Month: February 2011

The weekend brought more disorder to the Middle East, particularly Libya, but while the initial turmoil there had knocked stock prices down a bit, the decline has not been alarming and it was reversed on Friday.  Today futures and the European exchanges both suggest an up market again.  Even cooler-than-expected growth in US consumer spending isn’t hurting, and some suggest that Buffett’s bullish letter to investors may be the cause.

In the tech world, Cisco’s stock-price woes continue; the company has been largely flat since its earnings call while competitors Alcatel-Lucent and Juniper have been on a bit of a tear.  Fundamentals isn’t much of a motivation for stock movement these days, but it is clear that investors in the main believe that the latter two stocks have a potential for an upside and Cisco doesn’t have that same potential.  Objectively, I think that’s all true.  Cisco need to work through some very real product issues as well as redefine its internal sales-driven (as opposed to “value-driven”) culture.  Alcatel-Lucent and Juniper both need to learn how to sing better, but both have made what could be very significant product announcements in the last couple weeks.

OK, Cisco is in the dog house for now, but I still have to point out in fairness that the company could largely eliminate its problems in a stroke with some light-weight M&A and some heavyweight re-positioning and strategizing.  The service layer, which means the cloud-to-network binding for both enterprises and service providers, is the sweet spot of the future market.  Own it and you can hope to pull through your solutions en masse.  It’s still open territory.

There may be cloud architecture competition emerging from new quarters.  F5 today announced it had worked with IBM to develop a reference architecture for the cloud.  The architecture clearly covers the creation of private clouds based on virtualization, and F5 promises that it will be extended to envelope public cloud components to hybridize them with private clouds.  We see no reason why the architecture (which looks much like Eucalyptus, and that’s no accident according to F5) can’t be used for public cloud applications, including service provider clouds.  IBM has specific aspirations in the service provider space, and the reference architecture may be a step in helping prospective SP clients build cloud services that can then easily hybridize with enterprises.  It seems to us that the approach would also support SOA applications, but that’s not a specific part of the release.

Staying in the cloud, Verizon is planning to offer UCaaS, hoping to capture a share of business buyers who want unified communications and collaboration that includes users on mobile devices.  Generally, businesses embrace the notion of service-based pricing as opposed to building their own solutions because they like the cash flow better and because they may fear making a capital investment in a space that’s undergoing major change.  However, carriers have for years lost market share with hosted communications options relating to voice services, and it seems to me that this offering would be all too easy for OTT giants like Google to counter if they feel like getting into the space.

Moving to consumer social networks, JP Morgan says it’s going to take a stake in Twitter, and speculation is that will happen by buying out some existing investors.  The deal is said to value Twitter at over $4 billion, and it’s the sort of thing that already has the SEC concerned that private equity is circumventing the protections created by public corporation status while keeping the companies private in name.  I’ve got major reservations about any strategies that have the effect of empowering the “professional” investors and not the general public, which this would surely seem to do.  Further, I wonder whether we’re not creating another opportunity for bubbles by creating a whole new exit strategy set; companies don’t sell out, they don’t go public, but they sell pieces off privately to pay off early investors.  How do we avoid collapse when eventually the public has to bail out the last of the “private” investors like JP Morgan?

The murky regulatory area isn’t getting less murky.  Republicans have recently signaled that they’re not prepared to compromise on their rejection of any sort of net neutrality principles. While that doesn’t mean there won’t be any (Democrats can block any attempts to un-fund or weaken the FCC’s position here), it does mean that if the courts throw out the FCC’s latest order (which I think is likely) then there’s no option to create comparable rules through legislation.  That would mean market forces would decide what happens, always a risk but perhaps not as great a risk as bad explicit policy.  The current FCC order isn’t bad in my view, but I think there’s less than a 30% chance it will stand.

Another semi-regulatory issue is raised by Comcast’s announcement it would not be offering paid streaming video service to non-subscribers, something at least one satellite TV rival says it’s preparing to do.  That may raise an issue with regulators who think that Comcast must make at least NBCU content available to competitors on the same basis as they offer it internally.  Does not offering separate streaming video satisfy that condition?  Comcast may have another reason to appeal the FCC’s order—which is already a target of appeal by other players.  The Comcast/Level 3 dispute may even join the parade here!

Tech, overall, is in a bit of a state of flux, which may be why it’s off today when the Dow is up.  Good economic conditions overall don’t guarantee tech company success these days, and since bad economic conditions guarantee failure in most tech sectors, the industry may be headed for some whipsawing as investors try to price out the current muddy trends.

