Scientists have been able to capture some pretty unique images of a radiation belt inside the rings of Saturn.  They have been able to obtain one of the sharpest pictures of the planet's giant magnetosphere, and to also study in more detail the planet's rings.  Some of those images have revealed the presence of some faint rings, which may possibly indicate some previously unseen moons.

The Cassini-Huygens mission is a joint project that counts with the efforts of NASA, the European Space Agency and the Italian Space Agency.  Caltech's JPL (Jet Propulsion Laboratory) manages the mission for NASA and also designed the Cassini orbiter and its two onboard cameras.  The Cassini spacecraft entered Saturn's orbit in July of last year, kicking off a four year study of the planet.   

   more »

A growing proportion of cellular phone shipments are replacements (last year, almost 60 percent).  So it's no wonder every year, we see new handset models, as vendors try to incorporate new functionality to drive up replacements.  Over this past year, there was added emphasis placed on camera phones, which are great not only for the handset manufacturers themselves, but also for operators because they can hopefully increase data traffic.  So it was no wonder to see the so many models having built-in cameras at 3GSM, including the sleek Motorola SLVR V8 show on the figure (hat tip: Gizmodo).  The model is really compact and features tri-band capabilities (GSM/GPRS/EDGE), 5 Megs of memory, Bluetooth support, color display, push-to-talk, MP3 playback, and an MPEG4 VGA camera.

Camera phones are tracking at about 50% of new handset shipments in Western Europe, and part of the reason for their success is that vendors have been able to keep the costs down (the integrated camera models is about 5% of the total bill of materials for the handset).  One of my own biggest questions from 3GSM was - will users want a cell phone that can take pictures or a camera that can make phone calls?  Honestly, one could almost not tell the two apart.

   more »

Scientific American has an interesting note about breakthrough HIV research being conducted at Harvard Medical School, where Bing Chen and his colleagues have been able to get the first sharp pictures of the gp120 protein, which enables HIV to fool the immune system effort to destroy it, while continuing to infect host cells. 

gp120 is a highly variable protective membrane that allows HIV to evade the attack and survive in the body long term. The research undertaken at Harvard studies how the molecule changes shape once it recognizes and binds to the cell. This is yet another important milestone in understanding how HIV infects a cell and how to stop that from happening.

   more »

Upon hearing that MCI (Nasdaq:MCIP) shareholders had questioned the wisdom of accepting a $6.64 billion offer from Verizon (NYSE:VZ) rather than an initial $8 million bid from Qwest (NYSE:Q), it seems that Qwest executives decided to up their ante and submit a revised $8 million offer with more guarantees and upfront money.  The idea is to offer more cash up front and also protect MCI if Qwest shares drop by as much as 10% before the deal closes. 

The reason that some MCI execs initially were more inclined to accept the Verizon bid (even though it came in lower than Qwest's) was the debt that Qwest is currently burdened with ($17 billion).  The reality is that while Qwest is the incumbent in 14 Western and Midwestern states, it faces a lot of competition for HSI (High Speed Internet) and cable TV services.  The company incurred more debt when building its state-of-the-art national fiber optic network. 

This latest offer will get the full consideration from the MCI board, but predicting its reaction is difficult, considering the public manner in which the bid was made and the turbulence of some of the shareholders.  It could very well mean that MCI will leverage this latest Qwest bid to get more money from Verizon.  But the million dollar question is how much more MCI will be able to fetch from Verizon - some folks in Wall Street are guessing that increase will not be very significant (less than $0.50 per Verizon share). 

   more »

Alcatel (NYSE:ALA) announced a couple of days ago a partnership with Microsoft (Nasdaq:MSFT) geared towards developing an integrated IPTV delivery solution.  The idea is enable MSOs to deploy IPTV faster and cheaper.  This relationship is the first significant one that unites a major triple play solution vendor and one of the top IPTV software vendors.

Interestingly enough, prior to signing this agreement, Microsoft and Alcatel were competing in the IPTV middleware segment (in fact, Telecom Italia was trialing both platforms).  Alcatel had acquired a couple of software houses to create its own IPTV offering (the Open Media Suite).  But Microsoft kept piling win after win (SBC, Verizon, a trial at Bell South, and Swisscom. among others), and Alcatel had to implement Microsoft's IPTV system in deals where it was the primary equipment supplier for DSLAMs and routers (such as SBC, for instance).

So nothing like a partnership to solve the issues.  By jointly developing an IPTV solution with Microsoft, Alcatel saves considerable system integration time (since it can integrate Microsoft's software into its hardware).  Microsoft wins a way to tap into the world's large carriers.   

   more »

Om Malik reports on the ongoing broadband price war in India, where $8 a month can get you a basic 256 kbps service from Bharti Telecom (still nothing compared to the French broadband bonanza that was discussed previously here, but cheaper than Indian incumbents BSNL and MTNL).  The original story appeared on the Times of India, and is worth reading.

The part of Om's blog entry that I found interesting was his comment on how difficult it was to get a phone line in the old days (often entailing bribes).  Well, my experience growing up in Brazil was not much different: getting a line from TELESP in the good old days was very, very difficult, and the wait was enormous (baksheesh in India was better, because at least that is a bit faster - in São Paulo, there were bribes and on top of that, there were middlemen... really, there were "brokers" who got into the game and used to charge extra bribes supposedly for faster turnaround times).  Of course, this was way back in the late 70's-early 80's and many years before Telefonica entered the market.

   more »

Tom Evslin wrote a great insight on the topic of vertical integration and how it is no longer a sound business strategy (hat tip: Fred Wilson from A VC).  It is an excellent article that explains the transition from vertical to horizontal integration.  IBM, Tom argues, was in the older days of computing the typical model of vertical integration: it made the hardware, chips, application software, OS, and even peripherals, while also providing instruction, consulting services and accessories.  However, with the advent of the microcomputer, a paradigm shift took place, with the business changing to a horizontal model: Intel became the microprocessor king, while Microsoft dominated the OS and applications.  IBM had to change itself, and managed to adapt to the new environment thanks to initiatives such as IBM Global Services, although it is not the dominant force it once was.

