When a friend from Nortel sent me James Bagnell's story earlier this month detailing the strange circumstances in which Gary Daichendt was let go, it did sound a bit too bizarre for the average reader to believe in (note: the story originally appeared on the Canada.com portal that belongs to the Asper family of newspapers, including the National Post, Ottawa Citizen, Calgary Herald and the Montreal Gazette, however, it is no longer available; that piece claimed that Daichendt pulled a "mission-from-God power play" stunt that did not go very well with the Nortel Board).  But James Bagnell is a bona-fide reporter, earning the acclaim not only from readers but also other fellow colleagues.  So the issue is - did Bagnell's source deliberately feed him with an "embellished" version of what truly happened?  Or was it in fact a true story? 

Either way, one knew that Daichendt would categorically deny the Ottawa Citizen story and give his own account of how he resigned from his post.  And he has done so - in an interview with Globe and Mail reporter Gordon Pitts.  Not surprisingly, Daichendt flatly denied the original story, claiming that the allegation is "so false, it isn't even close".  Interestingly enough, he did acknowledge the fact that he did tell five or six people at Nortel that he and his wife had prayed for guidance before he offered his resignation.  But he said that the comment was part of a broader conversation about prayer involving the other people. "I prayed with my wife; that's a true statement because I am a man of faith."

While we will never really know 100% what truly transpired on that day, one reality is becoming increasingly clear (as per my original post on this story): what works for Cisco may not necessarily work for Nortel.  Mr. Daichendt might have ruffled a few feathers internally by making it public that he was hired to eventually be the next CEO (when the more politically astute thing to do would have been to remain quiet, even if this information was leaked to the press).  The only question is: by proposing a plan that was too radical (given the level of changes, and the suggested timeline, which really was very aggressive according to what I heard), was he really trying to find an excuse to leave?

Nortel might have not been on the list of the big winners of the 21CN RFP, but that does not mean the vendor was completely shut down from earning some other tenders at BT.  Late last year, Nortel won a deal to upgrade the BT Retail Customer Contact Center infrastructure (a US$ 5 million deal to bring 10,000 agents up-to-date with IP-enabled contact center technology, which will make them more productive compared to legacy ACD technology).  Then, in June, Nortel won another tender (valued at US$ 42 million) to be the main supplier to deploy a nationwide communications network that will be used to provide secure managed voice and data services to the UK Ministry of Defence (MOD) and British Armed Forces.

Then, today came the news that BT placed an open-ended order for the Nortel CS 2000 softswitch, enabling the British carrier to provide hosted IP telephony services to large enterprises.  While the value of this IP Centrex win was not disclosed, it does open the door for more future deals (e.g. IP VPNs, security).  The CS 2000 already counts with a long list of customers, including Cable & Wireless (UK), Timico (UK), Sprint and Hong Kong Broadband.

   more »

Recent news from Brazil indicated that the adoption a surprisingly rapid proliferation of VOIP within Brazil.  Abrafix, the Brazilian association for wireline carriers (a non-profit organization that could be more or less though of as a Brazilian version of the Stentor Alliance) published some research on June 20th claiming that 36% of Brazil's outgoing international calls are being made via "unlicensed" operators selling some sort of LD (VoIP toll bypass).  "Unlicensed" in Abrafix lingo means not having the license from Anatel to participate in this sector. 

Estimates indicate that there are roughly 1,9 million VoIP users in Brazil (not including Skype), representing over 3% of the existing line base already making VOIP calls.  This is enough to begin having a measurable impact on traffic, and eventually, on pricing.  Obviously, the biggest pressure will be on the LD market, due to the exorbitant tariffs for LD calls.  However, given the growing number of subscribers making free peer-to-peer local calls, local tariffs will obviously also have to drop (it is about time! - we are so spoiled over here in Canada and in the US that we do not even know what a local impulse is). 

Also worthwhile pointing out that Skype CEO Niklas Zennström mentioned on an interview earlier this year with Valor Economico that Brazil currently is the fourth largest base of Skype users in the world, with roughly 2 million users.  This figure is only surpassed by the US, Taiwan and Poland.  More impressively, the Skype growth in Brazil has been about 7,000 new adds per day, which would translate to roughly 3.3 million users by the end of 2005 (about 6% of the installed fixed line base).  That is a huge result that can have a lot of ramifications - no wonder the telcos have already begun to actively lobby the government to update the General Law of Telecommunications, in order to address the new realities of the marketplace and issues such as convergence and the impact of VoIP.

   more »

It was interesting to see the news on the FT that Telefonica will buy a stake in China Netcom (more precisely, 2.99% for a price tag of €240 million).  While this is still not enough for a board seat (the minimum holding to warrant that is a 5% share), Telefonica is hoping to eventually attain that goal.  The price, according to some Wall Street estimates, came at about a 15% on the fair value of China Netcom - the reasons being the high exposure to broadband growth (given that China Netcom has a near monopoly in local loop, particularly in areas such as Northern China), and the defensive nature of PHS versus wireless (with prices roughly half CDMA).

