Not to be outdone by BT, Cable & Wireless (the U.K.'s second largest network operator) unveiled plans earlier in April to build its own NGN over the next few years.  The total CAPEX earmarked for the project is roughly $988 million (from 2005-2008).  The plan calls for the collapsing of 70 voice switches to 10 softswitches, to be housed in "IP routing nodes".  Furthermore, the few hundred metro nodes will be replaced by roughly 80.

Last week, C&W announced Tellabs as the first supplier for its NGN.  C&W will rely on the 8800 metro Ethernet multiservice routers, whose strong suit is the aggregation of multiple types of legacy traffic.  Another deal was announced with Marconi (this one on the access side).  Marconi is one of the incumbents on the Bulldog Communications (C&W's access business unit).  However, more announcements are expected within the next few weeks.  Cisco could likely win a piece of the IP core.  Nortel could potentially get a piece of the softswitching piece of the upgrade, thereby capitalizing on its Succession win last year.

   more »

As more details emerge from BT's 21st Century Network tender, I started making some back envelope calculations to see what the potential opportunity would be for several vendors.  The consensus estimate is that about 40% of the total $19 billion kitty is to be earmarked to equipment vendors (or about $8 billion), to be spent within a 6 year horizon.

So which vendors will be the ones that will be able to reap the most rewards?  Cisco, for one, should get a good portion of the core and metro routing piece, as it sells its products directly, rather than relying on a channel.  By contrast, Juniper has to go through two channels: Lucent (core) and Siemens (metro).  Cisco could get between $800 - $850 million of the total $1.4 billion part.

Lucent was the winner thus far in the professional services category, getting about $100 million in network services and network management software (for the Juniper routers).  As the migration from the legacy equipment installed base to the new IP continues, the company should be well positioned to get more professional services business, albeit it will get a good run for its money from Siemens.

Huawei gets a substantial piece ($1.5 billion) of the total CAPEX pie, although it was very aggressive on its pricing.  On the other hand, BT is a great reference account for the Chinese-based manufacturer.

Of the remaining vendors (Ciena, Alcatel, Ericsson), I would like to emphasize Ericsson in particular, since it was the winner of the VoIP / softswitching sweepstakes of 21CN RFP.  That particular portion of the contract is worth in the vicinity of $500 million.  The win by Ericsson also signals BT's intent to keep its wireless / MVNO options open.  This represents yet another instance of an incumbent vendor being able to leverage its position to win the next-gen piece.

   more »

A bit more than a week ago, Ericsson became the latest Euro vendor to form a relationship with a Chinese player to jointly develop TD-SCDMA offerings.  Known as the "Chinese version of 3G", TD-SCDMA (Time Division Synchronous Code Division Multiple Access) has been originally specified by the Chinese Academy of Telecommunications Technology.  The standard combines older TDMA with the Time-Division Duplexing (TDD) technique of broadcasting over a single chunk of spectrum, rather than the normal two bands.

The three big Chinese manufacturers now all have TD-SCDMA agreements in place with foreign vendors (i.e. Huawei is teaming up with Siemens, ZTE is pursuing Ericsson and Datang struck a deal with Alcatel).  The announcement was consistent with Hakan Eriksson's statement during the Ericsson analyst day the previous week: "TD-SCDMA has not been prioritized but Ericsson will have a presence".  Eriksson added that Ericsson regards this as a Chinese-centric technology, and as such, the company could rely on partnerships.

It is anticipated that the Chinese government will issue three 3G licenses later in 2005, and chances are each will be for a different wireless technology: CDMA EV-DO, WCDMA, and TD-SCDMA.  There could be some delay, as the Chinese government keeps on promoting TD-SCDMA.

   more »

The IMS (IP Multimedia Subsystem) framework is gaining more and more notoriety these days, becoming one of the hottest topics du jour in today's telecom world.  However, much work remains to be done, including issues such standardization, the refinement of voice QoS mechanisms (for a conversational class of service) in the wireless environment, and the availability of some new "combinational" services that can bundle features such as location and presence.  So fittingly (after playing a bit with Photoshop), I spent some time creating the above scoreboard that provides a good snapshot of the current state of the IMS market:  we are still in the 2nd inning of a 9 inning-game.  2005 is shaping up to be more of a year of experimentation rather than full-scale deployments.

Note: I wrote an article regarding IMS that appeared on the May issue of the IPCC Newsletter, and will be giving a talk about IMS at Supercomm.

   more »

Ben Charny picked up an interesting thread from a Deutsche Bank report indicating that the 3G evolution at Cingular might not be going as fast as previously anticipated.  It seems that the integration with AT&T Wireless and the deployment of EDGE are taking away substantial resources from Cingular's 3G initiative.  Read it all here.   more »

It looks like the uptake of data wireless services has been quite good in South Korea.  The data portion of the ARPU for mobile operators has grown steadily in 2004, with a gradual crescendo that peaked in Q4 (16.5% as opposed to a 13.9% increase in Q2).  Year-over-year, the results were even better, with data services growing at a rate of 36% YoY in Q4.  SK Telecom has lead the way, notching almost a 22% growth, up from 17.5% from two quarters before.  The operator is currently investing in a WCDMA/HSDPA network in parallel with its EV-DO network, with almost 40% of its CAPEX being earmarked towards this new HSDPA initiative.  About 35% of SK Telecom's subscribers are on its 3G EV-DO network and 19% have VOD (via the "June" service offering).  The Korean Network Information Center claimed that more than 40% of the mobile users utilized Internet services (the data is from a survey taken in the second semester of 2004).  The future prospects look bright, since 84% of teenagers have these services and 63% of folks in their 20s. 

   more »

South Korea perhaps represents one of the best examples of where convergence is taking hold.  About 12 million out of a total of 15 million households have Internet access (typically around the clock, 24 hours per day) for a flat rate of 30,000 won (1 US$ is roughly equivalent to 1,004 won, so this is equivalent to under $30). 

There are quite a few reasons why this is happening, including, among others:

• high household broadband penetration: 73% versus 26% in the US and 22% in Europe (South Korea is the global leader)
• high throughputs: South Korea has among the fastest speeds available to consumers (average of 14 Mbps compared to only 4 Mbps in the US)
• favorable demographic characteristics: 65% of the population lives in MDUs (Multiple Dwelling Units)
• excellent broadband facilities for the masses: South Korea counts with roughly 22,000 "PC Bang" cyber cafes and 1.2 million PC terminals

These factors translate to one of the most favorable customer profiles in the world.  South Koreans Internet users spend an average of 20 hours online per week (compared to under 13 for the US).  Roughly 12% of the retail sales are online (versus only 2% in the US) and 68% of stock trading is done online (versus less than 25% in the US).  Interesting to see how pragmatic regulation played a small positive role here - the restriction of Japanese imports made the PC the primary gaming platform in South Korea, and many feel that was one of the drivers for these impressive broadband numbers.

