believe for a while that we are communicating in (additional) new ways that is something out of this world and then offers simple low cost telephony. I bet you a thousand dollars that Jaxtr did not have plain old telephony in the original business plan. Jaxtr’s offnet calling is somewhat analogous to SkypeOut type services. And yet if you think about what is wrong with Skype, it is the dependence on those termination revenues.
If Skype had sought an alternative direction for monetization, we could have had a true Voice 2.0 company today. Termination business has made Skype complacent and at the same time threatened its prospects due to thin margins. There is something not so graceful about a Voice 2.0 company offering simple termination. This is emerging merging with the ordinary.
In a recent short report on Voice 2.0 we took an early look at how various vendors and service providers are approaching the delivery of these communications aware mashups. I think it is fair to comment that the delivery models are yet to be established. These short interactions with Sylantro and MetaSwitch present a couple of contrasting views and should be viewed along with posts like the ones on Broadsoft and Ribbit.
Attendance was low for sure. A lot of people walking along the central isle …. But exhibitors along the ends of the hall saw very little traffic.
Lots of Analysts and Press. Announcements were not so exciting though.
Exhibiting companies: mostly infrastructure guys. No startups, no emerging companies ….. and no debates around Telco 2.0 on the floor.
The overall theme appeared to be access again. Not much talk about apps. Some video and IPTV stuff was being talked about. Wimax and LTE ... not much around that topic either.
Broadsoft will be announcing availability of Web 2.0 APIs next week. The company had announced an open API developer program a couple of months back. The REST interfaces promised to the developers will be released along with the announcement next week. So far Broadsoft has managed to enlist over 300 developers in the program. With the developer friendly Web 2.0 APIs, the company expects that number to increase substantially. With such APIs in place, the ‘Sandbox’ will start working off the web where developers can test their applications.
The timing is certainly right for Sonus to position itself in the VoIP over WiMax market, which is starting to ramp up in parts of Eastern Europe, Africa, and Asia. There is nothing new in terms of the capabilities that Sonus has to support for VoIP over WiMax. The existing product line will do. However there is a much wider choice or codecs that Sonus offers for VoIP over WiMax implementations. A wider choice of codecs gives a Wimax service provider flexibility in terms of the bandwidth it provisions for VoIP.
Once mobile Wimax kicks in, we can expect vendors like Sonus to tweak the platform a bit more in order to be suitable for VoIP over Mobile Wimax.
Robby Benedyk, Senior Manager Global Marketing, Tekelec co-authored PacketCable ENUM Server Provisioning Specification that deals with the way ENUM is handled in Cable networks. Tekelec has announced a commercial ENUM solution, which Robby discusses in this video. Cable MSOs are very active in the area of VoIP peering.
The specs have also been submitted to the IETF and as a result could be extended over to the GSMA IPX project initiative. Once that happens, Tekelec will substantially increase its target market for the newly announced ENUM solution.
GSMA IPX initiative could change several things for cell operators. To start with, it can lower down interconnection costs substantially for cell operators. There are several value added services that could be enabled across multiple cell operator networks. Dawn Hogh, VP Marketing, Veraz Networks spoke to me today about her company’s support for the GSMA IPX spec.
The spec support involved enabling features such as certain type of transcoding and end-to-end QoS. Veraz has been pushing its peering solution for the last several months now. With the GSMA IPX support, it is now able to sell the peering solution to cell operators as well. But it will be the VoIP wholesale carriers that Veraz will be targeting first. VoIP wholesalers like iBasis, Telecom Italia and BICS are looking to cash in on the new opportunity (the multilateral GSMA IPX peering contract was awarded to NeuStar some time back). Veraz already has three customers for this GSMA compliant peering solution. BICS is one of them. Veraz will be making the BICS announcement on Monday.
Veraz has introduced a smaller version of its softswitch. iMN is a mini ControlSwitch (Veraz’s flagship product). I thought this was the long awaited Class 5 application server that Veraz has been working on. It turns out that iMN is a small version of ControlSwitch, supporting both Class 5 and Class 4 applications in a single platform.
It is pre-packaged i.e. many of the parameters on the ControlSwitch which can be customized are pre-configured to make it easy to install and turn-up quickly for smaller applications. The mini platform supports 15,000 subscribers. Carriers that deploy the iMN and later decide to scale beyond 15,000 subscribers would then migrate to the ControlSwitch.