It is interesting to note that Telefonica is continuing with its strategy of making investments in foreign telcos, despite not being terribly successful thus far in some of its other plays (e.g. Cesky Telecom) and also despite the lackluster performance of other euro players in Asia (e.g. Vodafone's acquisition of a 3% stake in China Mobile has proven to be unfruitful thus far). 

Hopefully, Telefonica's persistence will pay off, and show the way to astutely invest abroad for North American service providers, which have typically have retreated after mixed results investing in regions such as Latin America (e.g. Bell South, MCI, etc.). 

   more »

Ben Charny had a great post on the backlash of VoIP in Germany. Vodafone Germany made it known to German telecom regulators (during a tariff filing) that it plans to disable calls from Skype and other Internet-based VoIP operators.  Obviously, wireless operators are feeling threatened by VoIP - that is pretty much a similar reaction to their wireline counterparts.  However, with the advent of seamless roaming ,VoIP running on a flat-fee based WLAN model and a pretty decent QoS (using CODECs such as GIPS) can certainly be a big problem to wireless operators.  While the article does state that "Vodafone's other 15 divisions serving nations in Europe and Asia have not enacted such a policy", French operator SFR made a similar announcement back in March (i.e. that it will block VoIP and peer-to-peer streaming traffic).

I wonder what legal recourse Skype would have against this, but these are yet another couple of examples representing proofpoints that whoever owns the last mile typically has a huge turf advantage.  And even companies such as Vonage can be quite vulnerable.  OK, so Madison River Communications was caught in the act of port blocking and got slapped on the wrist by the FCC.  But that was sloppy on Madison River's part.  What if instead of port blocking, it would have just put on additional traffic, or added some randomly generated noise / delays / jitter / etc.?  Such an action would be much harder for Vonage or any other VoIP SP to prove, and hence would create a big barrier for a wider adoption of VoIP.

   more »

Mark Evans had ample coverage of the Nortel AGM, including his post-mortem, where he credited Nortel's brass for managing very well during the Q&A crossfire.  I could not attend the event due to a conflict, however, had I been at the Toronto Congress Center, here is the question I would have directed straight at the Board of Directors: "Why are you taking Toronto out of Nortel's R&D equation?" 

I understand 100% what the CEO/Admiral Owens meant when he talked about Nortel's Brampton HQ.  It is a very large facility that was once one of the main manufacturing plants of the company, but now is the base to only roughly 1,000 management, HR, sales and various administrative personnel.  From a strict accounting perspective, it almost represents the proverbial "white elephant" that despite being a great facility, with a fabulous work-out gym, cafeteria, visitor center, etc.  I understand that the company has been trying to sublet some of the vast unused space, without success, and the area taken up by this space is quite above and beyond its current needs.

But is moving all these operations and consolidating everything from the Brampton HQ to Ottawa the right thing to do?  That will effectively kill any future potential for Nortel to ever recruit young bright engineers in what is Canada's largest market: the GTA (Greater Toronto Area).  Some of the brightest engineers graduate every year from institutions such as the University of Toronto (Nortel collaborates with U of T on a lab and some joint research projects) and the University of Waterloo (which granted, is not in Toronto, but pretty close).  Throw in a few other institutions that are also within the area (York, Ryerson, McMaster, etc.) and then one realizes the great recruiting potential of Toronto. 

Moving whatever small pieces of R&D (probably more like a very little R and a very big D), PLM (Product Line Management) and product marketing that are left here in Toronto to Ottawa makes no sense whatsoever.  Many brilliant folks refuse to go work in Nation's Capital (famous for its brutal winters) and instead choose to remain here, where they can work at other established vendors such as Avaya, Cisco, IBM, RIM or Siemens, service providers such as Allstream, Bell Canada or Telus, or even startups such as Newstep, etc.  In fact, many Nortel competitors, realizing the great potential of the GTA's graduates are expanding their operations here (such as Siemens, for instance, which picked up a couple of startups within the GTA over the past few years: Trango - a contact center vendor - and Chantry Networks - a WLAN vendor).  Now, does it make sense for one of our nation's greatest R&D spenders to completely shut down its operations in Canada's largest city?  I would say no - au contraire.  Closing Brampton might be necessary from a cost saving perspective, but there is plenty of cheap office real estate to be had/rented (such as for instance, the old BNR location in downtown Toronto, at 522 University Avenue, where Nortel once had a sweetheart of a deal).  Move some of the bureaucratic functions back to Ottawa, but keep and grow the R&D in Toronto.  Hopefully the new directors will hear this message loud and clear.

   more »

Many folks trying to predict the uptake of 3G in regions such as Europe and North America are looking at the Japanese and Asian experiences in order to obtain important insights into how things will develop in their own turf.  However, wireless telecommunication is all about the subscribers' personal choices, with cultural differences definitely playing a key factor.  Therefore, extrapolating the Asia/Pacific 3G experience thus far to another continent is definitely a risky proposition.