Tom wonders whether or not SBC will be able to make vertical integration work with its acquisition of AT&T.  Clearly, the vertical strategy failed at AT&T, since it lost the opportunity to be an ISP powerhouse and a key local access player.  Tom makes it clear which strategy is his preferred one:

"A horizontal company has a high surface to volume ratio. It sells all of its outputs in a competitive market. It is free to buy all its inputs in a competitive market. Its managers are not isolated from the markets they compete in. A vertical company spends most of its time and energy dealing with itself rather than the external market. Meetings are dominated by esoterica like transfer pricing between divisions because this is what determines internal success. The real marketplace is distant from most of the managers trapped inside the vertical structure. Buy vs. build and capital allocation decisions inside a vertical company are made by office politics in a vain attempt to optimize across the whole vertical organization. Horizontal competitors optimize only for the layer they are competing in and so end up being superior to the vertically integrated company layer by layer."

Read it all! - This is definitely a great insight.

   more »

The search for growth opportunities is getting to be difficult, since mobile operators in some regions such as Western Europe are staring at nearly 90% penetration rate. In North America, the take rate of cellular phones is in the 50-60% range, so the market is not quite as saturated, but the growth prospects are not as bright as they once were.

Increasingly, wireless service providers are seeking to make investments in emerging markets such as Asia/Pacific (penetration in the mid teens), Caribbean and Latin America (penetration at about 20 percent) or Eastern Europe (penetration moving closer to 40%). Take Norway’s Telenor, for instance, which has been that country’s incumbent for nearly 150 years. Faced with a domestic market that was nearly tapped out, Telenor started taking ownership positions in a few foreign operators in countries such as Bangladesh, Pakistan, Russia and Ukraine.

The Pakistani fully owned subsidiary is expected to be launched in April, and there is optimism about the success of that operation, given the low penetration in the region (less than 5%). In Russia’s case, Telenor only owns a 30% stake at Vimpelcom, but the future is looking bright, considering the nearly 8.5 million net adds registered for the entire Russian market last December, which represents more new customers than what all Chinese operators gained during the same period. These foreign investments catapulted Telenor to the 12th spot in the world, with an installed base of around 50 million subs and roughly 19 million net adds in 2004.

Of course, closer to home Cingular prevailed over Vodafone in the AT&T Wireless stakes, thanks to a last-minute bid that triggered a late night board meeting (the idea being not to give Vodafone a chance to up the ante because of the time zone difference).  But after a few so-so investments in regions such as Brazil, are American operators ready to take the plunge again and try their luck with emerging markets?

   more »

There was a flurry of WiMax events last week, which definitely were positive for a wider uptake of the technology. WiMax got quite a bit of attention at 3GSM, partly because of the high throughputs it can offer (15 Mbps, which is comparable to HSDPA). Despite that, most of the talks focused on WiMax as an alternative for fixed line broadband, emphasizing that the opportunities lie mostly in rural markets (with WiMax being used instead of ADSL buildouts). Here are a few announcements:

- Alcatel and Intel announced that they will jointly develop WiMax products based on the WiMax mobile spec (IEEE 802.16e) in mid-2006.  Alcatel already sells fixed WiMax (802.16d) products from Alvarion.  Alcatel plans to price the technology to consumers somewhere between DSL and 3G.

- T-Mobile introduced WiMax net access for UK trains, relying on the technology as a way to uplink to the Internet.  WiMax allows T-Mobile users with Wi-Fi access to get Internet access on the go by sharing the 32 Mbps WiMax connection, which should be plenty for a large number of concurrent riders. 

- In Malaysia, a number of equipment manufacturers are already trialing handsets that use mobile WiMax and other new wireless technologies, with commercial launches expected within less than year.

- Cisco is paying more attention to WiMax

- Sprint has joined the WiMax Forum

- DirectNet (leading Brazilian ISP) ordered WiMax gear from Alvarion totaling $1 million over the last few months.

Hopefully, some of the WiMax skeptics will start seeing the glass half full rather than half empty

   more »

Cannes certainly is a very charming setting and has gained a lot of acclaim with events such as the famous Cannes Film Festival or the Formula One Grand Prix in nearby Monaco. But the overall consensus was throughout the 3GSM show was that the French Riviera town is a bit stretched to the limit of its current resources to host a conference of this magnitude. So next year, the host city will be Barcelona, although by 2007, the show could be back in Cannes, when the Palais de Congress is expected to be ready (reportedly, it will undergo some renovations geared towards increasing its capacity). Of course, some analysts will miss being briefed on yachts, or conference rooms of classy hotels such as the Majestic or the Carlton Intercontinental.  But the reality is that, given the 35,000 attendants, the current floor space is tight, whereas the infrastructure needs some upgrades or capacity expansions.   more »

One strong indicator of a rebound in the telecom sector has been the wider attendance in the traditional big annual events, and 3GSM is no exception to that trend. This year’s audience has been one of the biggest over the past few years, with quite a few notable developments, including, among others:

- gizmos galore: lots of mobile handsets that I have never seen on the other side of the Atlantic, including the introductions of a variety of new phone models.  There have been a few new models introduced, including Motorola's impressive slim black Razr.

- Microsoft made a full court press with its wireless initiatives in the enterprise and carrier spaces.  The Peabody is a new GSM/GPRS low-cost platform based on the Windows Mobile software.  Although Microsoft finally overtook Palm in the handheld market, the mobile phone market still remains an elusive target.  The company claimed that Microsoft Windows Mobile-based devices are capable of yielding a higher jump in ARPU than its competitors (37% higher).  On the enterprise front, Microsoft secured another ActiveSync licensing agreement, this time with Nokia, which joins Motorola and palmOne.  Microsoft also added Flextronics to its Smartphone ecosystem. 