Alcatel-Lucent will be demonstrating its Voice 2.0 capabilities at NXTComm next week in Las Vegas. Alcatel-Lucent is currently engaged in setting up a developer program. The purpose is to provide a Service Creation Environment based on its IMS application server allowing development of Voice 2.0 applications via its existing Web 2.0 APIs.
The developer program is intended both for extending legacy IN features on to the Web as well as to introduce new web-enabled applications. Alcatel-Lucent’s IMS application server has IN interfaces that will be exposed through Web 2.0 APIs, while existing products such as the Contact List Server, Presence Server, etc are already offering Web 2.0 interfaces.
Peter Blackmore is actually going to take charge as CEO of UTStarcom from July 1st. The present CEO - Hong Lu, an equally charming guy - updated us yesterday on NGN voice and the service providers they are trying to win over. Hong revealed that one of its softswitches in China (just outside Shenzen) is serving around 3 million subscribers. That is perhaps the biggest single softswitch POP I have heard about.
Anyway, back to the interview with Peter. I caught up with him right after the company’s analyst event here in New York today. We talked about (1) UTStarcom’s ability to cross sell its complementary triple play product lines to customers, (2) company’s strong focus on IPTV, (3) reasons for offloading the IP CDMA business, (4) strategy to address opportunities in Eastern Europe, and (5) strategy to address the issue of lower gross margins in the handset business.
Apart from the video content above, I would add the following notes from his presentation that we sat through earlier during the event:
It has been over two years since some of the first peering federations were announced. Following the Netherlands setup facilitated by XConnect there was a lot of enthusiasm about the CableLabs’ peering RFP. However, two years into trials, there has been no official announcement from CableLabs. An equally bigger peering RFP put out much later by GSM Association was finalized during the early part of 2008.
What I had heard some six months ago was that the CableLabs peering contract was awarded to a smaller ENUM server company (my guess is NetNumber since they are the most widely selected routing database vendor in the cable segment). However a bigger player in form of NeuStar was brought in for some handholding.
Our estimates suggest that 5.3 million IP PBX lines were sold in 1Q08. These include those IP PBX lines that were shipped as part of hybrid TDM/IP systems. We have previously reported line estimates from pure IP PBX systems only.
Geographical split shows that EMEA leads. The 5.3 million IP PBX lines sold during 1Q08 generated $875 million in revenues worldwide. Avaya leads the market with 25.1% market share worldwide. That lead is followed by Siemens with 23% and Cisco at number three with 17.7% market share worldwide. (Market shares by line shipments).
This announcement from ECI caught my eye yesterday. It has been a long time since I last heard about MSAN+Voice. ECI has integrated VoIP access gateway module into its MSAN product. There are several MSANs with VoIP access gateway modules out in the market. ECI claims to be unique in that the vendor has integrated the module onto the network interface card, freeing up an additional slot for a subscriber line card that would otherwise be occupied.
Anyway, MSANs have a clear business case for operators that are unnecessarily running two parallel voice networks: narrowband voice and broadband voice. A VoIP enabled MSAN bridges the two disparate networks. A customer can either migrate to a VoIP line or choose to keep its PSTN line. Both will be served by the same network. Obviously there needs to be interoperability between MSAN and the softswitch being used by the operator. ECI claims to have proven the interoperability in the trials and deployments it is involved in.
I can see the value that this Echoes platform creates for a mobile carrier. However this is surely not for a wireline operator. If Microsoft’s Echoes is for real and the company is targeting wireline telcos as well, this could turn out to be yet another attempt to pitch applications to a telco without due consideration towards the lack of OSS/BSS integration. Wireline telcos are not slow adopters of fancy apps because they just happen to be conservative to the core. They have a dozen hooks to worry about.
And they prefer to have one humungous underlying platform that can be extended to accommodate all Voice 2.0 services. Echoes does not seem to be that extended underlying platform. It seems to be an extended IM platform, not an extended telco app server, which is what wireline telcos utilize and open up to offer new services.