Nevertheless, even considering this caveat, there is definitely potential for some takeaway lessons from the Asian development up to now.  One observation that is applicable elsewhere is that 3G rarely changes the popularity of apps - chances are that the same apps that were popular in the 2G world will be the most widely used in 3G as well, with the main difference just being faster download speeds. 

There is much debate on the issue of the uptake of video telephony.  In Europe, for instance, it is regarded as an important component of 3G offers, with operators such as 3 or Vodafone having it as part of their portfolios.  In Japan, despite the fact that it was one of the cornerstones of the DoCoMo 3G portfolio, the uptake has been pretty disappointing thus far.  Culture might play a role here - chances are Europeans are less intimidated by video calls than people in Asia (a similar phenomenon to VM), at least when the call is informal.

Currently, there has been no Asian "killer app" on the data side - albeit data usage is increasing.  News, entertainment, ring tones, sound clips, games and screensavers all compete for the title of most popular non-voice app, but there is no current clear winner as of yet.  This can certainly be the case in other geographies, until some more refined applications emerge. 

So these are a few "uber" conclusions that can be drawn from the Asian experience - 3G still represents a technology in flux, with much growth to come in the years ahead.

   more »

It was really interesting to see how many people jumped on the CRTC bandwagon, praising its VoIP decision as being a savvy one, and one that really promotes competition by taking ILECs such as Bell Canada and Telus out of the game (at least on their own home turf – and let’s be honest about it: many folks seriously doubt that we will see those players actively promoting and campaigning VoIP offerings outside their incumbent territories).

Amazing how none of these so-called “expert” analysts (with a few exceptions, of course) picked up on the fact that even under a full deregulation VoIP scenario, the ILECs would have faced margin pressures, and would have had to reduce prices to maintain their market standings and stay competitive. As Michael Sabia himself declared in a Globe and Mail interview a few weeks ago, the telcos would need to cannibalize their own base to keep their market shares. But the consumer would have been the winner in that case, since a new lower price equilibrium would have been reached. Competition and lower pricing is what drives progress, and Japan is a great example of that (check out this TF entry about Yahoo! BB).

However, by preemptively taking the ILECs out of the VoIP game, the CRTC pretty much assured that the existing status-quo on pricing would continue (there will be some competition, but price drops will be small compared to what they would have otherwise been. Let’s take a look at the MSO VoIP launches thus far: most cablecos have thus far been pretty “rational” and “disciplined” in their pricing (e.g. Shaw’s $55 / month offering). The lone exception is Videotron, but the fact of that matter is that the MSOs will balance profitability and volume. So it is very unlikely for us to see a major price battle happen any time soon.

There is some hope that to stem the loss in local market share, Bell and Telus might resort to offering some discounts in Internet, video (Bell Canada’s ExpressVu service) and wireless segments, or perhaps offer some bundling savings. However, thus far, Bell announced that it was taking away its $5 offer for 1,000 LD monthly minutes anywhere in North America for customers subscribing to the Bell bundle (local, wireless, Internet and satellite service).

In the meantime, the incumbents are appealing the decision to the federal cabinet, a process that can take at the very least 6 to 9 months, particularly considering the current turbulence in Ottawa.  The fact that there were two dissenting opinions should provide the ILECs with some valuable ammunition during the appeals process. 

On a separate note, Industry Canada recently formed a 3-member panel to review the Canadian telecommunications policy. This initiative is essentially a broad review and is not expected to affect current CRTC proceedings. Chances are that the panel will probably zero in on less sensitive matters like broadband access and deployment, rather than VoIP and/or competitive issues, and will provide its recommendations to Industry Canada before the end of the year. The findings may have implications on the role of the CRTC in telecom regulations in the future.

   more »

Here was a great find - it turns out that Brough Turner (the CTO of NMS) is also a blogger and he recently posted a great insight on IMS.  In fact, based on discussions I have been having with various wireless operators around the world, there has been a bit of skepticism on the actual value proposition of replacing mobile infrastructure based on circuit switching with packed switched gear - the extra bandwidth, the lack of QoS (which will come, albeit in 3GPP R6), and the handset availability are all issues that will take time.  In the meantime, most services will be based on streaming of audio and/or video, location, gaming and presence.  Instead of focusing on OPEX savings and cost justification, the early adopters of IMS will revamp their networks due to strategic purposes, and not because some ROI benefits.

But one quote of Turner's points that I really enjoyed was his comment #7, namely:

SIP (with SDP, RTP, etc.) was originally a solution for peer-to-peer communications sessions, notably peer-to-peer telephony.  Subsequently, various IETF working groups have extended the SIP family of specifications to support every kind of telephony that has ever existed, specifically including central control of dumb devices.  It's difficult to see how SIP proponents can complain that 3GPP has distorted the original vision of SIP (as a peer-to-peer protocol) when the SIP community has already done the same thing.  It sounds like the real issue is "who is in charge?"