- the IMS (IP Multimedia Subsystem) architecture is being covered in several sessions, including Convedia co-founder Grant Henderson’s talk today (Convedia is a key IMS enabler, being the de-facto leader in media servers).  IMS is a framework that provides a SIP-based control layer with open interfaces to the transport layer below and the services layer above. This will allow wireline, wireless and cable operators control over services on a per-session basis. Even though the original IMS concept was originally defined as a 3GPP specification, the IETF also is developing the standard, and even CableLabs is incorporating the IMS concept into its planning.

The pace has been quite hectic today, and considering that Monday is supposedly a "slow" day at 3GSM, perhaps this is indicative of many more new product launches and announcements to come.

   more »

Nortel (NYSE:NT) and Research in Motion (Nasdaq:RIMM) signed an agreement under which BlackBerry devices will have added VoIP capabilities via Wi-Fi.  This will require a tight integration between RIM's BlackBerry enterprise platform and Nortel's CSE 5100 multimedia server.  As announced at VoiceCon 2005, the first model to incorporate this feature will be the BlackBerry 7270, albeit it will be a VoIP-only device (to be used not only in enterprises, but also over public WLAN hotsports).  However, the future is really all about seamless roaming, so the market is IMHO ready for a dual mode BlackBerry that could function in 2.5/3G mode or in Wi-Fi/WLAN mode. 

Note: RIM also signed a similar partnership deal with 3Com, which will use its own IP PBX and WLAN switch (licensed from Trapeze).

   more »

Andy Abramson reported on the EU´s recent "light touch" approach in regulating VoIP.  It looks like the folks from Strasbourg are taking a page out of the FCC's book (i.e. the VoIP friendly strategy employed by Chairman Michael Powell).  The key decision driver for the EU is to increase the broadband penetration for all its current members (which currently stands as only 6.5 percent on average for the EU 25).  The idea is to offer new services such as VoIP at more attractive prices by stimulating the competitive environment.  The ruling clearly states in one paragraph:

The Commission favours an EU-wide “light touch” approach to Internet telephony as the best way to encourage competition between internet carriers of telephone traffic and traditional telephone networks. As the market develops, the European Commission intends to ensure, jointly with the national regulators, that throughout the EU, the roll-out of new IP-based services will not be hindered by regulatory hurdles. It is in the interest of Europe’s businesses and citizens that new technologies should be able to flourish and deliver better services at lower cost.

Wouldn't it be nice to see the CRTC follow suit and also establish an equal level playing field in Canada?  Hopefully, if enough regulatory bodies worldwide endorse a similar position, this has a better chance of happening.

   more »

Loïc Le Meur (hat tip: Dragos Novac from Argumente) had an interesting write-up on the Russian entrepreneurial scene, and how doing business there changed over the past few years.  Among other interesting findings, Loïc reports on the vibrant new Moscovite generation (unbelievably, there are more Mercedes 500 sold in Moscow than in New York and the real-estate prices are catching up with the largest cities in the world, at roughly $500 per square foot).

But it was really interesting to find out the appetite of this new burgeois young generation for mobile services.  The latest handset models are being sought, and the monthly spend ranges between $20 to $150.  MTS has become a wireless powerhouse, amassing 30 million subs, and becoming the world's fastest growing operator last year.

Read it all!

   more »

Business Week Online has an interesting section on nanotech, which is definitely worthwhile reading. Here is a quick excerpt:

For now, nano is starting out modestly. The biggest markets for nanoparticles remain in familiar products, from the black rubber filler in tires, a $4 billion industry, to the silver used in traditional photography. Lux Research Inc., a New York nanotech market researcher, estimates that only $13 billion worth of manufactured goods will incorporate nanotechnologies this year. That's little more than a rounding error in the global economy.  Yet new nano-based products that could have a far bigger impact are only a step or two away. Within the next two years, diagnostic machines with components built at the nano scale should allow doctors and nurses to carry pint-size laboratories in their briefcases, perhaps to test for HIV or count white blood cells on the spot. Nano sensors will scour airports and post offices for anthrax and sarin. Toward the end of the decade, scientists say, new computer memories composed of nanoparticles could conceivably pack the digital contents of the Library of Congress into a machine the size of a yo-yo. By that point, Lux predicts, nanotechnologies will have worked their way into a universe of products worth $292 billion.

Nanotech has been discussed often here at TF, including this entry, in which the Merrill Lynch Nanotech Index was mentioned.  I always wondered if with nanotech, the famous prediction behind Moore's Law will finally be a thing of the past.  There are quite a few semis firms with interesting concepts, including Emcore, JMAR Tech and Ultratech. 

   more »

David Appell reports on the MIT Technology Review blog that British PM Tony Blair is using the G8 forum to emphasize the climate change issue.  But this happens just as there is a lot of controversy of the "hockey stick" study that concluded that the big climate change over the past century is due to gas emissions.  You be the judge, as there are cases being made here about the flaws of the study, and here in support of gas emission controls. 

   more »

Rodrigo Sepúlveda Schulz (hat tip: Om Malik) reports on the incredible broadband price wars going on in the French market.  For instance, free.Fr offers 20 Mbps service for €29.99, or $1.50 euros per Mb/s a month.  Neuf (subway ad shown to the right) has an 8 Mbps access monthly package for €14.90, whereas Noos (an MSO) offers 10 Mbps for €10 a month. 

So what's behind all this?  Could it all be due to France Telecom's decision to "be all it can be" (to borrow the catch phrase for US Army recruitment)?  France Telecom (FT) opted to offer the maximum throughput available on all ADSL lines in France, having pretty much reached an almost universal plateau of 6 megs to the majority of its customers last year.  This is in sharp contrast to its neighboring incumbents (BT and Deutsche Telekom), that are still engaged in limiting their DSL data rates.  FT's DSL penetration is roughly 90 percent and is expected to grow over 95 percent this year.  Also, this year, ADSL 2+ will be deployed and speeds will go up to 16 Mbps (pretty amazing).