DU blocking Skype looks a bit out of place because: 1, DU is a progressive new provider/competitor of Etisalat, and 2, DU itself benefits from the use of IP transportation of voice minutes. Not only does DU secure low termination rates from VoIP providers globally, it also facilitates exchange of VoIP traffic between some of the Middle East carriers – possibly earning a cut in the middle.
DU chooses not to pass on the benefits of lower ILD costs to end users because it has regulatory protection. The regulator in the country is interested in VoIP within the country and not the cross border VoIP. Apparently the regulatory features like emergency services and legal intercept are yet to be tested resulting is substantial delay in VoIP roll out.
FreeConnect is a service available to Australian businesses that works a bit like Freephone. Businesses receive calls from online visitors to their FreeConnect profile page. The idea is quite straightforward but the concept of a click-to-call portal seems to be the first of its kind.
Click-to-Call is yet to realize its potential. You can have plugins like Skype but the caller pays for the calls. There are click-to-call solutions available from VoIP providers that work pretty much like Freephone services i.e. calls are routed by the service provider to the numbers (either nominated by the business or assigned by the VoIP provider). FreeConnect effectively provides - all those businesses who sign up - the ability to turn their existing phone number to a Freephone number. It has gone for the portal approach rather eStara-like case-by-case approach whereby the vendor typically implements click-to-call capability over the client website.
Truphone has finally ventured into bridged mobile VoIP i.e. using GSM network for the last mile. In view of the limited WiFi coverage and problems related to VoIP-over-3G, it seems like a sensible thing to do. Mobile telephony arbitrage in form of mobile callback and bridged mobile VoIP is potentially a huge market to tap into and Truphone has to monetize its technology after all. I hope, however, that the story does not end here.
You could, like Jajah, try to generate meaningful revenues, but if you do not have the bells and whistles around your core offering, the industry interest and ultimately the investor interest just evaporates. Arbitrage is a bad long term business to be in. None of the resellers from 90s are in business anymore. Once the cell operators drop their artificially inflated rates for international calls, companies like Truphone could be in trouble. We are probably looking at around 5 years for cell operators to drop those prices. In the interim Truphone has to cook something appetizing. Something beyond cheap ILD calls.
There were estimated 69.2 million VoIP subscribers worldwide as of 1Q08. USA had the most subscribers (18.1m) followed by Japan (16.8m). France was at number three with 11.5m subscribers. Chart below gives VoIP subscriber numbers in top 10 countries.
SFR’s takeover of Neuf, the largest alternative provider in Europe, was approved by the French authorities last month. SFR has been offering SFR One, a converged IP Centrex offering to SMEs. Neuf has SME VoIP offering of its own. Neuf Pass is based on Open Source PBX software which the service provider integrates with a CPE in order to provide SIP trunking service.
SFR’s convergence play is based on Thomson solution. Neuf uses SER open source VoIP platform.
Post acquisition, SFR is likely to choose just one of the two service offerings. The choice will be a strong indication of where the operators see SME VoIP moving. The deciding factor is perhaps going to be convergence rather than the Open Source issues. With a hosted Centrex offering - that SFR has in place – it is easier to enable convergence across a fixed centrex phone and a mobile phone. It is easier to achieve a single centralized voicemail and other features with a centrex setup. With a premise based PBX solution such as the one Neuf has in place, it is difficult to centralize voicemail and other features. It is also a challenge to offer value added services-from-the-cloud.
Veraz media gateways and softswitch will to be deployed by Spice Telecom, cell operator serving two major states in India. Spice serves over 3 million subscribers and ranks among the top 6 cell operators of India.
Veraz has consistently been performing well in the wireless segment, which seems to be turning into the biggest source of revenues for softswitch and IMS vendors going forward. The chart below shows proportions of Veraz’s VoIP equipment that landed into wireless networks –across the last 4 quarters.
1&1
• Added 15kVoIP subscribers during 1Q08. At the end of March08, the total VoIP subscribers 1.91 million
• The number of VoIP minutes completed per month grew from 950 million to over 1 billion during 1Q08
• 1&1 uses open source platform to serve VoIP customers
Kabel BW
• Added 33,000 VoIP subscribers during 1Q08. Total VoIP subscribers at end 1Q08: 272,500.