Interestingly enough, this is happening a lot with IMS.  SIP is about smart endpoints, P2P, etc.  IMS provides for a centralized call session control (the famous CSCF, also dubbed the "IMS server"), a rich centralized database (HSS), and the separation of the transport, control and application layers.  Can the two go hand-in-hand?  Yes, but in between, as we might expect, there will be a lot of "proprietary" versions of SIP - Lucent, for instance, has its SIP-Assist technology that offers a greater capacity (by 40%) on its CSCF (Lucent Session Manager or SM) via some sophisticated "parsing and messaging techniques".  SIP purists will have a field day with this - what would someone like Henry Sinnreich say?  On the other hand, the vendors will always try to build a competitive advantage, and that is to be expected.

   more »

Brazilian telecom regulator Anatel announced appointments to its new organizational structure earlier in June, signaling yet another step in its quest for restructuring.  The Brazilian national telecom agency decide to reshuffle its deck following a process rather than service based organization.  This means no special treatment of public versus private services under the new regime, in contrast to the modus operandi of the old system, which was geared towards protecting the wireline voice services.  This would tend to indicate a much lighter approach to regulation, enabling mergers that can form full-service national competitors.  The real competition would then be among these nationwide new organizations - with companies such as Telesp, Brasil Telecom and Embratel expected to take advantage of the change in rules.

On a separate note, Anatel is also trying to hold its 3G license auction as quickly as possible (perhaps even this year).  Despite some skepticism, I do believe that Brazil would be able to attract a new entrant (like Hutchison or Vodafone).

   more »

Rogers announced today the launch of its Home Phone telephony service in the GTA (Greater Toronto Area) starting July 1st.  The much anticipated launch announcement did not bring in too many surprises, as Rogers Home Phone will apparently follow the pricing of Sprint Canada's local and LD offerings.  In order to give customers an incentive to bundle, customers who sign for the telephony service will receive a 15% discount, albeit that will come at a price (a two-year contract). 

The discount over Bell Canada will depend on the service option and the bundling discount.  For instance, for a basic line plus 3 services (from a list that includes voicemail, call display, call waiting, call forward, call return, 3-way calling, call screen, call transfer, and speed call) the Rogers price will be $37.95 without the bundle (a 14% discount over Bell's $44 rate) or $32.26 with the bundle (a 27% discount over Bell's $44 rate).  However, for the standard plan with one feature and no bundling, the Rogers discount is only 3%.  This is in line with my expectations - in other words, "rational pricing".  This is economics 101 - there is no major incentive for Rogers to really drop the pricing too much (a la Videotron), as that would not necessarily be profit-maximizing, particularly considering the fact that Rogers also can parlay its vast wireless infrastructure.

What can one expect for uptake?  Perhaps a conservative estimate would call for 25k subs by year end and 10x that (or a quarter million lines) by YE 2006.  But a lot of that depends on features, initial system performance / customer satisfaction and what Bell Canada might do to counteract this expected move by Rogers.  So how does Rogers pricing stack up against its cable peers?  Not as attractive as Videotron, but certainly better than either Cogeco or Shaw.

   more »

Earlier this spring, I met with Sir Terry Matthews at a Mitel Analyst event here in Toronto and we briefly spoke about one of his investments (Solace Systems), which is focused on application aware networking (i.e. XML routing).  Having heard about Solace's deployment at Allstream, I was already familiar with the company, but his enthuasism was so contagious that I had to go and check the company out myself at Supercomm.

And I came out impressed with seeing some real live demos of application-aware networking, and the notion of being able to classify, filter and parse massive amounts of incoming data as it flows through an enterprise's network.  XML will be an enabler for powerful machine-to-machine interactions in the future.  There are several protocols that already rely on XML to handle and manipulate data streams, including, among others, SAML, SOAP, UDDI, XML-RPC, etc.  There is an obvious need to have some gear that can manipulate, parse and transform XML at network speeds.  We are talking about a potentially destructive technology that can profoundly alter the landscape of the Layer 4-7 routing market. 

The key is achieving this goal and stay within the key CIO goals of not increasing performance bottlenecks or security vulnerabilities.  Market timing is important, and as usual, Mr. Matthews has correctly anticipated the need for XML processing infrastructure.  Hence, it was not surprising to see companies such as Solace and MA-based DataPower get the R&D jump on a few other competitors.  One of the questions on every analyst's mind, though, always was that if this technology was so great and potentially disruptive, how come none of the technology giants made too much noise about it thus far?  Other than Solace and DataPower, only a handful of other vendors had been active in the application aware networking market - including Ciena and NetScaler. 

That is no longer the case.  After a few rumors on the press (most noticeably on this insightful article), Cisco has finally come out with more details on its AON (Application-Oriented Networking) initiative during its Networkers 2005 customer and partner conference in Las Vegas.  This will be Cisco's seventh AT (Advanced Technology) growth market (there are 12 ATs in total, with 6 of them already announced and 6 yet to be announced). Enterprises having Cisco's Catalyst 6500 Series switches or 2600, 2800, 3700, and 3800 Series routers will not have to replace their existing infrastructure - all that will be necessary is for them to plug in AON blades when they will be generally available.  During the event, there was also talk about a likely AON wireless and RFID module for the ISR router, the availability of some management software, and the announcement of some technical relationships with several companies including IBM, SAP and Tibco.

   more »

SMBWorld, a multi-city event put together by Jake Gordon & Associates, had its final day in Toronto last week.  The show was pretty interesting in that it catered specifically to the Canadian small and medium business segment.  I always felt that there was a need to fill in a conference void in that space, and SMBWorld certainly achieved that, presenting a good view on market landscape. 