But despite all that progress, Rodrigo wonders whether or not these service providers are committing harakiri in dropping rates so low that even costs are no longer covered.  I am drooling thinking "what if" Bell Sympatico and Rogers could ever offer us such value (dream on).  But what will the French net citizens do with so much bandwidth?  That's a pretty valid question.  HDTV is roughly 19 Mbps, and one of the highest riders.  Perhaps video conferencing? 

What if Vonage were to offer a three-way conference calling service in France and call it "Vonage à Trois"?  Are you laughing?  It certainly would not be beyond the realm of possibility.  Mark Evans reported on a poll that Vonage Canada conducted with the help of Decima Research, geared towards finding out who people would want to engage in phone sex with.  The result?  Brad Pitt (the top choice for the ladies) and Angelina Jolie (the top gentlemen's pick).  This is all a part of a branding campaign launched just in time before Valentine's Day.

   more »

Despite some gloomy forecasts out there about the wireline capex impact of the fusion between SBC and AT&T, there will be also a few opportunities to be had.  Management's guidance was that there is little impact in capex plans for 2005 and that there will be some savings of 3% of their total budgets (roughly $200 million).  It is debatable whether or not those will hold true, but the fact remains that AT&T (NYSE:T) and SBC (NYSE:SBC) mostly have a non-overlapping footprint (with the exception of some of AT&T's metro networks). 

From a TDM switching perspective, the legacy Lucent and Nortel Class 4 and 5 switches will stay in the SBC network until either their useful lifetime expires or there is a real compelling business case for migration to IP.  SBC has been evaluating softswitches from Alcatel, Nortel and Sonus for quite sometime, and there are some rumors that Nortel might have a slight edge over its competitors.  While SBC has been slow out of the gate in replacing its Class 5 infrastructure (when compared to Verizon), it could be conceivable that the new SBC might accelerate those plans, particularly considering the IP-centric nature of AT&T's core.

On the AT&T side, a capacity upgrade of the backbone might become necessary to support the extra SBC traffic, with such a deal being pegged at a quarter of a billion dollars.  The usual DWDM suspects will be in the running for such an RFP when it is announced, including Lucent and Siemens.  It is also likely that AT&T will develop some video infrastructure to support the offering of triple play services.

Regardless of the above-mentioned opportunities, it is true that historically, carrier consolidation disrupts vendor sales.  At the end of the day, a vendor like Cisco might sell less IP gear because SBC might no longer have to build its backbone on a nationwide scale.  But the flip side of that argument is that if carriers are integrating networks, this might accelerate the migration to an IP infrastructure. Also, the pursuit of synergies might lead a carrier like SBC/AT&T to pursue a newer architecture such as IMS (IP Multimedia subsystems).

   more »

Tyler Hamilton writes about the imminent launch of Virgin Mobile Canada (around March 1). He believes that the clues are manifold: Sir Richard Branson (the flamboyant multi-millionaire) is coming to Toronto to participate in an upcoming Canadian Club lunch event (in March), more front-line sales activity and perhaps the fact that there is no longer a low-entry option for the young, consumer market segment, now that Rogers killed the old City Fido plan, so what better time than to grab those customers.  Coincidentally, Virgin will go after that exact same clientele that Fido once did (before it got neutered).  Which makes one wonder why Virgin and Bell could not have made a stronger play for Microcell after the initial low-ball offer from Telus?  Granted, it was a different network, but it would have gotten Bell a piece of the GSM action, something which Virgin Mobile is quite familiar with.  Besides, just having the same infrastructure will not eliminate other headaches such as OSS/billing integration, etc.

So here is some hope that the entrance of Virgin might shake up this otherwise oligopolistic market - although to follow the musical thread from the previous post, I would not even hold out as much hope for that, slapping on an andante tempo on cellular price drops... (the reason being that Virgin will rely on Bell Mobility's network).

   more »

Characterizing the market development of VoIP and the overall adoption of the technology can be tricky sometimes: one has to take the good news items and the bad ones and then try to see how they fit within the big picture.  There are many people drinking the VoIP Kool Aid and believing that the Class 5 switch to softswitch migration can take place within five years or that the entire enterprise installed TDM PBX base can migrate to IP PBXs by 2006. History shows that these transitions take far more time (witness how long it took for switches like the DMS to be deployed worldwide or how long the "digital PBX revolution" took). 

Despite that, there has been an incredible amount of optimism within the VoIP industry particularly during the past year (witness the great increase in attendance of conferences such as VON, Internet Telephony, and VoiceCon).  Perhaps the best way to temper some of these high expectations is to use a musical term: allegro ma non troppo.  This Italian tempo marking literally means "quick but not too quick".

Here are a couple of examples that suggest that VoIP's tempo is definitely not the presto that some are suggesting.  Om Malik wrote in his blog about a Forrester survey that found that only 13% of consumers were "interested" or "very interested" in VoIP.  Furthermore, that interest only goes up when consumers have a tangible benefit such as savings.  And, closer to home, Mark Evans mentions a Pollara survey that suggests that only 9% of Canadian Internet users would buy Internet service if it were priced at $20 per month.

   more »

While voice and data convergence in the enterprise has been considerably refined in the past three years, very few vendors have put much thought behind convergence of devices. There has been a lot of emphasis put on the so-called softphones, which are CTI (Computer Telephony Integration) applications that can run on PCs and provide end-users with a rich set of functions.

While the softphone has been labeled as the “desktop of the future”, that goal has remained quite elusive as most vendors did not understand that these applications should ideally work with desktop phones and not necessarily replace them (except perhaps for special cases such as traveling salesmen who can load them up on their laptops and take them on the road). Business cases were made for the IP softphones to replace IP hard phones, but nobody figured a way to get more functionality of a desktop phone via a complementary CTI application that worked in conjunction with that desk terminal.