Note: The South Korean wireless market will be discussed in a separate post. 

   more »

Om Malik writes about VDSL2, and how much bandwidth it can deliver. It turns out that the new VDSL2 specification is about to be ratified by the ITU sometime next week, and that seems to be generating a lot of excitement among FTTN / FTTC proponents. Does this mean that last-mile copper solutions will be in fact effective for all customers, no matter how many miles they happen to be away from the CO?  As I wrote before, there is an inverse relationship between the loop length and the bit rate that can be supported by such last-mile copper solutions (be they ADSL2+, VDSL2, etc.).  For a minimum level comfort zone, for HDTV, around 19 Mbps are necessary (from a future proofing network perspective).  This means that the maximum loop length supporting that would be somewhere between 5 and 6 kilofeet, which won't cut it in about 57% of the cases for a carrier like SBC.  So FTTN or FTTC solutions represent a "go after the low hanging fruit" type strategy. 

On the other hand, in Canada, the situation is quite different, as Canadian carriers such as Bell Canada and MTS have very good loop length profiles.  In fact, according to Bell, 85 to 90 percent of its customer dwellings are within 1,200 meters of the node.  So FTTN with VDSL2/ADSL2+ seems to be the right approach for Canadian ILECs such as Bell to bring the triple play to all of their customers as quickly as possible.

   more »

Despite making a push with a lot of handset promotions, the numbers for CDMA net adds in April for China Unicom were substantially down from a year ago (from 680k to 495k).  There were rumors of the lack of handsets, albeit the problem was mitigated with a big order or low-end CDMA mobile phones.  From a CDMA camp perspective, these results might not look so great - despite the fact that China Unicom faced stiff competition from China Mobile.  Just earlier this week, China Unicom had announced plans to upgrade its national CDMA network to 3G, and one has to wonder how soon this upgrade will come after these results.  Despite these results in China, the outlook looks much brighter for CDMA in countries such as Japan, Korea, India and Brazil.

   more »

A new concept is getting a lot of attention in the UK, namely that of a virtual broadband operator. An FT story documented the advent of a broadband Internet service that is custom-tailored to a football fan (ed.note: in the UK, as in my native Brazil, they call the game football, not soccer ;-).

Viatel, a UK-based broadband operator, introduced earlier this week a new branded broadband service specifically geared towards Chelsea FC fans (ed.note: Chelsea won the English Premier League title this year).  Chelsea Broadband will cost fans £28.99 per month and offer a 2Mbps high-speed Internet data connection with unlimited monthly downloads and up to five personal e-mail addresses.  Another "sticky" element added to the bundle deal is a free year subscription to Chelsea TV online, and as an added bonus, for the first 1,000 subscribers to sign on, free video highlights from Sky Sports. Chelsea will promote the service online, in its matchday program and around its Stamford Bridge stadium.    

Undoubtedly, this is yet another major coup for Viatel, which already had similar deals in place with Manchester United (from the English Premier League) and Hearts FC (from the Scottish Premier League).  The company also has a site (www.fastfooty.com) where fans can register their interest in having a similar service for their own soccer clubs.  But the math is impressive: between Chelsea and Manchester Utd. alone, the fan base represents a huge total addressable market opportunity of 14 million fans in the UK alone (about one fifth's of the UK population). 

Here are a few more ideas added to the mix: what about a similar MVNO type deal for wireless operators?  Special screens, ringtones, phone updates, and video streaming of a few highlights can be a part of a great "sticky" wireless bundle.  Would a player such as Virgin step up to the plate?

I wonder if on this side of the Ocean, other broadband service providers would consider a similar play.  Closer to home, here in Toronto, Rogers could definitely come up with a similar type packaging for a Blue Jays (owned by Rogers) branded broadband service.  And they can even include a few free tickets as an extra bonus.  I know a few die-hard Blue Jays fans that would definitely find this idea very appealing.

   more »

Microsoft and Sprint are teaming up to introduce a wireless location based service geared towards helping businesses locate, track and direct mobile workers through the use of text and voice messaging.  The service is expected to be available across all of the US and is part of the Sprint Business Mobility Framework, which is being introduced after recent successes with some target customers over the past few months. 

Sprint will rely on Microsoft's Map Point Location Server (MLS), which can do a lot more than just mapping and routing information from the MapPoint Web Service.  The MLS software also can perform real-time tracking and management of highly mobile employees (such as "road warriors") or vehicle fleets.  This requires a partnership with a service provider and Sprint represents the first US operator to sign on.  Last year, there were many similar deals announced with service providers in Sweden and the UK.  Closer to home, Bell Canada was the first carrier to provide the service, in a deal announced last year.

Location and presence definitely are part of a recipe for a potential killer app: the possibilities are endless. Imagine an end-user getting pushed with some information when he/she is at a mall, being able to find out a movie is playing in that complex, and then finding out a person from his/her buddy list happens to be there as well, and can receive an SMS invitation to watch the movie together.  The IMS (IP Multimedia Subsystem) framework can be instrumental in allowing service providers to start experimenting with these so-called "combinational services".  Therefore, expect location based services to get more and more sophisticated in the future.

   more »

A week ago, Trapeze Networks announced yet another partnership, this time with Enterasys.  Trapeze was picked as Nortel's WLAN partner after Cisco bought Airespace (the previous WLAN partner that Nortel had).  As I already pointed out earlier this year, the WLAN space will still see some M&A activity happening this year, given the continued growth (despite the fact that Cisco pretty much owns half of the market).  So after seeing Trapeze engage in Airespace-like behavior and establish other relationships with vendors such as 3Com, Enterasys and D-Link, will Nortel risk having its own WLAN product roadmap derailed again (in the event that one of the above companies decide to acquire Trapeze)?  Stay tuned - we will soon find out the answer, but I am willing to make a strong bet that Trapeze is the next WLAN company most likely to be bought.

---

Note: Airespace had OEM agreements with Alcatel, NEC and Nortel before being snapped up by Cisco.

Note2: Proxim can also be a potential acquisition target, as there are continuing rumors that the company is in a messy financial situation.  Last week, Proxim shares dropped almost 60% after the company revealed it had "an immediate need" for additional financing, and might seek bankruptcy protection unless it is able to be sold or obtain enough financing in Q2 of fiscal 2005.  Many feel that the Orinoco product acquisition from Agere did not go well, and then, the failure of Proxim to introduce a WLAN switch further exacerbated the problem.  But Proxim still has some good IP (Intellectual Property), and the last drop in the stock might also present an opportunity for a perspective buyer.

   more »

There was a lot of interesting coverage on the newspapers today on the aftermath of the CRTC VoIP ruling.  Mark Evans and Paul Vieira had a story on the National Post (where my opinion is stated).  I egregiously mentioned the fact that there is nothing in the ruling to prevent the MSOs from engaging in anti-competitive practices - ironically the very same behavior that the CRTC wanted to prevent the ILECs from adopting - not because I am actually 100% certain that the MSOs will do that.  Rather, my message was that should any player engage in such conduct, the CRTC always has the option to step in.  If that's the case, why not do the same with the ILECs and start from an even level playing field?  Also interesting to see the regulator quote right at the end of the article:

"No persuasive arguments were presented to the effect that ... the phone companies would not have the motivation, the means and the opportunity to engage in below-cost pricing that would have the effect of stifling competition."