One of the highlights was hearing ex-Raptor GM, Glenn Grunwald speak about his current responsibilities as the President & CEO of the Toronto Board of Trade.  Glenn mentioned his current efforts in helping new immigrants get integrated within the Canadian workforce (i.e. here in Toronto).  I asked Glenn what is currently being done at the municipal level ot help encourage more telecom investment in Toronto.  He mentioned that much more effort needs to be spent in that direction, above and beyond the Toronto Board of Trade.  I couldn't agree more:  maybe the time has come for the municipal and/or provincial government to give some tax incentives to lure more high-tech to our city.

   more »

Robert Liu from TMCnet posted an interesting story on some research done by Deutsche Bank on Consumer VoIP.  The numbers this year also are showing strength, with VoIP residential subs growing from 1-1.2 million (at the end of 2004) to 1.6-1.8 million (at the end of Q1 2005).  Interestingly enough, the report also mentions the advent of IMS and the prospects of wireless VoIP.    more »

India is ripe for a major broadband expansion, to be followed by a wireless broadband uptake.  Currently, there are 183 ISPs that are operational, with 388 licenses having been signed.  There has been a recent easing in regulation that has as a result lowered the market entry barriers, allowing infrastructure sharing among distinct service providers.  The idea is to enable cost reduction via optimal utilization.

VSNL (Tata's Videsh Sanchar Nigam Ltd.), Reliance Infocomm, and the state-owned BSNL (Bharat Sanchar Nigam Ltd.) have been testing WiMax services since the start of this year.  They are regarding WiMax as a last mile wireless connectivity solution to be used particularly in rural and/or remote regions.  If things go as expected, WiMax will provide relief for these operators, as they struggle to meet their rural telephony mandates (there is a huge gap between teledensities in urban regions - about 22% versus rural areas - under 3%).

Also in the WiMax front, Bharti has been already running market trials in Bangalore, Bhopal and New Delhi.  Thus far, the results have been good and the company is pretty bullish on the service.

   more »

Another key takeaway from Supercomm last week was the fact that the FCC (at least for now), despite its recent E-911 decision, will attempt to keep its VoIP-friendly posture.  It was good to hear new FCC Chairman Kevin Martin repeatedly express his belief that cable and phone companies should compete on a level playing field.  Therefore, MSOs have been able to enter the voice business world with few regulatory barriers.  With the phone companies battling local municipalities over the need to obtain franchises to offer video service, Congress will likely get a chance to review the current federal communications law.  Regardless of the uncertainity of future regulation reform, the one sure bet is that all players will increase their network spend in the future.  Anyone at the CRTC listening?   more »

The big news today, as reported by Om Malik: Yahoo! got serious about adding VoIP to its portfolio and acquired Dialpad, the well-known VoIP operator.  Almost everyone I know used Dialpad for free calls to the US at some point circa 1999 (this was due to the agreement that it had in place with GTE, to put some traffic on GTE's network).  Then, as usage went up, calls were being cut-off after three minutes (or was it a minute?), and then Dialpad became a paid service.  Granted the uptake might have not been as phenomenal as with Skype (and the freebie calls were only for the US), but Dialpad certainly had a huge base of customers, many of which were lost as the company transitioned towards prepaid services.  Fast forward to 2005, and now Yahoo! sees potential to make inroads in the telecom space and offer new services such as PC to phone and inbound calls.  Interestingly enough, Yahoo! will integrate Dialpad functionality into its IM client, which is the mirror image of what Skype did (i.e. add IM on top of a softphone platform).

Yahoo! gets a quick entry to VoIP services on its own terms, rather than relying upon the services of other companies (such as Yahoo's own past use of Net2Phone).  Moreover, Dialpad also has a lot of billing/OSS expertise, in addition to fraud management detection. 

So the million dollar question is: how will MSN and/or Google react?  Can this open the door to some new acquisitions of VoIP SPs?  What about Skype?  Mr. Zennström (the CEO) certainly does not appear to be in a big rush to sell, but will the VCs and other Skype execs consider a sale at an attractive valuation?  We will soon find out...

   more »

Nortel bashers had a field day last Friday, as they took the stock on a roller coaster ride (down, for the most part) on a negative sentiment due to the resignation of recently-hired COO Gary Daichendt (followed by the departure of his handpicked CTO and ex-Cisco colleague, Gary Kunis - likely a solidarity move).  As usual, there has been a lot of speculation (including today's interesting story on the Globe - hat tip: Mark Evans) on the reasons that led the two Garys to leave.