Amazon has further complicated the already-complex world of streaming online video by announcing their own service, which is included in the free-shipping Prime membership.  The service currently includes about 5,000 items (movies and TV).  Obviously this isn’t good news for Netflix or even Apple, but it’s also a new step in the growing challenge faced by network operators.  I commented on that in detail yesterday, so we won’t reprise that topic so soon!

Apple’s problems with its subscription policy may be headed for clarification, but if so it’s not taking a giant step as yet.  An email attributed to Steve Jobs said that the policy of requiring subscription services (something that requires periodic payment, according to the app store policy statement) work only through Apple and pay 30% isn’t for software services but for publications.  The problem is that’s not what the policies say, nor how Apple appears to be interpreting them.  The company has been taking growing heat from developers now that a few additional apps have been tossed, including some that would appear to fit within Jobs’ description of a software service.  Obviously they’ll have to do something to prevent developer defections, because some publications say there could be tens of thousands of apps already in the store that will be impacted by the new policies.

The underlying reality here is that everyone in the online game has to make money these days.  Apple has brand awareness and coolness, and they’ve spent a lot of money and R&D to develop and maintain that.  They want a return on their investment.  So do app developers, and publishers of content, and portal sites and network operators.  For over a decade we’ve been in an age where Internet eyeballs and customers were their own currency, but now we’re seeing a wave of demand for monetization.  A good part of that, in my view, comes from the death of the “buy up startups to help the VC industry” mindset of the ‘90s and even into this decade.  More and more startups have to go public to cash out their investors, and that means showing a profit, and the demand is particularly great for online dot-com players.  Fifteen years ago, virtually no startups who managed a successful exit were profitable; the companies who bought them were getting technology or image.

Infonetics released its latest numbers on the carrier switch/router market, and they show Alcatel-Lucent has taken second place (after Cisco, still the clear leader), overtaking Juniper.  The only reason we can gather for this shift from our survey data (which doesn’t include questions on market share) is that Alcatel-Lucent has sustained higher credibility in the wireless space as well as in access, optics, and the service layer.  They can then pull through their network-layer assets.  I noted when Cisco’s earnings came out this quarter that Cisco apparently was not leveraging its service-to-network opportunities well so far, and that may hamper Cisco’s turning around its share losses versus Alcatel-Lucent unless Cisco can address the issue.  Of course, Alcatel-Lucent probably won’t stand still either.

The issue of cybersecurity and the “kill switch” on the Internet continues to raise blood pressure on both sides.  Reworking of the cybersecurity bill to pull the explicit notion of a kill switch isn’t satisfying some who think any restrictions on the Internet constitute a control of free speech.  The problem is that there are legitimate cybersecurity issues, and criminals or terrorists attacking critical websites could create situations where the government had to intervene in some way.  The challenge is to create a framework that permits intervention but doesn’t open the door for the mechanism to be used to censor free speech.  Normal judicial review may not be enough because the process could take too long—too long to get a warrant to act, and too long to appeal an unreasonable application of the authority.  Logically, I think this should be an FCC issue because that agency is the federal expert on communications, but so far there doesn’t seem to be much movement on either side to support a compromise.  It may be that the best strategy would be to do nothing and let the government convince ISPs to act voluntarily, if only because that seems the likely outcome if no specific approach can be agreed on by all parties.  The debate is getting more polarized and ridiculous every day.

We now have an Institute for Civility in Government; maybe we need to expand its scope.

Apple, already facing an anti-trust review or two, is now getting growing push-back from app providers over the subscription-sharing rule.  Apple wants a cut of every subscription, meaning that they want apps that sell something to sell only through Apple’s store and not directly to the consumer.  If dissent spreads here, it could be a worse problem for Apple than government scrutiny.  From the very first days, Apple has fostered a closed ecosystem model to the greatest extent it could, bucking a general industry trend toward opening software and systems to third-party exploitation.  Google, of course, announced its own program for an Android store that’s considerably more financially friendly to publishers and streaming audio/video apps, which only puts more pressure on Apple.

The rumors of a lower-priced iPhone and even iPad are further indications that Apple is worried about competition from Android.  Just as the PC-compatibles shunted Apple aside in the desktop wars of the 1980s, Android-based devices are threatening to diminish Apple’s market share and marginalize it with developers—those who aren’t already upset by Apple’s store policies.  But Apple loses in any price war even if they win, because they’re always seen as a player who sustains higher margins.  With Jobs’ health now clearly a problem, the difficulties could be harder for Apple to work through.