Enterprise telephony equipment vendors have struggled to understand that end-users typically prefer an interface that is not overly complex and with which they are already familiar with rather than something overly complex. A sophisticated IP phone with color displays and web browsing functions sitting right next to a PC is just an expensive toy that will hardly be used for these functions (the PC is just as capable for viewing some web sites). Of course, that same phone in a different setting (such as an airport lounge or a hotel room, where a PC is not readily available) would have an entirely different appeal and value proposition, but why settle for only niche opportunities?

Another forgotten point was human factors. How many vendors out there are selling softphones that have pictures of phones on the screen? How bad is that? Instead of taking advantage of an environment that is as rich as a PC, with nice GUI gizmos such as dialog boxes, dropdown menus, radio buttons, etc., let’s just go back to the same old 12-button desktop phone interface and emulate that on a PC.

This is why it was so refreshing to see innovative products such as Mitel’s Navigator, which was announced today at BCR’s VoiceCon conference in Orlando.  Mitel has brought a new form factor to the world of IP telephony: a phone bar that can deliver powerful functionality via a tightly knit interface to the PC world. This allows Navigator to work in conjunction with applications such as Mitel’s YourAssistant (YA) and environments such as Microsoft’s Live Communications Server. The new ergonomic design saves desktop space, is very intuitive to use and allows users to define their own functions via 9 programmable keys. These buttons can invoke PC applications, telephony functions or commands. Navigator also supports SIP and hands-free calling, but the key differentiator is the rich suite of functions that YA can deliver, including collaboration, instant messaging, presence and video. Users can drag and drop a few contacts on a container and initiate an impromptu conference call on Navigator via just a click of a button. The end result is pretty compelling: a phone with an innovative form factor that is easy to use, yet that can deliver a lot of powerful functions via PC client applications.

   more »

Mesh networking is getting to be quite a hot topic – so much so that even the IEEE is forming a new protocol geared towards mesh networks (802.11s). In most WLAN network setups, access points are typically linked to a wired backbone Ethernet infrastructure, but with mesh networks, the idea is to relay traffic throughout the network completely wirelessly. Of course, the purpose is to drive down the costs of transporting data between Wi-Fi hotspots and wired broadband networks. This concept can also be extended over large areas at a low cost, without major right-of-way issues, since the equipment can be located virtually anywhere.

Nortel (NYSE:NT) has been one company actively working in this area, having run trials with BT (British Telecom) and MIT (the wireless mesh network there provides connectivity to e-mail, Internet and access to campus files and content securely from any location with any device; that trial also features wireless data services created by the MIT Media Lab). The Nortel approach came from peer-to-peer and ad-hoc wireless network research carried out in its Ottawa labs. Another vendor (Silicon Valley startup MeshDynamics) also has an interesting solution, relying on a three-radio architecture (two 802.11a radios devoted to backhaul and one 802.11b/g dedicated to serving clients).

It will be interesting to see the market development of mesh networking and watch where the technology will be initially deployed. The municipalities, enterprises, military and public safety segments can all represent good initial beachheads for meshing.

   more »

From Israel comes this innovation that enables cell phone users to conduct conversations while being quiet.  Silent Communications, an Israeli startup, has a technology that allows users to carry a two way conversation without making sounds and bothering others in settings such as a meeting room, a concert or a lecture hall.

The secret is using pre-recorded voice messages.  The end-user is able to hear the other caller and can respond by selecting one of these pre-recorded responses.  These messages can be customized, and use either the standard ones that come with the product or personalized ones that are recorded by the user on his/her own. 

I'd be willing to bet that Max Bluvband (the inventor of this system, called TWT - Talk Without Talking) must have somehow gotten the idea from the voice messaging industry, where pre-recorded messages are taped by vendors such as Nortel and come with the standard voice mail / unified messaging packages (e.g. Meridian Mail or Call Pilot).  But what's unique here is the ability of the end-user to create his own unique set of personalized canned messages that could be chosen on the fly, depending on what the person on the other end is asking.

   more »

Much has been written about WiMax recently, and most of it hasn't been very promising.  First, in a Wi-Fi Networking News article from November of last year (authored by Glenn Fleishman and quoting Robert Berger extensively) stirred quite a lot of chatter within the blogsphere (it is January and I am still getting the link in some e-mails from my friends in the wireless industry).  Perhaps Berger comes out looking like a bit of a WiMax basher when he said that WiMax is "another zero billion dollar industry" with "no products". 

Of course, we know that not to be the case (just as an example, documented right here on TF was Alvarion's win with Altitude Telecom in France last year). Also, despite the fact that Intel hasn't started shipping out the Rosedale chip just yet, that does not mean that there are "no products" in the marketplace.  For instance, Redline Communications (just up the road from where I work) already has had products on the marketplace for quite sometime, lining up over 5,000 installations and 60 countries (granted, Redline was the first one to introduce an 802.16 compliant product).  Other independent chipset manufacturers (Sequans and WaveSat) already have silicon sampling (still seeking to achieve 802.16 certification) with big customers. 

But more naysayers (such as Om Malik) pointed out delays in the WiMax certification process, and how those are tied to a lack of chips from Intel.  No doubt about it, until Intel actually starts to back its own plans with actions instead of words, the industry will not take off as fast.  It is hard to argue against that.  I have also pointed out a few other restraints in a previous note (including the reluctance of carriers to deploy anything in unlicensed frequency bands).  Of course, the merger action has not helped (Sprint and Nextel now are the only holders of such spectrum, and the new AT&T plus SBC might not be as interested to pursue its WiMax strategy, considering the huge Cingular/AT&T Wireless network it can now leverage).  Another huge factor that not too many people are writing about are the challenges in the IEEE 802.16e subgroup, as it tries to define mobility specs for WiMax (the ratification of 802.16e will still take time, probably at least until the latter part of this year).