Again, in an unregulated regime, should the CRTC notice this, they could always step in.  The MSOs would always hold the ILECs in check, until the equilibrium price is reached.  So would a pre-emptive salvo against the ILECs further the cause of competition?

Tyler Hamilton from the Toronto Star had a piece with the views from the incumbents, cable operators, and new entrants (Vonage, Primus, and startups such as Montreal-based BabyTel).  I am glad to see that Steve Dorsey, the BabyTel CEO (also from my alma mater, and a seasoned executive and entrepreneur in the telecom world) is right in synch with my own opinion that there is nothing in place to prevent predatory behavior from the cable companies.  BabyTel represented the lone voice of VoIP SPs pointing out this fact, given that Vonage, Primus, etc. were all praising the ruling.

Simon Tuck of the Globe and Mail also included a view of the dissenting voices in the CRTC in his article (ed. note: the final voting tally was 7-2).  Two commissioners (Andrée Wylie and Andrée Noël) were more avant garde in recognizing the fact that VoIP is different than regular telephone service. Ms. Wylie went as far as saying that the commission should have let market competition set the tone, rather than issuing “static” restraints on some VoIP providers and not on others.  I could not agree more with Ms. Wylie's opinion. VoIP should be an information service, not a telecom service.

   more »

And the final verdict of the CRTC VoIP decision announced earlier today was pretty much what most expected: nothing more than a rehash of their original position made public earlier last year.  The chances that the public hearings held last September would have changed the final outcome would be really slim. 

The key question is: will it really promote the growth of a nascent industry?  Again, the key mission of a regulator is to attempt to set the course for a market to be as competitive as it can be.  And by taking out the incumbent telcos (including Bell Canada and Telus) with a pre-emptive salvo, it is my belief that competition was not best served.  By constraining the ILECs to a certain VoIP offering price level, there is nothing preventing the MSOs (the other major VoIP players in terms of human and capital resources) from setting a price just below that level. 

From a pure microeconomics perspective, no matter what duopolistic model one chooses (Stackelberg, where competitors move sequentially, the Bertrand oligopoly, where competitors simultaneously choose prices or Cournot's duopoly game, where they choose quantities simultaneously), the "equilibrium" price level will always be smaller than for the case that one of the players (in this case the ILECs) cannot match the MSOs with a competitive move.  Let's take broadband access as an example in the US.  Some would model this as a Stackelberg game, in which the ILECs were the first to move (with DSL), with MSOs making a move in response offering cable access at a lower price.  It took the RBOCs a while to match the price, and eventually there were other pricing moves downwards.  But the fact is, there was competition - albeit not a perfectly competitive market. 

However, if one of the players is constrained on pricing moves (at least on their incumbent region), the resulting price would be undoubtedly higher than the Nash equilibrium of the duopoly.  This is just plain old simple Economics 101.  So the best prospects for lower VoIP pricing would be if the ILECs decide to attack each other's territory as a CLEC, albeit it might be costly for them (in terms of defending their own turf).

Mark Evans made a good point about which players got the early control of the VoIP market in the US.  After Vonage, Cox, Cablevision and Time Warner, the rest of the pack (about 396 players) each has roughly $70k in annual sales.  So what that really means is that the bulk of the market belongs to the MSOs and an xSP (Vonage).  Let's take this to Canada, in one region (say out west).  Shaw now has a $55 VoIP plan and is no rush to lower that down, because it does not have to.  Maybe it will a bit when Vonage will compete more, but that's about all.  Telus will not really be able to match any price drop in the future with their consumer offering, due to ruling.  And Bell West probably will not engage in a lot of competition in the Telus footprint (knowing that this move would be matched by a similar Telus move in Ontario and Quebec).  Is this the competition that the CRTC hopes to achieve?

Regulation is necessary as a tool to guarantee that players do not engage in anti-competitive practices and to ensure that some public service goals such as emergency service (E-911) and law enforcement (CALEA) are met.  The FCC showed that this model worked - and it would have been wise for the CRTC to have taken a similar approach. VoIP is still in its blossoming stages and innovation in this area should not be choked before the industry has a chance to grow.  Therefore, in my humble opinion, the CRTC should have created an even level playing field for all the industry participants.  And, in the event that the ILECs would engage in anti-competitive practices, the CRTC could always have the option of penalizing them in order to avoid a large-scale industry consolidation.

   more »

For many followers of the speech industry, this was a scenario that was envisioned years ago as a way to further establish the speech technology industry, namely a merger between giants (and arch-rivals) SpeechWorks and Nuance. 

Well, that was no longer possible after ScanSoft bought SpeechWorks back in 2003.  At that point, ScanSoft was a MA-based imaging company that started up gobbling up speech companies such as L&H and Philips Speech Processing.  But why an imaging company with expertise in technologies such as OCR (Optical Character Recognition) would want to delve into speech in the first place?  Because many of the algorithms used in OCR have some commonality with those used in speech processing (e.g. recognition). Similar discrete-event simulation and Markov chain models can be used in both cases.

Fast forward to 2005, and the news today was that ScanSoft was about to acquire Nuance for $221 million in a stock and cash deal.  And hence, a speech giant was born, having a full portfolio consisting of engines in speech recognition, text-to-speech and speaker verification.  In the ASR (automatic speech recognition) market alone, the new entity will have a 75 percent share of the market.  Will regulators allow this transaction to go through?  Probably so, due to the increased traction of other vendors such as IBM and Microsoft.  But the integration of these two companies will be a challenge, as they compete in many product lines, so there is a lot of duplication.

In the meantime, another interesting thread on this story is that private equity company Warburg Pincus will buy $75 million in ScanSoft common stock in order to help finance the merger.  Warburg Pincus is the same company that has invested in companies such as Avaya, BEA Systems, Proxim, SS8, and Veritas Software, among others.  The company obviously believes in the potential of the speech industry, particularly with the proliferation of shrink-wrap or built-in small vocabulary speech engines in mobile devices. 

   more »

Earlier this year, I had the opportunity to catch up with Owen Matthews (Sir Terry's son and CEO of Vancouver-based startup NewHeights Software, a maker of PCM - Personal Communications Manager - software).  It was a great chat - Owen certainly showed some of the same character traits that have made his father so successful in the telecom business: enthusiasm, persistence and vision.  The interview can be viewed here (ed. note: no registration necessary) and it provides us with a great insight into the mind of an up-and-coming tech entrepreneur.

In case you are not familiar with NewHeights, the company designs a personal desktop application (Desktop Assistant) that manages an end-user's communcations.  Desktop Assistant manages voice, presence, IM, directories and all other aspects related to an end-user's communcations needs. The company has gained quite a lot of acclaim over the past year, including a deal with Bell Canada, acclaim from the VC industry and partnerships with vendors such as Ubiquity and Marconi.

   more »

Even though I was unable to attend N+I this past week, I got a few tidbits from other analysts and key industry folks in attendance.  One of the key takeaways of the show was that embedded security is quickly becoming a key catalyst to change the competitive structure of the enterprise equipment market.  In terms of competitive benchmarking, Cisco seems to be better positioned than its peers in terms of embedding security, switching and applications in its routing product lines.  The recently introduced ISR (Integrated Services Router) product line epitomizes the company's push towards this integrated design. 