So why so much turmoil?  Well, Mr. Daichendt came with high accolades since he ran the worldwide sales organization at Cisco during the period of phenomenal growth in the company back in the 1990s.  No doubt about it, Gary's credentials were unquestionable, as he not only ran sales, but also played an instrumental role in shaping up Cisco's manufacturing.  However, as the Globe article states, Mr. Daichendt might have committed the ultimate faux pas right after being hired, making public his plans to eventually become Nortel's CEO.  Granted that might not have gone over very well with the hierarchical culture instilled by a military officer such as CEO Bill Owens. 

But from what I have heard (from some insider Nortel sources), it was the aggressiveness of Daichendt's plan that ultimately sealed his fate.  Apparently, it was the proverbial CLM (Career Limiting Move) - not only audacious in strategic terms, but also perhaps a bit unrealistic in terms of timeline (and the hint was that apparently the board agreed more or less with about 80 percent of what was proposed, but just could not come to terms with the overly ambitious time milestones).  Perhaps an interesting question is whether or not Mr. Daichendt egregiously established such an aggressive restructuring plan, being fully aware of the consequences, just to gauge how much support he would receive from the Nortel board, or to assess how long it would take him to become the CEO.  In case he would not get the board's buy-in, then he would have the excuse needed to leave the scene. 

That has been the subject of a lot of speculation around the Internet, but regardless of the motives, the truth is that perhaps Daichendt might have not been the best suited exec for the job, and in all fairness, perhaps way too much was expected of him.  Remember this is a gentleman that was out of the telecom business since late 2000, and while he was a great exec in his previous life, what works for Cisco does not necessarily work for Nortel.  Nortel right now is seeking a turn-around, emphasizing operating income, profit and strategy rationalization.  An aggressive strategic mind whose previous modus operandi was shaped in a company set to always be the number one or number two in any market it competes might not quite be the best fit.

So why was Gary hired in the first place?  Well, the mystery remains, but regardless of the soundness of that decision, and all the negativity of Bay and Wall Street, the fact is that during his short stay at Nortel, Daichendt's mandate was to explore the organization and come up with a plan.  Three months into his mission, Gary did indeed propose a plan that proved to be a bit too much ambitious.  So he is gone after a short stint, but the flip side is that perhaps it is better that his departure happened sooner rather than later given the big gap in culture that existed.  Also, chances are pretty good that 90 days was hardly enough time for him to begin implementing his own ideas.

As far as Kunis' departure, undoubtedly, he is a sharp technology strategist who will be missed.  The board has placed this position under review (given that Brian McFadden, the previous Nortel CTO is now Chief Research Officer.  My suggestion for that particular CTO role is to promote from within, and my vote would be for Phil Edholm, who is currently the CTO and VP of Network Architecture for the Enterprise Networks Division.  And the search for the new COO goes on...

   more »

So Supercomm wound down on Thursday, and here are a few takeaways from this year's edition of the show:

1- IPTV: will be an important driver in the routing market, since the deployment of IPTV will entail a substantial upgrade of the existing routing network infrastructure.  But I doubt that the time frame will be 2 years, as often mentioned by players such as Microsoft and Alcatel.  The resolution of issues such as scalability, maturity of silicon, Microsoft software support, and regulation will take time.  So it might be more a five to eight year horizon for the deployment and actual full blown mass adoption of these services.

2- Chinese are here to stay: vendors such as UT Starcom, ZTE and Huawei had big booths there and there was even a Chinese telecom market mini-show.  Although they are still seeking a greater market penetration on this side of the ocean, they are a force to be reckoned with in wireline (NGN, DSL, triple play) and wireless (CDMA, W-CDMA, CDMA 450, TD-SCDMA). 

3- Blades are also becoming a mainstay - both Intel (ATCA) and IBM (BladeCenter) showcased their blade platforms.  The notion of reusing the chassis and having modules that can be incrementally added for a variety of different applications is gaining popularity.  Intel and IBM are both vying to have their products become the standard platform for carrier-grade applications.

4- IMS was again one of the buzzwords most often heard at the show.  Service providers such as Bell Canada, SBC, Bell South and Comcast are all investigating this framework and evaluating future deployment strategies.  While there were no specific plans / milestones mentioned, there certainly was some optimism vis-a-vis IMS, also tempered with a good dose of realism.

   more »

Last night, the Ditech dinner revealed a bit more about its acquisition of Jasomi Networks, yet another SBC (Session Border Controller) company to be acquired recently (following Kagoor, which got snapped up by Juniper).  The final price tag was $20 million ($13 million in cash and $7 million in convertible notes), which might seem a bit underpriced compared to Kagoor (which fetched $67.5 million).  On the Wednesday analyst/press dinner event, the answer given to me about this difference in valuation was pretty much the same as what was quoted on a LightReading article, namely, that Kagoor was funded by VCs, whereas Jasomi got most of its capital from angel investors.  The transaction is expected to close by the end of the month.