• VoIP represents 12% of customer account penetration
Seems like VoIP had a soft quarter after all. In 1Q08, vendors shipped a total of about 7.9 million VoIP Subscriber Feature Server licenses for deployment in service provider networks generating $144 million in revenue. The number of lines is down by 19% Q-o-Q. The 3Q07 and 4Q07 quarters however were unusually high growth quarters for VoIP Subscriber Feature Servers. If the 4Q07 seasonality in particular is normalized, there is a nice sequential growth.
In 1Q08, Nokia Siemens Networks (NSN) led the VoIP subscriber lines equipment market on a worldwide basis with a market share of 19.8 percent. That lead is followed by Italtel at number 2 and Cisco at number 3 worldwide.
Nortel has announced an add-on module to bring in Web mashups to the telco world. The module works over standard hardware across the CS and MCS platforms. Nortel has been able to achieve integration with the usual first set of suspects: IM, social networking, TV etc. There are also some interesting combinations of pre-integrated mashups available. Delivery channels are obviously the telecom service providers.
Most other vendors that come in the Nortel league have been working on such a module and the associated developer programs. However none have publicly announced such solutions/program yet. Nortel is the first one to do so. Among the large vendors, NSN’s CBVoIP offers web services based SDK but the vendor has not announced the capability yet; Alcatel-Lucent is in the process of creating such capability on its IMS applications server; Huawei has a developer partnership program going under its IMS Ecosystem project. Again nothing has been announced yet; Ericsson does not have a wireline specific Voice 2.0 program in place yet.
• Added 43,931 new subscribers during 1Q08. At the end of March 2008, the total Belgacom IPTV subscriber base was 349,250
• IPTV ARPU: EUR 17.4 compared to EUR 13.4 a year ago
• Added 9k new VoIP subscribers during 1Q08. At the end of March 2008, the total VoIP subscribers 16k
• During the quarter Belgacom and Twentieth Century Fox reached an agreement to make Fox films available on demand via Belgacom TV, Belgacom’s IPTV service
Session Border Controller (SBC) revenues during 1Q08 were up 22% from the previous quarter 4Q07. A total of 9.9 million ‘sessions’ capacity were shipped during 1Q08 generating $56.8 million in revenues. 4Q7 revenues were in the range of $46.6 million for the segment.
Acme Packet leads 1Q08 SBC segment with market share of 47%, followed by NextPoint. Vendors who made significant inroads into the SBC market during 1Q08 include Sonus and Veraz.
I will be posting overall 1Q08 VoIP revenues tomorrow.
There was an aggregated total shipment of 335 thousand IP media server ports shipped by independent media server makers in 1Q08. Shipments represented a slight increase Q-o-Q, however revenues decreased compared to the previous quarter – perhaps an indication that Average Selling Prices are decreasing in this segment.
Estimated $14.7 million were generated by carrier IP media server products in 1Q08.
This is for the first time that a major telecom service provider has acquired a major social networking company. There are several possible synergies in this deal. I will pick on one of them. And that is the small business (virtual PBX) offering bundled with the relevant Voice 2.0 apps (communications aware mashups).
All major service providers are exploring Voice 2.0 strategies. Only BT has gone public with its Web21C developer program so far. While most service providers recognize the value of teaming up with their existing telecom vendors to bring in the developers, these switch vendors in turn do not have access to all the developers. They definitely do not speak and Web services language and do not typically provide Web 2.0 APIs except a couple forward looking VoIP platform vendors. In order to gain access to a vast developer pool, service providers have to tap into social networking, which is one of the main driving forces behind Voice 2.0.
VoIP momentum is shifting within Europe with Eastern Europe coming in strong now. The new set of initiatives is driven by the Wimax service providers. There are over 40 Wimax service providers in Eastern Europe if you count the Baltic states as well. Nearly half of them have either launched a VoIP offering or are involved in such trials. (There is a partial list at the end of this post).
The present VoIP traction in Eastern Europe is different than what we saw two years back. Two years back it was all about the network upgradation of the incumbent telcos there. This time it is a bit different. The alternative carriers are going for the broadband telephony type option. Wimax operators seem to be the driving force right now.