So does that all mean that WiMax, as it was originally hyped up ("the broadband wireless standard of the future"), is dead?  Is it just a glorified backhaul access technology?  Of course not.  In answering that question, it is important to get quite a few points across first:

1- Check the history of 802.11 and how much time it took for the standard to develop and mature.  Why are people getting so impatient with WiMax, and declaring it "dead on arrival"?

2- 802.11 is not the answer to nomadic broadband access over a wide area.  WiFi really was geared towards the enterprise LAN space, the home consumer market, or public operation in confined areas (e.g. your local Starbucks or an airport lounge).  By contrast, WiMax is a MAN-type solution that 802.11 really was never geared to be.  That's why the way I view WiMax is more complementary than competing against WiFi.   

3- For greenfields (e.g. broadband for developing countries), WiMax can be deployed much quicker than ADSL or broadband cable, and the CPE costs should be pretty comparable in the future (circa 2007-2008).  This point needs to be understood - "emerging markets" want to get to progress by taking the quantum leap, rather than evolving the networks following the same footsteps of our telcos.  Just like there was a good correlation between GDP growth and teledensity, there must be another similar relationship between GDP growth and broadband access.  These emerging markets want to make that transition and increase the penetration of broadband as soon as possible.  And the people in these countries want that progress to be achieved quickly, too.  Just think of all the color TVs inside the favelas (shantytowns) in Rio.  Therefore, there is still ample time and opportunity for WiMax to really flourish in these areas.

4- Even in countries like the U.S., in competing against DSL and cable, WiMax still has one key differentiator: nomadic access.   Of course, WiFi can also offer that, but not with such a long reach.  Therefore, WiMax could find a beachhead with alternative carriers and mobile operators.  Plus, WiMax can also serve to extend ADSL networks into rural areas, for instance.  No doubt about it, WiMax will face competition from other fixed wireless technologies, but here is where the marketing machine of vendors such as Intel comes into play.

Other industry catalysts include a partnership between a next-gen carrier offering WiMax and an electric utility.  In such a partnership, the utility would act as a landlord, and lease its feeder networks as a backhaul access to the wireless broadband service provider.  Maybe some of these next-gen operators can also work out cheap access with localized, targeted advertising on a WiMax client device (the ads would change dynamically, depending on the region of a city that the user is).  

Bottom line is WiMax still is a nascent industry, but there is still time for it to develop as something more than just backhaul access.  Costs will come down, as vendors such as Intel and Fujitsu further refine their 802.16 chips.  As the standard gets solidified more high-volume low-cost manufacturers will be attracted to WiMax. 

   more »

Nortel (NYSE:NT) announced today its 1H 2004 restatements, without revealing any major surprises.  The 1Q EPS of 2 cents was higher than the break-even estimates of some Wall Street analysts, mostly due to margins that came out just a bit higher and a bigger chunk of forex income.  However, the Q2 results were a tad lower than expected due to forex losses.  Nortel still has to produce the 2H 2004 numbers, which will be filed as soon as it is "practicable".

The good news is that some of the overhang is beginning to lift from Nortel, and the "new" CFO (Peter Currie, who was Royal Bank of Canada's CFO until last September) was well received by Bay Street (coincidentally, Peter was NT's CFO from 1994-97).

I got some mail yesterday regarding my Verizon VoIP build article.  Of course, despite the fact that this particular deployment has not met some pretty lofty expectations, it was still a big win for Nortel in a key Lucent account.  In fact, it showed that Nortel was ahead of Lucent in softswitches at that point (and in fact, Lucent ended up buying Telica to catch up).  Perhaps it is time to acknowledge John Roth's famous "right angle turn" that emphasized the development of IP technologies.

   more »

A bit more than a year after Nortel announced its landmark deal with Verizon, the reality is that the RBOC only replaced between 5 to 10 of its Class 5 switches, which represents less than 1% of its total Class 5 switches in service (2,500).  This result is not surprising.  I recommended a lot of caution right from the early going.  I recall using a hockey game analogy in an interview I had at about the same time last year.  It was just the first period, and Nortel had just scored a goal, I said.  But it was still early. 

Also, while a couple of legacy Lucent switches were replaced, at such a low pace, it would seem that Lucent was able to just about maintain its share of the Verizon Class 5 installed base, with the other replacements being older Nortel DMS switches.  Since Nortel's exclusivity expires in July of this year, Lucent will still be given a few chances to get a significant portion of the replacement opportunity.  This just goes to show that, despite the initial Nortel win, it is still very hard to replace the incumbent, and in the worst case scenario, Lucent will still be the secondary supplier.  So expect that the months following July will be a good acid test for the recently acquired Telica softswitch (company acquired on May 2004 for $295 million).

Verizon's own Class 5 VoIP plan called for a an agressive 5 to 6 year deployment, but right now, a more likely scenario would be a 5 to 10 year exercise, with a rough annual spend of $100 to $200 million.  But that is assuming that Verizon will stick to its original allocated capex for class 5 switch replacement of roughly $1 billion.  And that other high priority items (e.g. FTTH project, IMS, etc.) will not draw more resources.  Those are two big ifs.  Nothing but a big dose of reality for those engaged in a bit of "irrational exuberance". 

   more »

Of course no sooner that I wrote a slightly more positive spin on a particular spammer (Amir Gans, the top Israeli spammer, courtesy of the MIT Technology Review blog), another bad consequence of spamming came to light. 

A few months ago, I discovered an interesting blog listing site called Blizg.  According to its own old "raison d'etre", Blizg was "created out of a desire to use blog metadata in interesting ways. We want to seek out and understand the relationships between and among blogs. We want to help you find blogs of interest."   Meta tag content was indeed a great way to connect like-minded bloggers.