At N+I, Cisco announced a line of smaller devices in the ISR family, including Ethernet switching, managed security and 802.11 a/b/g access - all this at a price point below $1,500.  The products were showcased during the event, including some demos of its wireless features (albeit the incorporation of Airespace functionality is still a work in progress).  The ISR can serve as a pretty effective solution "out of a box" for SMEs and branch/remote office locations.  Juniper will combat the Cisco with its management features and its operating system.  But the big installed base and number of yearly router shipments (about 900k units per year) make Cisco the player to beat.

   more »

Linux enthusiasts got yet another reason to smile when mobile chipset giant Qualcomm announced on Thursday that it would support the operating system on its chips for CDMA mobile handsets.  The built-in Linux support will come on the company's Mobile Station Modem (MSM) 6550.  Running Linux on the cell phone eliminates the need for a separate co-processor.  Qualcomm also plans to embed Linux functionality in future chipsets that support future generations of 3G (i.e. HSDPA and 1xEV-DO Rev. A), which will likely be introduced by mobile operators around 2006.  Up to now, Qualcomm phones typically only had its own proprietary OS (called BREW real-time operating system), so this is the first time that the company supports another third-party OS. 

   more »

The CRTC got some praise by issuing its VoIP operator 911 service mandate, albeit the Commission has moved slowly in some areas such as VoIP and local number portability.  The sluggishness in taking and implementing a final stance on VoIP has been blogged ad infinitum by many, and the overall consensus is that the CRTC will hold its initial view on the subject.  In other words, that the ILECs will be subject to the same rules and regulations as traditional voice services in their own incumbent areas. 

This will restrict Bell and Telus inside their own regions, but not so outside their incumbent areas.  However, the costs to deploy VoIP remotely can be quite high, so there probably will not be too much action in that direction.  This will give MSOs (most noticeably Rogers, Shaw and Videotron, among others) and independent operators (such as Vonage and Primus) a chance to take market share away from the incumbents. 

Subjecting ILECs to having to file tariff applications is onerous and limiting to them, because every time a player wants to change the rates, it must file an application.  The new pricing, if granted, has to be applied across the band.  The only way out of price regulation is if the CRTC believes that the particular market is competitive.  Then, price regulation can be achieved via a forbeareance application, a slow process that could take up to two years.  Therefore, the only recourse of the ILECs (other than forbearance) is more aggressive bundling and/or pricing.  So under this scenario, we can expect more Michael Sabia-like moves of offering 1,000 monthly minutes worth of LD in North America for an additional $5.  The cablecos and other VoIP service providers will have an excellent opportunity to take market share away from the incumbent telcos in the highly competitive residential voice market.

But the key question is: will the key goal of creating a market as perfectly competitive as possible be served if the CRTC's initial stance is held?  I have already stated my case and believe that heavy-duty regulation should not be applied.  In due time, however, the details will be uncovered and we will know which way the CRTC is leaning towards.more »

Back in March, I wrote the following in a note regarding the importance of five 9s reliability:

"Of course five 9s is highly desirable, because of the possible consequences of a service outage (even as temporary as it might be).  Here, the notion of expected value takes hold - i.e. even starting with a low probability event, when factoring in the bad consequences associated with it, the expected value could be high (e.g. someone who is relying on VoIP service as a primary line and all of a sudden cannot call an ambulance due to a service interruption)."

Not surprisingly, a Florida couple is trying to shut down Vonage service after they could not reach 911 during an emergency, resulting in the death of their baby girl (hat tip: Om Malik).  Regardless of the stance of a regulatory body on VoIP ("light" versus "full" regulation), there are a few social obligations that have to be fulfilled no matter what.  E911 and CALEA should be addressed (at the very least, if there is no reliable E911, consumers should be warned, and any provider claiming to offer this feature is obliged to comply with the strict guidelines for the enhanced emergency service.  If it cannot, then it is incumbent upon that provider to explicitly state that the end-user is on the hook for another E911 alternative (e.g. cell phone).

Three states (Texas, Michigan and Connecticut) are now taking action against Vonage due to the E911 issue.  The emergency service does not come pre-installed with the basic Vonage VoIP package; in order to activate the service, users need to provide the company with their addresses. In Vonage's defense, the E911 technology is typically handled by the local carriers, and getting them to open up that technology is a challenge (as claimed in this Technology Review article).  Thus far, only Bell South has been responsive, albeit it is demanding Vonage to get a CLEC certificate.  In Rhode Island, Vonage could get an E911 trial off the ground because the state owns and operates their system.

Also not surprisingly, the new FCC Chair is rumored to propose at the May 19th meeting that the Commission should require VoIP service providers to begin delivering 911 calls by September.  In other words, Martin would give a 120 day warning for all these VoIP operators to implement the service or else... Now, the "or else" clause can range from explicitly state that the VoIP service is strictly "add-on" and not a "replacement" of traditional telephony to a drastic, outright "shutdown" of the service.  I am particularly in favor of the former rather than the latter measure.  Under such a ruling, Skype, which clearly states that they do not recommend that people drop their regular phone, would be OK.  But in either case, the FCC should demand that RBOCs should open up their E911 centers to VoIP operators - and why not?  Those ILECs are losing money with E911, so this should give them an ability to cover some of their costs.

---

Note: For more information on E911 technology, this insight that I wrote back in 2001 should be quite useful.   more »

If Siemens is trying to unload its ailing mobile handset business, it will have to try doing this just a bit harder.  According to a report in yesterday's Frankfurter Allgemeine Zeitung German daily, just as Motorola execs were about to sign an agreement between both companies, they had a last minute change of heart and backed away from the deal.  This is a bit of a setback for Siemens, but I'd be willing to bet that there could be a potential suitor down the road (maybe an Asian player such as Samsung or LGE).

   more »

Business Week Online's blog (Blogspotting) has had quite a few interesting articles since its recent inception. One of them in particular struck a chord with yours truly, namely the one about the crunkie concept.  The idea is to post entries to a blog that can be pushed to cell phones of members of a buddy list whenever any of them are present in certain locations (for instance, a message saying that a certain product is now on sale whenever a friend walks through a shopping district).  It is one of the first attempts to connect presence and location to generate a new service - interestingly enough, this blog entry is called a Crunkie.  The product is the creation of Wavemarket - an app company funded by Qualcomm and early-stage VC firm Draper Fisher Jurvetson (which also has invested in companies such as Skype and SocialText).

   more »

After making a nice return on the MCI transaction, it looks like Carlos Slim Helu is ready to take on a next challenge.  And why not?  His astuteness already made him move up to #17 in the Forbes 2004 list of the world's wealthiest.  Although he has been successful in selling MCI for a tidy profit (he bought the shares at Filene's basement prices when MCI was emerging from bankruptcy and made close to $1.1 billion by selling his 13 percent stake to Verizon), the sale of his shares effectively leave him without a major foothold in the US market.  But the reality was that he made it quite clear that his position in MCI was strictly "a financial investment".