If Nimbuzz is planning to play with social networking and all the buzz words, it might be a bit late in the day. The market is looking for a monetization story, not a mobile VoIP app that hooks into every IM-client / social network / handset on earth. What is the value in that? There are plenty in the mobile VoIP market already signing up customers without much revenue to show. The only time they generate revenue is when a call is terminated outside the community cloud i.e. into landlines and mobile phones that are not Nimbuzzed.
An IM or a voice chat between Nimbuzz-to-Nimbuzz is not going to generate money. It will remain a challenge to monetize that aspect – I mean for ever! So something is wrong here: You create a community but make money only when you communicate outside that community!!! What incentive is there is expand that community then?!!!
For off-Nimbuzz calls, the company will make money on international calls mainly. Like EQO and iSkoot, Nimbuzz will also re-direct the calls to its local VoIP PoP from where it transports the call over public Internet backbone. So for the amount of time you are on an international call, you are effectively making and paying for a local call in addition. That is like the early PC-to-Phone days when you had dial-up connections. Outside the US, where local calls were metered, you paid for a local call (to your telco) plus the international call rate charged by Net2phone and others. That was not the main factor responsible for slow uptake of PC-to-Phone, but it was a turn off nonetheless.
• 1.09 million VoIP subscribers at the end of 1Q08
• Added 125,700 new subs during 1Q08, the most active VoIP quarter so far.
• VoIP revenues nearly doubled to $121 million compared to pro forma revenue of $63 million in the year-ago quarter.
• VoIP represents 11.4% penetration into the overall customer base.
• VoIP subscribers in Norway 133K. Added 2k new subs during 1Q08
• VoIP subscribers in Sweden 209K. Added 7k new subs during 1Q08
• VoIP subscribers in Denmark 107K. Added 8k new subs during 1Q08
• Telecom Italia ended 1Q08 with 136k IPTV subscribers, an increase of 56 thousand in the quarter.
• 1.5 million VoIP subscribers as of end 1Q08, which represents 23% penetration into retail broadband access lines.
• Added over 639k VoIP subscribers during the 1Q08.
• Comcast ended 1Q08 with a total of 5.1 million VoIP subscribers.
• 5.1 million VoIP customer base represents 12% penetration of the total Comcast addressable market
• Revenue from VoIP service more than doubled to $573 million in the 1Q08 compared to the same period of the prior year. Circuit-switched phone revenue declined $69 million to $14 million in the 1Q08.
• As of 1Q08, Comcast had 66,000 circuit-switched customers, and expects to wind down that business by mid-year 2008.
Jangl investors should try to sell the company assets to a telecom vendor. Some of telecom vendors are developing features that enable ad supported communications apps. Still early days but a Jangl platform would place them ahead of most competitors.
We are going to see two relevant trends going forward. One is the re-positioning of telecom service providers as some sort of media companies enabling various forms of entertainment, content, and social networking through PCs, TVs and mobiles. The other is the re-thinking among Voice 2.0 companies to productize their platforms.
• VoIP represented $160 million revenue during 1Q08 which is approximately 8% of company’s consolidated revenues for the three months ended March 31, 2008
Tata Communications will be deploying the high density gateways from Sonus. This is for the international footrpint, not for the domestic Indian network. The domestic network deployment would be a much bigger opportunity if it materializes for Sonus. Sonus has been involved in trails with Tata as well as Reliance in India.
The international footprint where Sonus gets its foot inside was inherited through the VSNL acquisition. Here comes the history:
- ITXC, one of the largest VoIP ILD wholesalers, deployed multiple platforms including Excel Switching (now part of Dialogic), Huawei, Vocaltec, Cisco, Clarent (now part of Verso gone bankrupt)
- ITXC standardizes around Cisco platform, the famous AS5300
- ITXC gets acquired by Teleglobe. VoIP traffic growth remains flat for a while.
- Teleglobe gets acquired by VSNL. By this time VSNL had already been acquired by Tata I think
- VSNL also used Cisco VoIP gear in its international footprint
- Teleglobe/VSNL Intl did not migrate to high density MGX platform of Cisco
- TODAY: They announced move to Sonus platform
Total revenue $27.9 million, a 20% decline from 4Q07
VoIP product revenue $20.7 million, compared to 4Q07 revenue of $20.9 million
1 customer accounted for 10% of 1Q08 revenues: MTN. Top 10 customers accounted for 52% of Q1 revenues. 6 customers each accounted for $1million or more in Q1 revenue.