Unfortunately, the site went down late last year (around December sometime?) and I always wondered why (thought it might have been lack of funding).  So I went on a fishing expedition to discover what went wrong.  And I finally found the answer on Blog Cafe: spamming.  Brandon Stone, the webmeister of Blizg, had to shut it down because of the incredible amount of spam the site was getting.  He did not have the time and/or resources to clean that mess up.  It is sad that spammers infiltrate every corner of blogsphere, including a unique site such as Blizg.  I hope Brandon comes back with a new site name and a bit more security to prevent that next time.
more »

With all the talk about how the RBOCs should start building out fiber-based video networks instead of relying on the satellite approach, one would think that the fiber deployments should be happening in full force in 2005. Verizon (NYSE:VZ) is favoring a FTTP (Fiber To The Premises) strategy, whereas SBC (NYSE:SBC) and others are more inclined to a FTTN (Fiber To The Node) approach.

No doubt that FTTN is the "quick fix" method, being faster and cheaper, however there are risks (i.e. is there sufficient bandwidth for when the loop length exceeds a certain threshold, or for HDTV). Also, the standards issues persist (VDSL or ADSL2+?) and then, there is also the question of the cost and availability of the set top boxes. 

FTTP is slower, and entails a higher upfront cost, but it should give Verizon the best network asset among its peers.  But the buildout will be slow, as Verizon plans to launch FTTH to 10 million homes over the next five years. 

SBC's Project Lightspeed is highly ambitious in terms of their video share gain assumptions, which call for a higher portion of the pie than what DBS players get today.  But their roll-out pace, opex and capex estimates presented in November of last year look reasonable.  However, despite the fact that these numbers were well received by Wall Street, some naysayers still point to past experiences as the best predictor of how long this deployment will take (who remembers Project Pronto?).

Another factor that might definitely dampen the RBOCs pace of fiber deployment is the potential SBC and AT&T merger.  Om Malik highlights a Merrill Lynch report that should the deal close, US wireline CAPEX might decrease by 10% (or roughly US$ 2 billion).  The SBC and AT&T were planning to spend about $7 billion of wireline capex this year, but a combined organization would mean a shrinkage of 30% (or $2.1 billion).  Of course, some of this reduction might also trickle down to Project Lightspeed, as SBC focuses on integrating the AT&T network.  In such a scenario, it remains to be seen whether or not Verizon, after seeing a combined SBC/AT&T spending less, will also follow suit.

Regardless, the MSOs will have a significant time-to-market advantage in bundling voice service versus the RBOCs in bundling video with their fiber buildouts.

   more »

Trying to capitalize on the success of electronic pets, high Christmas sales of Robosapien and the increasing popularity of MP3 players such as Apple's iPod, Sega Toys revealed plans to launch iDog, a new "music entertainment pet".  The electronic pet is able to actually "feel" four emotions: anger, humor, joy and pathos.  These emotions are expressed via seven colored LED lights mounted on the face. 

Built into this gizmo is an MP3 player that can store roughly 70 children themes and is also able to mimic one's voice (via playing sounds that are in a similar wavelength).  The pet is touch-sensitive: touching its noise will trigger a happy mood, while touching the tail will cause the dog to become upset.  The iDog also comes with a built-in microphone and analog audio input, which makes it also a rather different type of karaoke machine.

SegaToys is planning to distribute the iDog in 57 countries across the world, and the launch in the Japanese market is planned for April 2.  The company hopes to sell half a million units in Japan alone, and the iDog will definitely be distributed in the US, so it is expected it will be a hot seller over this side of the Ocean as well.

   more »

About Robotics.  I received feedback on a couple of fronts.  First, Swan Solutions Managing Director Bart Delgado (whom I had the pleasure of meeting at Inter-Tel's analyst conference earlier this month) had an interesting insight about the Seiko Epson flying micro robot video link that I included in a TF entry last year.  Bart makes a few interesting observations, including the pendulum-like motion and the blue background.  I went back to the Gizmodo entry on the same topic, and did not see any updates.  Funny how the Gizmodo folks also noticed the "enthusiasm" of the demo girl.  But I will have to see this flying micro robot for myself before making any judgments. 

Bart, being a big heli enthusiast, mentioned the Hirobo XRB Sky Robo Silver Lama (shown to the left), a radio controlled electric indoor heli, as being the closest available robot to this Seiko Epson model.  Bart also gave me a verbal update on UK sales of Robosapien, the robot that I nearly bought while visiting Boston last summer with wife.  According to Bart, this was one of the hottest toys sold during Christmas, for an 80£ price tag.  Indeed, according to this Times Online article, Hamleys (a London toy shop) expected to sell 180,000 models, whereas online gadget shop sold 20,000.

Another TF reader from the Netherlands, who has been actively involved in robotics research, designed quite a provocative scientific calendar.  The calendar is meant as a nice joke, showing robots in some insinuating positions (please see photo to the right) or showing their "private parts".  Of course, the idea of the calendar is to provide information about robotics conferences and events, but with a slightly different, funny twist.  No porn is tolerated here at TF, but the eRobotica 2005 PlayBot Calendar is given a "poetic license" here, since it disseminates robotics conferences info in a quite creative fashion.  So be sure to check out their web site and download your copy of the calendar!   more »

Simson Garfinkel, one of the major contributors to the MIT Technology Review blog, wrote an interesting entry at the beginning of the month, talking about a Haaretz article in which Israel's top spammer was interviewed.  Since the article was written in early January, the link is no longer valid, but there is a Google cached copy of the interview with Amir Gans, the fellow who is responsible for sending out 50% of the total volume of junk mail sent in Israel.  The 30-year old entrepreneur advertises all sorts of items, including discounted ink cartridges for printers, security systems for businesses, calendars for the new year, Plasma televisions, water filters, plans for tuition payment for students, poker clubs, luxury watches, flowers and telephones.