So where will he spend the proceeds from the sale?  Many felt that it would be in a region with which he is intimately familiar with (Latin America).  However today, at America Movil's investor day, management mentioned a few times that it is interested in entering the Spanish market (Mr. Slim owns America Movil and Telmex, among other holdings).  This foray into Spain can be achieved via a license purchase or via an MVNO play.  Recently, the Spanish telecom regulator (CMT) announced that it might auction 10 MHz of wireless spectrum under the same "beauty contest" rules of the March 2000 tender.  America Movil (NYSE:AMX) believes that the tariff levels in Spain are very high, and there is an opportunity for them to be a player.  With Telefonica getting more and more aggressive in Mexico (thanks to funds from its operations in Brazil and Spain), AMX currently has no means of striking back in enemy territory, despite its presence in Brazil.  So the move would make sense from a defensive or counter-striking perspective, however a full Brazilian-like offensive strike would not be productive.  This is an ongoing development that will get a lot of attention in the next few months.

   more »

Looks like Microsoft is "enabling" more and more telecom-based applications.  The latest one is the contact center, with the introduction of its Customer Care Framework, a software platform that can integrate existing OSS, BSS and CRM systems from service providers, consolidate different applications and automate workflow.  The product is based on .NET and the Web services architecture.  This development could potentially open the door for future expansion into the communications area (including call control, etc.), whenever Microsoft chooses to attack that space.

   more »

After a long battle, it seems that Qwest pulled out of the MCI auction leaving Verizon as the winner by default. Verizon's offer ($8.54 billion) was roughly 13 percent below the $9.85 billion offered by Qwest. After the MCI board rejected the higher Qwest bid for the fourth time in a row, the Denver-based company announced that " It is no longer in the best interests of shareowners, customers and employees to continue in a process that seems to be permanently skewed against Qwest".  The key question is what is next for Qwest.  The company might target assests that giants Verizon and SBC might be forced to sell to mitigate government concerns that the mergers will be detrimental to competition.  Other alternatives include going after companies selling services to businesses, pursuing a merger with Sprint or becoming itself an acquisition target.   more »

Om Malik had an interesting piece on the ascent of Huawei, particularly on the coattails of the impressive win in the BT 21 CN RFP sweepstakes.  Huawei was one of the underdogs in the access piece (most folks had expected Alcatel and Marconi to be the winners in this category instead of Huawei and Fujitsu) and in the optical portion of the BT tender. 

Undoubtedly, the selection of Huawei in the two segments of the BT network (access and optical transport) clearly signals a validation of the quality of their portfolio, along with the ability to support a large European operator.  But the bigger questions to be asked are: 1- Does this victory also indicate a growing long-term presence of low-cost Chinese suppliers into Tier 1 accounts? and 2- Does the win also trigger a negative trend for all major non-Chinese suppliers (such as Cisco, Ericsson, Juniper, Lucent, Nortel, etc.)? 

The answer to the first question is definitely yes, but a more interesting consideration is that thus far, Huawei was picked for areas that have already been commoditized due to very high competitive pressures.  These low-margin LOBs include broadband access and optical networking (as in the 21 CN case).  However, other areas such as softswitching, network intelligence and routing are apparently safe thus far.  However, that is how most new entrants in a market begin attacking the bases of older, more established vendors - first with the low-margin products, and then, moving up the value chain.  So while the presence of Huawei in long-haul transport implies a potential margin erosion of vendors such as Ciena, the other vendors will be watching this development closely - because they know they can be next.

The second question is a bit more challenging, as other carriers might start asking: "If Huawei's gear was good enough for BT, then why can't it work for us?" - and traditional vendors really need to get a good handle on that question.  If more telcos start buying Huawei equipment, then Western vendors would really be faced with an uneven playing field, given the price differentials (due to the artificially undervalued Yuan) and the massive loan guarantees and subsidies being given to Huawei by the Chinese government (to the tune of US$ 10 billion).  This will not mix well with the already fragile telecom sector.  So what can these vendors do?  Lobby for Embraer-style subsidies (ed.note: Embraer is the Brazilian plane maker which receives a great deal of incentives from the Brazilian government)?  Press the WTO, their governments and other authorities to put pressure on the Chinese to float the Yuan?  Move up the value chain and focus on services and software/applications?  Partner with these players?  These are challenging questions, and their ramifications are even more complex.  But the one clear result of this increased competitive threat is margin erosion and lower pricing: carriers are getting to be very good at playing vendors off each other. 

Vendors such as Huawei, UT Starcom, and ZTE, among others are buying assets (CommWorks, and now even perhaps Marconi), partnering (3Com) and now winning bids in major tier 1 accounts (such as BT).  Their threat to North American and European vendors is real and has finally arrived.  They represent formidable competitors that have great resources (human and capital) at their disposal.  So we better start getting used to the idea of including Huawei in the potential winners list, even for carrier RFPs on this side of the Atlantic.

   more »

Mark Evans wrote about the compensation package that Nortel CEO Bill Owens gets (including sallary, bonus, pension, and other incentives).  A close inspection of the benefits (on the company's latest 10-K) reveals a pretty generous package, including a base salary of US$1-million a year (with a chance of bonuses worth up to 170% of that base), a special pension worth about US$33,540 a month for the five years following Mr. Owen's retirement, and an added "special pension benefit" in March of US$4.5-million to compensate Mr. Owens for equity incentives he forfeited when he resigned from the boards of other other companies after he was hired as Nortel's president and CEO last April.

While that might seem like a pretty good deal on the surface, in terms of CEO compensation packages, it is nowhere near what some top corporate czars typically command.  Let's take a few comparative examples mentioned in this article.  AT&T CEO David Dorman got a $2 million bonus for 2004 (note: bonus only - that does not even include his base salary), despite a drop in revenues of nearly 12% for that year.  Lucent CEO Pat Russo got a generous package from Lucent's board last December, totalling $13.6 million, including stock, options and a $2.95 million bonus (ed.note: at least in this case, Lucent did register a profitable year, the company's first since 2000).  And here's another data point: Ivan Seindenberg, who made $20.8 million last year (salary plus incentives).

So the real question is: compared to the above figures, how is Mr. Owens' total compensation package?  I would say Nortel got a pretty good deal there (compared to other peers).  But having established that and leaving the comparisons aside for a moment, what would be the best formula in general to determine how much a CEO should be compensated?  I would suggest performance-restricted stock is a flexible way, since this can be tied to a set of metrics including profitability, return on assets, etc.  However, a short term goal (e.g. EPS) should be combined (from a compensation perspective) with a longer-term objective (e.g. return on invested capital).  Another alternative is premium options.  In the future, boards will begin scrutinizing a bit more the compensation their CEOs receive, but they might have a tough time determining some performance targets and appropriate pay scales. 

   more »

Earlier today, Avaya (NYSE:AV) and Juniper (Nasdaq:JNPR) announced that they formed a global strategic relationship to deliver enterprise class VoIP solutions to enterprises.  The partnership will leverage Juniper's routing and security solutions with Avaya's enterprise communications equipment, with each company being able to resell each other's products and services.