The rationale behind the acquisition is to complement the PVP Packet Voice Processor product (an offering that was talked about at the Spring VON show).  The PVP is expected to begin generating revenues in calendar Q1 of 2006.  One idea is to develop the PVP into an SBC, or even perhaps as a media gateway.  In order to shorten the time-to-market, the company chose to purchase rather than to develop such an offering in-house.  Jasomi's 23 employees are mainly based in Canada (the HQ for product development is in Calgary), with a few others working out of Mountain View, CA. 

Another possibility is parlaying Jasomi's SBC competency in the IMS space, albeit thus far Jasomi's biggest customers have been in the enterprise market (versus carriers) - can that be scaled up or is more development needed?  There were rumors from my Wall Street connections that sometime ago, Cisco had tested Jasomi gear in Singapore and passed up on the opportunity, but a good chunk of time has passed since then.  Jasomi's current partners include the likes of Microsoft, Broadsoft, Sylantro, RADVision, and General Bandwidth, among others.

The million dollar question now is: which will be the next SBC vendor to be bought?  I will be willing to make a couple of bets: one is Nextone (since au contraire of vendors such as Acme Packet or Netrake, Nextone's solution can be put on a blade - and hence be a part of the IBM Blade Center / IMS ecosystem).  The other one?  Ditech itself - since the company could be positioning itself to be bought at some point in the future by an even bigger player.

---

Note: Jeff Pulver was a guest speaker at the Ditech dinner and gave a great keynote.  I have posted a couple of photos of the event in my photo album.

   more »

Nortel execs take note: Siemens joins other vendors in exiting markets in which the company underperforms (other recent examples: Lucent, exiting the GSM business, or Ericsson, which left the CDMA market for the most part, except for emerging opportunities such as CDMA450).  In Siemens' example, the Munich-based multi-national finally unloaded its wireless handset unit to Taiwanese vendor BenQ.  As reported here before, Siemens' quest to unload its underperforming mobile phone division has been going on for a long time.  The transaction, expected to close in September 2005, calls for Siemens to purchase $61.4 million (€ 50 million) of BenQ stock, in addition to taking a $429.5 million (€ 350 million) charge.  The division's workforce (totaling roughly 6,000) will remain in Munich.  Siemens and BenQ combined in 2004 to amass about 10% of the overall global market (4th in the world, behind Samsung).  Geographically speaking, the deal makes sense, with BenQ being strong in APAC and Siemens showing strength in Europe and Latin America (including Brazil). 

   more »

Tuesday and Wednesday also brought a flurry of new announcements at Supercomm, including, among others:

• Adtran unveiled the Total Access 1100F, which delivers IPTV and deep fiber connectivity for its Total Access OSP and Mini-DSLAMs.  The company also announced the next version of the OPTI-6100 Optical Multiplexer, which comes with enhanced Ethernet service capability and SDH for foreign markets.
• Alcatel made public its UMA (Unlicensed Mobile Access) solution, which delivers cellular/WiFi convergence.  This offering works by providing seamless roaming for subscribers in GSM (and GPRS) networks and 802.11 (and Bluetooth) unlicensed bands.
• Foundry announced the NetIron XMR(TM) series family of Internet and Metro MPLS Routers.  These routers (IPv4, IPv6, MPLS) routers are geared towards delivering triple play services, converged backbone and broadband VPN.
• Lucent introduced new features for its Stinger product line (i.e. DSL Access Concentrators that will empower SPs to deliver triple play services to their customers.
• Two Nortel partners (Calix and Keymile) debuted new voice gateway functionality to their broadband access products that make them compatible with Nortel's VoIP offerings.  Both products have been tested with the Nortel softswitches.
• Sonus announced its own IMS solution, and a lunchtime presentation revealed some detail - the IMX App Server will serve as the service broker (SCIM) component, and also store data in the HSS.  The company will also have its own ASX Feature Server, HSX HSS and PSX BGCF.  Sonus will count with its own IMS partner ecosystem, including the likes of IP Unity, Sylantro, Genesis, IPeria and BayPackets, among others.

   more »

Lucent is going full steam ahead with its plan to amalgamate its wireless and wireless LOBs.  After taking that decision back in April, Lucent is currently undertaking a 4 month period to analyze its entire portfolio of products and resources.  The vendor is promoting its capabilities in wireline (VoIP, optical) and wireless (CDMA, 3G) as key components of IMS.  Relationships with players such as Broadsoft will continue to be important as Lucent builds its IMS partner ecosystem.  During a breakfast meeting with analysts, the company also stressed its strength in the services business (LWS) and its relationship with Cap Gemini.  On the wireless front, Lucent will try to get a piece of the Ericsson market share, as the Swedish vendor exits the CDMA market (albeit Ericsson will still continue to focus on emerging CDMA market opportunities such as CDMA450).  There have been some hints from Wall Street that Lucent might get an EV-DO contract next week.  Last but not least, I also sat a bit on the Lucent Labs table and felt a great dose of optimism coming out of that organization, particularly with the return of Dr. Jeong Kim as the President.

   more »

The two IEC IMS events were pretty interesting (the first one, a full day IMS session on Monday - where I gave a talk - and the second one, an hour-and-a-half long wireless IMS perspective on Tuesday). 