4 new switching customers in 1Q08.
Revenue split: 16% North America, 84% International. Compared to 4Q07 split of 23% North America, International 77%.
Service revenue: $5.6 million. Gross margin: 53.6%.
It remains a mystery why Sonus is not keen on acquiring a feature server company such as Broadsoft or Sylantro. They should have made an acquisition of a Class 5 feature server three years ago when consumer VoIP got going. Netcentrex was another option available at the time.
Sonus has a residential feature set in their access server product but it has not done so well. Sonus has enabled consumer VoIP offerings of AT&T (Callvantage), Qwest, AOL, Carphone Warehouse and others. Only Carphone has scaled up somewhat.
The absence of an alternative vendor in consumer VoIP has meant that most business is ironically going to the legacy vendors. Sylantro and Broadsoft are small players in consumer VoIP. I reckon Sonus with one of these products would have done very well. Sonus placed its bets on over-the-top VoIP rather than other VoIP-to-the-edge options. It turned out to be right but did not translate into any fortune for Sonus.
A few open questions on the recent Jajah/Yahoo deal, perhaps our readers can answer:
Whatever happened to the Dialpad acquisition by Yahoo? Was that investment a waste then? Yahoo had acquired the company because apparently they needed an insight into the call routing science.
Jajah does not do PC-to-Phone application. Is Jajah going to develop that capability for Yahoo Messenger, or is it going to be a similar application to Jajah i.e. web initiated telephony that does not necessarily use the public Internet? Jajah voice quality is great right now because the company does not use public Internet.
It has been a while since I heard any major development on the consumer FMC side. I am assuming that mobile VoIP and Femtocell have pushed this thing back a few years. But there must surely be something in store for enterprise FMC during 2008. There is much more activity around enterprise managed FMC solutions. All major PBX makers have announced their products now.
The early adopters here have been healthcare, education, and warehouse operations. But there is interest coming from new segments such as the banks, we are told. With early implementations in particular there was need to have on-campus mobility as well as remote mobility so segments such as warehouse operations - where WiFi access can be very important - picked up some of the FMC first. Hospitality and healthcare also have that campus aspect where you can put dual mode device on campus on WiFi and same number can work on cellular.
ThinkPanmure thinks telcos worldwide will set up an old style cartel with shades of 19th century settlement rate system. It is supposed be a joint initiative to offer Skype type services. I am not sure if such an alliance is cooking, but if that is the case, the stuff being cooked is likely to remain half baked. Here is why:
• It will be an intergalactic challenge to bring two dozen telcos under one roof. Previous attempts to set up alliances - Concert, Global One etc – have failed.
• The proportion of IP-IP calls is miniscule. It is in the region of 6% of the overall voice traffic in France, the country with world’s highest VoIP penetration. On a global scale IP-IP calls proportion is less than 1%. What that means is even with a Skype type service you are still paying for terminations, the same old story as in the landline business.
As has been the underlying characteristic of VoIP industry over the last 7 years, the growth pattern refuses to change direction. Our 9th annual industry update published today reveals a yearly increase of 67% in VoIP access server licenses shipped worldwide, and a 35% increase in VoIP traffic.
There is no evidence yet of a long term slowdown in carrier spending. It seems that despite the risk of recession (especially in the US) carriers have no choice. They have to build out the IP infrastructure if they are to remain relevant.
The new investment cycle that started in 2004 to do with access part of carrier networks continues into 2008 and is likely to remain so through to 2009. That can be said with confidence in case of European countries at least. On a global level, even if the investment slows down due to financial issues, the carriers have to spend nonetheless. They have little choice, it seems. Due to the losses in landlines, wireline operators are expanding into mobile and mobile operators on the other hand are looking to leverage fixed broadband networks for voice offload or over-the-top services. The underlying technology that lets them both achieve those objectives tends to be VoIP.
The report covers various aspects of VoIP industry and is quite comprehensive as the previous annual reports. A lengthy post would not be possible. I have however listed a few bits from the Executive Summary below. A big thank you to those who ordered the report in advance. We appreciate your vote of confidence. There are also two other reports we announced today (Voice 2.0 and Mobile VoIP).