From Juniper's perspective, this deal enables the company to tap into Avaya's large installed base of PBX / IP PBX gear, and will be a catalyst to more sales of Juniper's enterprise routing solutions (including products such as the J-Series enterprise router).  This move is another piece of Juniper's puzzle in its quest for expanding its enterprise networking portfolio (after NetScreen - firewalls/VPNs, Perbit - WAN acceleration and Redline - application front end technology).

Avaya gets another relationship in the routing space (where it already had a deal with Extreme), and a best-in-class partner to jointly develop products geared to enterprise employees in remote, branch or mobile locations.  Of course, Avaya maintains that the relationship with Juniper is "complementary" to the one with Extreme, but that remains to be seen.  A more interesting consideration is whether or not we will see a subset of the Avaya call processing software being productized on a future Juniper routing solution targeting the SME market segment, and what is the time-to-market for such a product.

   more »

Mark Evans wrote a thought provoking article on the "Americanization" of Nortel and then posted the story in his blog.  It is the story that gets your thinking going (that is what makes an article great, no matter what your viewpoint happens to be - as long as you can get together with some friends for a beer and have a discussion about the material, the article's mission is accomplished).

So I felt compelled to get on my soapbox and write a quick comment, but that got me thinking a bit more.  Mark might have a valid point about of the loss of Nortel's Canadian entity, but I believe that the major "Canadian content" loss happened way before, when the ownership of the company changed.  When Jean Monty sold the controlling share that Bell Canada had of NT stock, Nortel certainly lost a bit of that identity.  There is still the Bell Canada/Nortel joint research center in Ottawa, but you do not hear so much about BNR anymore (the lab I worked at here in Toronto used to be BNR but it no longer exists).  However, that was not really the "fault" of the Nortel executives, or the board, etc.  It was just a process - Monty astutely cashed in the BCE NT chips (near the height of the stock), although what he did with the proceeds of that sale might be open to another debate altogether. 

But the point is that the Bell Canada sale of Nortel's stock definitely marked a loss of the company's Canadian identity (for one thing, obviously the shares changed hands and perhaps the foreign ownership percentage of the stock went up).  Bottom line is that this will probably continue to happen, although not quite as much.  I am talking about the change in RRSP foreign content rules, which might lead some Canadian investment houses to slowly start moderately divesting from their Canadian stocks (that had to make up 70% of their fund portfolios in order for the fund to be considered domestic).  This might lead to more of NT stock going to American hands.

As far as R&D goes, yes, it is true that Nortel already had centers in Mission Park (CA), Richardson (TX) and Raleigh/Research Triangle Park (NC) for a long time.  But that makes sense, given the fact that a lot of the telecom action in North America happens to be in the US (which has far more enterprises and big telcos - Verizon, SBC, Bell South and Qwest).  But I expect Nortel to still continue having its own R&D centers in Canada, as Canadian engineers still represent better bang-for-the-buck than their American counterparts (which are more expensive).  So I do not foresee a JDS-like massive migration to the US for Nortel.  On a separate note, from all the foreign telecom vendors that have Canadian representation, only Ericsson has a Canadian R&D center (a lab in Montreal).  That is interesting, considering how many talented people we have - perhaps Ottawa should consider offering more incentives to strengthen our own Silicon Valley (in Ottawa).

   more »

After much anticipation, the winners of BT's £10 billion (or about US$19 billion) RFP for its 21st Century Network (21 CN) were finally announced today.  The key objective is to upgrade UK's legacy PSTN and evolve it to a multi-service IP based network carrying both voice and data services.  The project will span five years and the migration of customers onto the new network is expected to start next year.  A grand total of eight vendors were selected for six major areas of this massive undertaking:

• Access (DSL, Metro WDM): Fujitsu / Huawei
• Metro Routing: Alcatel / Cisco / Siemens
• Core Routing: Cisco / Lucent (with Juniper's routers)
• iNode (network intelligence / softswitches): Ericsson
• Optical (long haul, switching): Ciena / Huawei
• Services: Lucent

The results have disappointed some Marconi investors and rumors already have started about a possible sale of the company (ironically to Huawei, one of the 21 CN winners).  Marconi's loss also has negative implications to Sonus, which was hoping to leverage its gateway product as part of the Marconi solution.  Huawei's wins on two areas (access and optical tranmission equipment) is a proof point of a higher penetration of low-priced Asian manufacturers into Tier 1 carriers.  Also, the commoditization of low margin business (broadband access and optical networking) is expected to be a contuining trend.  Cisco's win is another validation of its CRS-1 routing platform, which will be used on BT's IP/MPLS core.  Lucent comes out as a winner, getting a lucrative portion of the puzzle (services, which carry good direct operating margins of about 10%).  Lucent will resell Juniper's routing systems, but these will be souped up with Lucent's own OS software and extra tools (developed at Bell Labs) to optimize network efficiency.  The selection of Ericsson might be indicative of BT's future plans to provide its customers with a seamless wireline/wireless service.

   more »

Cisco (Nasdaq:CSCO) has also been quite active lately from an M&A standpoint, and its most recent acquisition was announced on Tuesday: Sipura Technology.  The transaction, valued at $68 million (a cash and stock options combo) solidifies the Linksys wireless division (Linksys was another acquisition made in 2003 geared towards the SOHO/small business and home markets).  Cisco had been OEMing Sipura's ATAs since 2004; analog phones and fax machines can be connected to these ATAs to create a phone line over a broadband connection (e.g. DSL).  Om Malik had an interesting post on this transaction (including the past Komodo story).  Sipura is the third company bought by Cisco this year (after Airespace and Topspin).

   more »

Besides the PEC deal, Nortel (NYSE:NT) also got a lot of press (particularly across the Ocean) on a rumor that it has been linked with Siemens on some cooperation deals involving both wireline and wireless products lines.  According to an article in German magazine Der Spiegel, executives from both companies met "secretely" near the Munich Airport to discuss product partnerships and maybe even some co-development (ed. note: well, the Genie is out of the bottle by now, but one always wonders how the press gets this info - which company was this info leaked from?).  Obviously, many are thinking about a possible spin-off of the German vendor's money losing handset division, but I doubt that Nortel would make a play there.  However, perhaps a wireless equipment play could be a remote possibility (Siemens next-gen wireless sales globally trail only Ericsson and Nokia), as could a VoIP-related deal (Nortel holds the edge over Siemens here, with almost a quarter of the global share, compared to roughly 10 percent by Siemens). 

The current consolidation in the carrier space, and the difficulties that Euro vendors have achieving a higher penetration on this side of the Atlantic (and vice-versa) help fuel the rumor mill of such deals. But, speaking about Nortel specifically, the company needs to continue on the path of building partnerships (such as the ones that it already has with LG Electronics in Korea, Putian in China, in addition to other smaller arrangements in the US and India) in order to boost its performance in the highly competitive telecom space.

   more »

Nortel (NYSE:NT) was back on the headlines today spending money to the tune of $448 million (in cash) to snatch up IT services firm PEC Solutions (the deal is expected to close later in June).  PEC is a player in the government IT Services industry, which is being pegged at an estimated $60-65 billion in 2005, and growing at annual rate of about 4%.  The idea is to merge PEC with Nortel's Federal Network Solutions unit and to create a new business unit (a US-based wholly owned subsidiary called Nortel PEC Solutions) - thereby avoiding any restrictions a Canadian company might have in competing for US government contracts.