Meanwhile, the smartphone and tablet wars are putting pressure on mobile providers, and in particular creating competitive drive to migrate to LTE.  Operators tell us that their LTE migration strategies have been advanced “over a year”, and this is creating its own set of issues because the faster move means that a lot of users will either have to be driven to change out their phones or there will be a considerable number of older 3G devices still in service when 4G rolls in.

One impact this has had is increased interest among providers in using packet-mode infrastructure to backhaul circuit-switched connections.  Running a single fast feed to a 4G tower is bad enough, but doubling it up with TDM backhaul for circuit-mode voice is truly bad news.  Extreme Networks did a mobile backhaul announcement that focused on synchronous backhaul and integration of packet-sync traffic with TDM, and now Juniper has snapped up the IP of a startup with synchronous packet backhaul capability.

Another issue created by pushing 4G into the fast lane is the change in the dynamics of the metro network created by all the mobile backhaul paths.  A typical metro area (LATA) in the US today would have an average of about 200 central offices.  There would likely be ten to twenty times that many mobile cells (in addition to the femto and WiFi sites).  How will all these new locations impact the metro mission?  And given that 4G will almost certainly drive packet-mode voice (VoLTE), could there be pressure to migrate wireline users to VoIP?  Operations clearly has to contend with this.

Wireline broadband has its own issues.  It is very clear that video streaming services are growing, partly in response to the tablet opportunity, and this creates special problems for broadband operators, most of whom see video services in the form of channelized TV as a big revenue opportunity.  So now they’re faced with having OTT streaming services using the operator’s own lowest-profit Internet service to compete with channelized video that’s supposed to be that operator’s highest-profit service!

The debate here, crystallized in the Comcast/Level 3 dispute, was recently punted by the FCC even though no formal complaint has yet been filed.  But not only may Level 3 file such a complaint, Netflix is making noises that it might do the same.  Cable MSOs are particularly sensitive to OTT video competition because of the fear that cord-cutting will catch on.  They’re similarly concerned that HD and 3D services, when streamed, will create even more traffic and pose major congestion issues unless the cable companies build out more.  Remember, their cable spans are shared-capacity, so they may have to make more radical changes to scavenge bandwidth for online services.  And they want somebody to pay.

In the cloud space, Huawei has pulled its offer to acquire 3Leaf’s virtualization property, after a US government panel recommended against the deal.  The Chinese government expressed regret that the company had taken the step, though I think it’s unlikely they were really surprised by the move.  What’s likely to happen now is that Huawei will shop for software intellectual property and other assets outside the US, which may be harder in terms of finding candidates but easier in terms of getting the deal approved.  Remember, though, that there’s a lot of virtualization and cloud assets out there in open-source form, and Huawei would be free to exploit these as long as they complied with the license restrictions.

AT&T is continuing to push its notion of the cloud as being a natural extension of the network, enhancing its cloud computing offering.  One of the new features is tighter binding with AT&T VPNs, creating a “virtual private cloud”.  This is also a model Verizon seems to favor.  It seems to me that this would naturally create interest in cloud infrastructure as both a customer service framework and as a platform for carrier IT and feature hosting, but hey we’ve ignored similar evidence of symbiosis for a long time!

In the economic/geopolitical space, the situation in the Middle East (particular in Libya) remains very tense, and it’s far from clear whether the “democratic” movements will really result in any meaningful shift toward democratic government.  The US exchanges are closed for a holiday, but international exchanges have taken a hit and US stock futures are down.  Oil prices are rising too, and the situation could worsen if anything truly bad happens.  But I think the major US economic risk is still the shift away from a manufacturing economy to a service economy, which can’t produce growth in wealth large enough to satisfy people’s desire for upward mobility and can’t generate taxes enough to bail out revenue-strapped states.  Will this derail our recovery?  Not likely, but we have to watch the domestic developments as closely as the international ones.

Well, we’ve got the usual regulatory flap as we end this week, with the same players and the same issues.  Republicans in Congress are looking for a way to derail the FCC’s neutrality order, and the strategies range from a disapproval vote (which only buys some time) to pulling funding for the measure (a cop-out that has no chance of passing and getting by a veto).  It’s hard to say how much of this is politics, how much is lobbying, and how much is posturing to get on the “right” side of an issue the courts are likely to throw out eventually.

The FCC, meanwhile, has said explicitly that its neutrality order doesn’t cover the dispute between Comcast and Level 3, and this statement could be good or bad depending on your slant on the neutrality order’s legality.  Presuming that the FCC could craft something that passed legal muster, it would make sense for it to cover the critical issues in the industry rather than let them blunder on some random path.  The question of Internet settlement is one of the most critical of all, and so I’d have liked the FCC to have taken some strong steps there.  However, you can’t take a strong step from an order without legal foundation, and given the questions on the neutrality order it may be best that the FCC doesn’t intend to intervene in key issues based on its authority.