72% of Voice 2.0 developers prefer to work with Open Source telephony platforms like Asterisk, OpenSER, and FreeSWITCH and offer services direct to the consumer - according to our survey. Open Source platforms mentioned here are now considered carrier grade. For a standalone Voice 2.0 applications open source telephony platforms meet the developer criteria. Although working directly with telcos like BT (rather than going via vendors like Microsoft or Sylantro) is the second most favoured choice, it seems that Voice 2.0 developers overall prefer to take control of their development by utilizing open source platforms and then going direct to the end user.
Our survey of MVNOs reveals that nearly one-fourth of MVNOs are already offering/trialling mobile VoIP. By 2010 over two-thirds expect to have a mobile VoIP offering in place. The survey is part of our report ‘Mobile VoIP: 2008 Status Report’ published today.
According to the report, MVNOs could drive the first phase of mobile VoIP apart from the independent startups. The eligible providers are MVNOs, MNOs, ISPs, and Wimax operators. As far as Wimax operators are concerned they have to offer VoIP to differentiate themselves from other broadband providers. Mobile Wimax however is not expected till 2009. Wimax operators at present are looking to offer VoIP over laptops and fixed phones through ATAs.
MNOs can potentially use mobile VoIP feature to dump voice traffic transparently on IP. But they have cannibalization issues. ISPs on the other hand would see this as a completely new service since they have not dealt with mobile services before. So for now, MVNOs are expected to show more interest in mobile VoIP. And since MVNOs are likely to drive the first phase along with the independent startups, the take up is unlikely to be huge as per the report.
I noticed the keyword Azaire Networks being searched on iLocus all through April. So I knew something was going on (or the lack of it). Azaire Networks has indeed and unfortunately been shut down. It happened towards the end of March. Azaire had a nice data continuity story rather than pure voice handover among the FMC solutions in the market. Azaire also had a FMC security gateway product, something similar to Reefpoint (NextPoint) and Netrake (Audiocodes). Acme Packet and other SBC vendors have been trying their luck in the FMC networks. Azaire and others were supposed to have an edge there. Apart from security gateway and multimedia continuity solution, Azaire had about half a dozen trials going on for its VCC solutions. One such trial was at Orange.
Another recent casualty is Verso that collapsed under its own weight. Never easy to make a business out of acquiring 50 different stressed companies. One of the companies that Verso had picked was Clarent, a first generation VoIP vendor of the nineties. Cisco once called Clarent as the biggest competitive threat Cisco faced in service provider VoIP business.
My writeup on Skype Journal posted day before yesterday ........
Skype’s recent mobile VoIP announcement is an admission that VoIP over 3G is not practical yet. This is evident from the fact that Skype has chosen to implement iSkoot type architecture for its own mobile VoIP offering that it announced last week.
• 435k IPTV subscribers. Net additions during the quarter: 41k. Number of subscribers increased 10% q-o-q
• “1.3% Local revenue decrease and the 4.4% Domestic Long Distance revenue decrease, mainly due to mobile and VoIP substitution”.
• 4.28m broadband subscribers (including ADSL and FTTB) at the end of the 1Q08. 3.6m ADSL subscribers, 634.5 thousand FTTX subscribers, which is up 97.7 thousand in the quarter.
Why have you so far stayed clear of direct-to-consumer B2C offering?
That part of our business has not received much attention yet. But that will change and my guess is that millions – if not tens of millions – will be using our direct-to-consumer service. So there is going to be much more emphasis on viral marketing, search engine optimization etc to drive that. There will be more handset OEM deals.
Like your B2B offering, your B2C offering will also be dependent on MNOs somewhat. Right?
Not so much. Rather than hosting the termination management POPs at the operator NOC, we will host them on our own NOCs. We have NOCs around the world that we use for direct-to-consumer business.
How do you make money from your B2C offering?
The SkypeOut feature is turned on in this B2C offering and that is how we generate revenue, generating SkypeOut minutes.
How many B2C customers do you have?
Right now we have hundreds of thousands using our B2C service.
Going over to your B2B solution, why would a mobile operator want to work with you? Are you not cannibalizing their high margin long distance business?