The reaction thus far has been mixed, with some naysayers pointing the fact that this was an expensive cash transaction in an industry that is growing at a relatively modest rate.  But the flip side of that argument is that the transaction opens the door to a very profitable LOB for Nortel (13-15% margins are typical) in a market where the company was not previously able to firmly establish itself.  Plus, of course, it is a way to get CEO Bill Owens to leverage his military connections to help boost the company's chances of securing US government contracts. 

This is just one of many moves Nortel will be making this year - there should be more M&A activity, as well as some consolidation. 

   more »

Siemens' handset business has been suffering for quite sometime, despite some innovative models that the company has launched over the past year (including the SK65, which was reported here last year). Rumors have been everywhere, including a report that appeared this weekend on the Sueddeutsche Zeitung.  In that particular report, Taiwanese electronics powerhouse Acer was linked with being a potential investor in the German vendor's mobile phone unit.  Motorola has also been linked as a potential candidate in other articles.  In the meantime, losses the that cell phone unit have reached 100 million € (roughly US$ 130 million), and despite all these sales or partnership rumors, the exiting administration has given no guidance on the direction the company will take to find a solution.  The appreciation of the euro versus the dollar is further exacerbating the problem, and the company believes the € will rise over $1.35 this year.

Siemens shareholders expect incoming CEO (Peter Kleinfeld) will find a solution (i.e. spin off, seek a partner or even close the operation).  But Acer is denying the newspaper report, and the Motorola partnership does not seem a very likely scenario anymore, given that the US vendor has already made good progress into the European continent, so teaming up with Siemens would not be as advantageous (the Motorola story has been floated around for the past couple of years, before the Euro progress made by Motorola).  I had an "Eureka" moment thinking about this issue and thought of a more interesting partner that has not been mentioned too much... LG Electronics.  This could potentially create a Sony-Ericsson type entity, which can pickup strong shares in markets such as Asia/Pacific, Europe and the Americas.

   more »

Another interesting panel that I participated on at VON Canada was the IMS (IP Multimedia Subsystem) general session.  So what is IMS?  Many folks still think of IMS as it was originally by design, i.e. a spec geared towards delivering SIP services to the wireless space (as defined by the 3GPP folks).  In reality, it is much bigger than that. 

The best way to define IMS is as an access agnostic spec that enables service providers to deliver a new generation of SIP-based multimedia apps (in the SS7 and SIP domains).  This allows for the convergence of wireline and wireless network architectures and the benefits are OPEX savings (due to a decreasing incremental cost of deploying apps within this framework), a faster time-to-market (due to reduced complexity) and the new apps, which are geared towards increasing subscriber stickiness and the uptake of enhanced services.  The key idea is the re-use of common functions: IMS takes the network from a vertical integration modus operandi to a horizontal architecture model where these common functions are re-utilized.

In other words, IMS is supposed to be everything the intelligent network was supposed to be plus more.  But before drinking all the vendor Kool Aid, much work still needs to be done, both in terms of the actual IMS standard itself, the value proposition of vendors to service providers, the conceptualization of new services and the establishment of new pricing models that will make this network evolution a bit more palatable to service providers.

The session proved to be really interesting - I was particularly impressed with the candor of all the panelists.  Some folks might have expected that given the fact that there were four vendors (IP Unity, Lucent, MetaSwitch and Nortel) and one lone service provider (Bell Canada), the tone of the session would have been based on more hype and promotion.  However, that was not at all the case - in fact, the session provided a realistic view of the current state of the market, the realities facing both vendors and service providers, as well as the issues that still need to be worked on.  I will be posting more on IMS in the future, so stay tuned.

   more »

Intel announced a couple of days ago the volume production of its IEEE 802.16 standard WiMax modem chipset (known as Rosedale), geared towards equipment vendors that are targeting the broadband wireless market.  The chip, now officially called the Intel PRO/Wireless 5116 broadband interface device, is fabbed externally and utilizes a programmable architecture.  Intel is only delivering the silicon, with the vendors being able to implement their own software up to the MAC.  Intel expects most vendors will use 5 MHz for channel bandwidth thereby yielding roughly 10 Mbps of IP traffic. 

The chips will be used in CPE gear and eventually in wireless base stations.  The advent of the 802.16e standard (expected to be ratified later in 2005) will lead to better power efficiencies and the silicon will be targeted for desktops by 2007.  As of today, Intel is producing the Rosedale chips externally and as the uptake increases, the production is expected to shift in house.  Despite the fact that 2005 volumes are expected to remain relatively low, there is a lot of hope that with 802.16e, reduction in costs, higher volumes and economies of scale can be achieved. 

Despite the fact that this is still probably the bottom of the first or top of the second inning in terms of WiMax (to use the baseball analogy), I believe that this is still a very important milestone for Intel, because the company has finally delivered the next step in its vision.  Also significant the fact that Intel currently has 12 hardware customers and 100 operators planning to sample their Rosedale chips (the manufacturer customer list includes the likes of Airspan, Alvarion, Aperto Networks, Axxcelera Broadband Wireless, Gemtek, Huawei, Proxim, Redline Communications, Siemens Mobile, SR Telecom and ZTE).  To add a further dose of realism (and silence some skeptics), a total of 15 carriers worldwide will be announcing their intentions to trial the Intel 5116 WiMax chips.  The list includes: AT&T, Altitude Telecom, BT, Brasil Telecom, ETB, Iberlanda, Millicom, Qwest, Sify, Speakeasy, South Africa Telkom, Telmex, Tower Stream and UHT. 

How long before the Centrino model takes over WiMax as well?  Well, it will be a while, as power efficiencies of 802.16e will be needed.  So this second phase push probably will not happen until 2007, at which point the technology will start getting moved into notebooks.

   more »

I will definitely be posting more on the Skype story (got a couple of separate threads there worthy of separate postings).  But first, an insight on the analyst panel at VON Canada, which counted with Jon Arnold (a friend and ex-colleague of mine at Frost&Sullivan), Jeffrey Fan (director and equity research analyst at UBS here in Toronto) and yours truly.  I understand that due to force majeure reasons, Daniel Berninger had to drop out, and I certainly hope that everything will be OK with his kid.

The discussion centered around VoIP in three key areas: enterprise, service provider and the MSOs (the focus of most of Jeff's research).  Jon asked a few thought provoking questions, particularly about the rollout of fiber-to-the-x to various homes.  One thing I definitely brought up is that in contrast to their US counterparts, Canadian carriers such as Bell and MTS have very good loop length profiles.  In fact, according to Bell, 85 to 90 percent of its customer dwellings are within 1,200 meters of the node (that's about 3,600 feet, well within the 5 kft. distance to ensure a bandwidth of between 15-20 Mbps for a FTTN deployment with ADSL2+ or VDSL2 as a last mile technology - note Bell is going with a VDSL solution).  That is way better than SBC, for instance, which has over 57 percent of its loop lengths exceeding 6 kilofeet.  So FTTN with VDSL2/ADSL2+ seems to be the right approach for Canadian ILECs such as Bell to bring the triple play to all of their customers as quickly as possible.