The first day featured the perspectives of SBC (David Deas) and Comcast (Jason Livingood).  David's presentation focused on wireline/wireless convergence and brought up a few key points, including cellular/WiFi integration, and the big divide between wireline and wireless service providers. David mentioned that wireline carriers overwhelmingly believe in FMC (Fixed Mobile Convergence), whereas wireless operators are quite skeptical.  The latter part of his talk was also interesting, as he considered implementation issues (such as Parlay/OSA integration options).  Jason offered a very insightful perspective on the MSO view of IMS, including the ongoing standardization efforts and the issues faced by most cable companies in considering their strategies.  He also mentioned a few IMS applications that could be of interest to MSOs, and the ongoing Cable Labs activities.

 The second day featured the perspectives of Bell Canada (Philippe Jetté) and Bell South (Bill Smith).  Phillipe's presentation emphasized the evolution towards the seamless access in the world of IP (wireline/wireless and nomadic/fixed). IMS will play a key enabling role in the transformation from a single bill to a single network and to a single (user-centric) seamless application space, according to Philippe. Bill's presentation highlighted the IMS view not only from Bell South's, but also from Cingular's perspective.  He emphasized some of the consumer IMS apps, including managing a family's communications, seamless roaming, etc.

Overall, the consensus throughout the two days was that while IMS still is very much a work in progress, it will play a key role in the transformation towards a user-centric world in which applications can be delivered quicker and more efficiently to certain segments in the subscriber bases.  All in all, the IEC put together a couple of great sessions that provided a good current perspective in the world of IMS.

Note: I posted a few photos from my day on Monday (including the IEC session and an evening at Wrigley Field). 

   more »

The Supercomm Nortel update included the revelation that the MPE9000 switch/router that was announced last year will finally hit GA by the end of June.  The MPE is expected to be a "big product" (according to VP of Carrier Marketing Jim Dondero) for Nortel within six months, as wireline and wireless operators will begin to rely on it to aggregate ATM, Ethernet, Frame Relay and IP at the edge of their networks.  Nortel has been quoting some pretty impressive VoIP numbers: 55 carrier VoIP customers and 20 out of 35 MSOs that have made VoIP decisions up to now.

   more »

Blogging has been a bit light the past few days, as I have been caught up in a lot of work, office move, and a trip to Supercomm.  Speaking of which, a lot of announcements were made during the first day of the show, including, among others:

• Cisco, announcing its DDoS (Distributed Denial of Service) protection offering, which is now a part of the company's NGN IP architecture.
• Juniper giving demos of its newest software upgrade (5.0) to its Secure/Remote Access SSL VPN appliances; the company also unveiled its new E320 broadband services router, the industry's largest capacity broadband router.  The E320 is a pretty good platform for delivery of services such as IPT, VOD, and IPTV.
• ADC, Lucent, Nortel and Tellabs all revealed their new products targeting the emerging FTTX / "triple play" markets.  Lucent announced a collaboration with Polycom geared towards delivering IP voice, video and collaboration solutions.  Nortel introduced next-gen features for its optical network solutions.  These features will allow service providers to deliver high bandwidth content and application-based services in a much faster fashion.
• Sonus unveiled its GSX4000 compact gateway which caters to the lower end of the VoIP market.
• On the M&A front, Ditech announced it was acquiring Jasomi, in a marriage of SBCs and voice processing technologies (I will attend Ditech's dinner on Wednesday evening and learn more about synergies, and the rationale behind the purchase).

P.S. I will have a separate posting about IMS and will also share a few recent pics taken in Chicago with my camera.   more »

This was the claim made by Sir Terry Matthews, as Ubiquity Software became the latest startup in his universe of high-tech investments to go public in the London AIM (Alternative Investment Market) (ed. note: prior to Ubiquity, another one of Terry's companies, March Networks, also went public).  Ubiquity, which raised about $47 million in its IPO, had about $15 million worth of sales.  Sir Matthews told a group of technology executives in Ottawa late last month that the London AIM has a key tax advantage over other exchanges such as the NYSE and Nasdaq.  By holding on to a startup investment for three years, there is no tax on the gain for a company listed on AIM. 

Separate Note: Interestingly enough, Sir Terry admitted for the first time that if someone made a "big enough offer" for Mitel he would consider selling the company.

 

   more »

IPTV has been getting quite a lot of bandwidth lately, and there have been a lot of stories such as the ongoing SBC effort and the Alcatel/Microsoft alliance, among others.  However, despite all the hype, here's a reality check on the current status of the technology: last week, it was announced that Swisscom delayed its own IPTV launch.  IPTV is turning out to be quite more complex than previously thought; one of the technological challenges is scalability, and it still remains to be proven that IPTV networks can scale.  Another potential headache is regulation, as there are some regulatory risks that can potentially impede the ILEC rollout of TV services.  So in the next two years, expect there to be a lot of access network buildouts, with initial deployments (in small scale) only really beginning to ramp up in 2007.  Therefore, a wider consumer adoption of IPTV services probably will not happen before 2008-2009.

   more »

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 »