Not that I am trying to defend spamming, but this spammer is different than most of his fellow "spam professionals" - he updates his removal lists twice a week, and really keeps his word in not sending these 77k plus addresses any junk e-mail. He also distributes public service messages without charge, such as for instance, asking for a kidney donation for a sick person.  But what's interesting about the article are the details such as how he gets e-mail addresses to add to his lists, or gets lists of people who have already purchased at least one product from a spammer.  It is a great insight into foreign spamming.

   more »

Looks like Jeff Pulver has done it again.  After his very first FWD (Free World Dialup) back in the mid 90's (with Vocaltec's Internet Phone, a program which I was lucky to get a free license for, as a thank you for being one of the original beta testers), his second one FWD launched in 2002 (courtesy of a SIP registrar), Jeff now has topped himself again by having a new service called Bellster (maybe a wordplay meaning how to get Bell service the Napster way - i.e. for free).

Bellster is essentially a P2P (Peer To Peer) service that enables users to share their phone lines to make "free" calls anywhere within the reach of the global PSTN network.  The system does not provide actual PSTN access, but it assists a user to find other users sharing their lines via Asterisk (please see photo to the left), an open-source PBX (believe it or not, for an open source PBX, Asterisk has quite a decent list of features that is growing over time - it has been impressive to see this grow over time). 

Callers can call any phone number in the world by sending the call via Internet to a shared phone line near the person that is being called.  That is pretty similar to the very first FWD, except for certain rules.  The system is setup based on a "social rules arbitrage" code.  In other words, users have to earn credits before being able to make these free calls, and these credits are directly proportional to the amount and duration of calls that users allow to move over their private lines.  Jeff Pulver already listed in this blog the countries that are currently supported by Bellster, and the list includes the US, Canada, Russia, France, Italy, the UK, Germany, Argentina and Brazil, among others.

So what is the catch?  For one, the investment in the Asterisk PBX ($1,030, quite a steep initial investment to make supposedly "free" calls - a payback period of more than 20 months for someone making $50 worth of calls per month).  The second possible deterrent is privacy.  Because the call goes through the Internet, there could be very little in the way of assurances of a secure call.  Hence, security could be a major concern.  But still, this is cutting edge in that calls can be made from a PC to a phone anywhere else in the world.  It is also a service that is growing quite fast, with the recent tally showing 22 countries available and 487 nodes registered across them.

Update (26/01): Another issue that I forgot to mention is the cost of local calls in some regions.  For instance, in South America and Europe, each local call costs (taking Brazil as an example, each local call during business hours costs an "impulse" for the first few minutes, and then additional ones after that, much like a pay phone in the US; the story changes during the evenings and weekends when each call is one impulse, no matter how long it is).  These costs would be billed to the local user, so obviously there is an imbalance there versus in North America (US and Canada), where users can make unlimitted local calls paying the same monthly flat fee.  That said, on the plus side, the receiving end does not need to have a computer or any broadband access (in case someone wants to call their grandma who is not a computer user).

   more »

Is it just me or is the amount of hack headlines increasing?  I keep reading about the threats of "evil twin" hotspots (actually, a quick search on Google News produced 135 results) and how it can allow hackers can clone an AP that behaves as part of an integral WLAN.  As more attacks become known, it seems like some hackers enjoy the extra notoriety, leaving their "signatures" on the sites knowing that their hacks will be publicized later.

Here are a few recent ones:

- T-Mobile: hackers penetrated their systems and read e-mails from an agent from the US Secret Service, among other company customers (400 total, including sensitive information such as their social security numbers.

- Gmail: sometimes, people discover vulnerabilities by accident, such as on this instance, when a Unix community group found a way to retrieve username and password information from Gmail users.

- Harvard University: some hacks do not even need to be sophisticated due to really lax security practices.  In this case, confidential drug purchase histories of Harvard students could be obtained by anyone knowing their student ID numbers. 

This reminds me of my days at Nortel, where I once worked on the Symposium Contact Center product.  All user manuals, training, etc. highlighted the fact that the default password to the product needed to be changed by the system administrator.  Then, I remember seeing an article that appeared on 2600, one of the most widely read hacker magazines.  I was amazed to discover the amount of system administrators who just never bother to follow those instructions (this was not Nortel's fault, of course, but pure carelessness by the sysops).  Of course, this poor practice could potentially expose their systems to hackers. 

   more »

A couple of weeks ago, IBM announced it was going to offer open access to 500 key open-source software patents for the developer community at large to develop open source programs and work on operating system alternatives to Microsoft Windows (such as Linux).  My favorite French VC blogger (Marc Goldberg) had a good entry on the topic, including a link to a great insight from Business Week:

Why would IBM allow others to use its intellectual property free of charge? In a word: Microsoft. The move is central to IBM's efforts to fend off Microsoft and its Windows monopoly. While the computing giant will continue to innovate and gather new patents as aggressively as ever, at the same time it is stepping up efforts to bolster the world of open-source software. IBM figures that doing so will give it a leg up in selling the software and services that work with the open-source programs it helped develop.

Of course, since Marc wrote his article, Sun Microsystems announced it will open source a component of its Solaris OS.  This is just a precursor to fully opening Solaris 10 completely.  Apparently, Sun is not too worried about IBM or HP taking an application such as DTrace and incorporating it into their own OS versions.  Sun's COO and President (Jonathan Schwartz) challenged IBM in an open letter in his own blog to port its various packages such as DB2, Tivoli and Websphere to run on the Solaris 10 OS (which analysts believe will begin shipping in March). 

Despite all this open source action during the last couple of weeks, open source penetration remains relatively low, with Linux expected to reach 6% of the market by 2007, up from 3% today. By collaborating on the development of operating systems such as Linux and Solaris, players such as IBM and Sun can secure a solid foundation on which to build tech projects and save substantial funds in programming costs as they battle Microsoft for OS supremacy. 

   more »