For Skype-to-Skype calls what the operators have determined is that we are not cannibalising their business. In fact we are adding to their business because if I am on Skype and my friend in Hong Kong is on Skype, I was not going to call that person for two dollars a minute anyway. I would have waited to Skype them over PC and completely cut out the operator. So by putting Skype on mobile phone at least the operator is able to use up the free minutes. With SkypeOut which involves calls to a non-Skype user, you probably had no choice but to call with very expensive ILD rates. So some of the operators prefer not to turn SkypeOut on. At least not at first. They are however realistic about the ILD rates substantially reducing over time with the calling cards and callback and Skype etc. At some point therefore they will also turn on SkypeOut.
Sonus started displacing Huawei some 6 months back at Convergia. Actually as a first step Convergia stopped investment in Hauwei. They went back to original infrastructure and then selected Sonus to replace the systems.
Convergia had deployed Huawei NGN gear some 3 years back. At the time it was supposed to be biggest Huawei NGN deal outside China: “One of the largest cross country VoIP networks - 17 countries.” See slide 19 of this presentation.
RE Skype’s future, I had put up a post last week suggesting a telco acquisition would be the second best option (the best option being a spinoff) for Skype. A lot of you fine gentlemen do not agree with my suggestion. So let me highlight one more point today.
Skype can potentially save a telco $3 billion. Here is how: The proportion of IP-to-IP calls is increasing. It is directly proportional to the penetration of VoIP in a country. So if France has 25% VoIP penetration, expect 25% of VoIP traffic in France to be IP-to-IP. However even for these on-net calls, telcos use licensed call servers (apart from a few daring telcos who use open source platforms … I will be writing about those telcos next month). All their VoIP calls make use of a call agent to seek routing information and a feature server to enable features such as Caller ID etc. It is a client-server setup. Skype on the other hand is a peer-to-peer set up. There are no core network servers involved.
I have not been able to post many blogs lately … totally engrossed in the 9th annual VoIP industry report which I am trying to get published within the next few days. One of the bits I missed on this week was Atreus acquisition by Sonus. I am posting a few notes below:
Is this a good move from Sonus?
Sonus has been working with best of the breed providers for a while. They understand value of a ‘solutions sale’ rather than just the ‘systems sale’. With a solutions sale there is usually quite a bit of money associated in form of professional services. Almost every Atreus deployment requires a significant amount of customization which generates services revenue. There is an increasing focus on services revenue component by large vendors.
Will this acquisition result in new accounts for Sonus? What are the chances of this acquisition getting Sonus inside someone like Telus and KPN?
Skype is one of those faces that launched a thousand ships! Its big bang acquisition and open developer program led to the formation of numerous VoIP startups. But the company itself got entangled in a loveless relationship that …. as expected …. never proved synergetic. At this juncture when eBay is understood to be exploring different options for Skype, I am going to suggest a couple of places where Skype could find love.
Let me first touch upon the reasons why the eBay-Skype synergy never materialized. eBay acquired Skype in order to woo large retailers into setting up their storefronts on eBay. That did not happen. Even if it had, the kind of click-to-call capability that Skype would have been able to bring in would not have sufficed. The reason for that is the missing third ‘C’. The Context. Click-to-call in a consumer driven environment is what some of you would call a Web 3.0 application. We are not quite there yet. Translation: Suppose a retailer such as Apple set up a storefront on eBay. In order for its customer services agent to better serve the click caller, the agent needs to know the profile of the caller as well as the caller activity on the site prior to the call. That piece of info flashed to the agent screen at the time of the call is Context.
It is quite usual for enterprises to have multiple IP bandwidth links. That increases churn in business IP trunking. Obviously as a service provider, you would try to provide better QoS in order to counter that churn. If however your partner carriers use voice quality solutions from different vendors, I am told that you would have problems implementing QoS end-to-end. Twelve years of VoIP and you still have interop issues!
There are over 65 million fixed line VoIP subscribers worldwide. Nearly all of them – as well as a lot of business VoIP customers - are being served by public Internet. Being able to manage voice quality over public Internet has to be an important issue unless the service providers are content with keeping VoIP as a secondary option. Most of the service providers are burying their heads in the sand hoping the bandwidth in the backbones will remain abundant. That is a bad assumption. Shared transportation will not always work, especially since video is going to consume a lot of capacity.
On the other hand, the fact that there is a big player in the market validates the potential of the segment.