Jeff mentioned the triple play efforts of some MSOs, including Shaw and Rogers.  Interestingly enough, nobody from the audience asked any of us whether or not the $55 that Shaw is charging per month will be able to deliver the uptake in triple play services.  Obviously, since Bell and Shaw own the only two DBS licenses in Canada, that leaves Telus a bit vulnerable on the video side (until the latter launches its own IP-TV solution) - and until that point in time, maybe Shaw will just continue to keep its price at that level.

Jon also made a few comments about the uptake of VoIP in the enterprise, where thus far, the CPE approach seems to be winning.  The slow uptake of IP Centrex has surprised some, but given the dynamics of the market, and the typical lenghts of contracts offered, in hindsight, the modest ramp-up was to be expceted. 

All in all, it was an interesting session - it would have been nice to have Daniel in there, as well as Cody Willard (perhaps we can have him in a future session). 

---

Note: Mark Evans wrote about Merrill Lynch analyst Glenn Campbell's estimate that Shaw is signing up about 1,000 customers a week for cable telephony in Calgary, but the question still remains whether Shaw can reach 20% penetration of cable subscribers within 5 years.

   more »

One of the highlights of Day 1 at VON Canada was Niklas Zennström keynote speech right before lunch.  Mr. Zennström is the CEO of Skype, the peer-to-peer VoIP client that is taking the world by storm, earning huge coverage in the blogsphere, from Andy Abramson to ZDNet blogs (by authors such as Russell Shaw). 

Despite the skepticism of some bloggers such as Om Malik, the results of Skype over the past year are nothing short of amazing.  According to the numbers in Niklas' presentation, the company is attaining a growth rate of 1000% a year, reaching an installed base of roughly 35 million users worldwide.  Given that Verisign is pegging the global number of Internet users at 900 million, Skype has captured almost 3.89% of the total number of cybernauts out there.  But that also includes users still bogged down by low bandwidth, so a more telling statistic would be the penetration of Skype into the global installed base of broadband users.  A quick scan at ITFacts.biz reveals that 150.5 million users around the globe have broadband access.  This yields a Skype penetration figure of about 23.3 percent, which is pretty good for a company having a product that has been launched less than two years ago.

More importantly, Skype has also been very adept at increasing the take rate of premium services (i.e. getting the subscribers that have been obtaining the free service to pay for extra services such as SkypeOut or SkypeIn).  Thus far, the company is counting mostly on SkypeOut, and is getting a take rate of roughly 3.43% of its users (corresponding to 1.2 million SkypeOut accounts).

Figuring out the annual ARPU for SkypeOut might be a bit tricky, but here are a few datapoints: James Enck figures that $75 per annum is a reasonable figure.  After consulting the SkypeOut pricing list, one can see that the per-minute rates for the U.S., France, Germany, Italy or the U.K. are 2.3 cents US$ per minute.  So the $75 translates to roughly 272 MOU per month, which might be a bit optimistic (granted that the assumption there is that most users make calls to cheap locations instead of India (18.5 cents/min.), a mobile phone in Yugoslavia (28 cents/min.), Vietnam (35.8 cents/min.) or a U.K. wireless number (27.5 cents/min.).  I would suggest 3 hours per month at the "low" rates is a more conservative assumption (want to err on the low rather that high side).  That means $4.14 per user per month or an annual ARPU of $49.68 (let's just round it up to $50 for convenience).  Of course, this can be done in a more scientific way factoring in the top countries by users (Rodrigo Sepulveda Schulz has a list here), but the numbers wouldn't be that far off (Poland has a charge of 3.5 cents/min. and Taiwan 2.9 cents/min.). 

Now, fast forwarding a bit to the short term future, it would not be unreasonable to see Skype get 25% penetration of 200 million broadband subs by the end of this year (about 50 million subs).  Assuming that the take rate for SkypeOut grows to 5%, that yields 2.5 million paying customers by the end of 2005, at an annual ARPU of $50 resulting $125 million in revenues for SkypeOut.  In Niklas' presentation, he mentioned a SkypeIn annual fee of 30 € per user, which translates to roughly $39.10 (let's just use $40 for simplicity).  With a take rate of 2.5% (assume half of SkypeOut due to the fact that it is a newer service), this means 1.25 million customers generating revenues of $50 million for SkypeIn, or a combined total of $175 million for Skype services (for 2005), not including extras such as SkypeVoicemail or headset partner deals.

That is a pretty amazing result, particularly considering the very low marginal cost per call minute and the near zero customer acquisition cost.  How much will this make the company worth?  Valuation depends on the sensitivity analysis of various DCF scenarios, but regardless of the multiple, with this much revenues, the company should be able to invest in R&D and enhance the product even more.  For kicks and giggles, why not partner with an speech rec vendor and bundle in some speech rec features (that would really make Om Malik's discussion about passing the "mom test" academic)?  Cannot use a simple interface to make a call?  Simply say the number you want to call!  Of course the speech engine does not even need to be that sophisticated (it could have a very small vocabulary, which does not add that much to the cost and/or complexity of the application).  But this is just one of many ideas of what Skype can do with the revenue streams.  Another one certainly is to hire top talent.  One sign of good things to come is the hiring of Microsoft guru Lenn Pryor, who left the Redmond,WA software giant to join Skype - a great coup for Skype.

Of course, despite all this optimism, a certain dose of reality is also needed.  For Skype to succeed, peering agreements will be sine-qua-non, and while Skype has the jump on the competition, more of such agreements are going to be required over time.  Moreover, for SkypeIn, Niklas also highlighted in his presentation that the company will need authorization from telecom authorities around the world to offer local numbers.  Thus far, the only countries with SkypeIn numbers are the U.S., U.K., France, Sweden, Norway, Finland, Denmark, Poland, and Hong Kong (ed. note: later, I asked Niklas about why Canada was not on the list, and he mentioned nothing happening yet, but my suspicion is he will probably wait for the CRTC VoIP decision in May before making a move in that direction).

   more »

After making strategic investments in "advanced technology" areas such as security, storage, wireless LANs (including companies such as Aironet and Airespace), enterprise VoIP, Optical (SONET), and home networking (including companies such as Linksys), you can tack on the newest addition to the list: clustered servers.  Cisco (Nasdaq:CSCO) announced yesterday the acquisition of Topspin for $250 million. 

The Mountain View (CA) company is a provider of so-called server fabric switches, which are designed to connect interconnect servers in a clustered environment and also deliver network and storage connectivity to that server cluster.  Topspin was founded in April 2000, has about 135 employees in Mountain View and Bangalore, India, and will join Cisco's Data Center, Switching and Wireless Technology Group. 

The raison d'etre of this acquisition is the increasing popularity of the clustered server approach.  The idea is to pool IT resources into a single virtual machine that monitors system demand and accordingly tweaks the supply.  Moreover, Topspin may complement Cisco's Ethernet and SAN switches, making Cisco a top player (#1 or #2) in the SAN switch market.  Another Topspin intangible is its relationships with companies such as Dell, HP, IBM, NEC and Sun, all current Topspin partners.

   more »