Many of you who have read my stuff through the years know that I’ve been a strong supporter of explicit Internet settlement, to the point of being a co-author of an RFC on the topic in the mid-90s.  I believe that any business ecosystem that can’t settle payments according to proportional involvement of parties will eventually fail.  In particular, premium handing and services are hard to imagine in a bill-and-keep space, unless everything is “on-us”.  Thus, lack of settlement is one of the biggest issues to address if you really want an “open” Internet.  The FCC is saying that the market can decide, but of course that will work only if Congress doesn’t get in the way, and in any case the market hasn’t decided up to now.

Comcast’s numbers, which included broadband subscriber gains that far outstripped the estimates of analysts.  Basic cable subscription losses were less than expected, but still there was a loss.  The data suggests that the media blitz on “cord-cutting” was largely hype, which is what my model had showed.

TV viewing is definitely undergoing changes, and some of these are exacerbated by the increased availability of online material.  But the big problem is a growing dissatisfaction of viewers with “network TV” and defection to cable channels for an alternative.  This defection works for a time, but the material on cable isn’t inexhaustible and some of it isn’t much better from the viewer perspective than network TV.  The same can be said for online content via Netflix or Hulu or whatever; the pool of material is only so large and the subset that will appeal to a given viewer is even more limited.

Live sporting events continue to garner large audiences; better penetration of the viewer base than in the past, in fact.  That shows that it’s what you want to watch and not how you want to watch it that matters most.  Comcast’s NBCU deal, in this context, is critical.  What the company needs to do now is to create a different model of content, something tuned to the way that consumers have already voted (with their remotes).  They don’t want star power, or series TV that quickly grows stale, or successions of follow-on shows or re-makes.  They want stuff that’s entertaining, and some summer-season successes in cable show that can be achieved with low-budget production.  On their earnings call, their management suggested that it recognizes the cable-channel properties may be the secret value to this deal.  I think Comcast has an opportunity now to churn out dozens of new shows for what one big network series might have cost them, and by doing that collect a larger audience and greater loyalty to the channel model.  They can then focus on how to exploit their growing content inventory online, through multi-screen.

AT&T may be looking at its own video marriage, but rather than buying a network it may be buying a satellite company.  Rumors of a deal to acquire DISH have restarted, and there are certainly things that recommend the deal.  AT&T doesn’t have the economic demographics that Verizon does, and can’t easily deploy fiber to the home for a large chunk of its population (in fact, it’s behind even on fiber to cell sites).  U-verse, based on the slotting of DSL bandwidth for multi-channel delivery, is pressured by the demands of HDTV (and possibly 3DTV) and by cable company competition offering higher broadband speeds.  In theory, a strong satellite position could be a big help, and while AT&T has offered that in a resale deal, buying DISH would give them control and all of the revenues.  There’s also speculation that AT&T might want the spectrum space DISH controls.

In the telco space, MWC seemed to bring out support for VoLTE; both AT&T and Verizon indicated they’d be making the Big Move, though over a period of two to three years.  There are still open questions on how LTE networks will support voice, how they’ll support roaming and premium services, and how much or little IMS and more traditional interconnect concepts will drive the process.  But presuming that the migration happens on schedule and that operators continue to push hard with phones, it’s not unreasonable to think that by 2015 we’d be seeing some pressure on wireline voice.  Will this create that final push to FMC and IP infrastructure everywhere?

Sure, but probably not all that quickly.  The problem is that the great majority of households retain PSTN voice services.  Operators tell me that they don’t believe that they’d have reached 80% penetration of LTE handsets before 2016 or even later in some areas.  They believe that regulators would be distressed by plans to force upgrades either in wireline or wireless unless the operator could essentially create PSTN at the home dmarc and sustain compatibility with current in-home and in-business phones.  They’re worried about lifeline, about powering phones, and about 911 services.  But most of all they’re of the view that recapitalizing voice services when voice ARPU has nowhere to go but down is a bad move.  So they aren’t in a rush to change their strategy for wireline.

What VoLTE does show is that migration to VoIP and the resulting changes in infrastructure are most likely to come about because of some business initiative that creates new revenue.  Transformation is a proactive process, driven by opportunities to make money or requirements to make it on something different.  Because LTE is 4G, 4G is the future of wireless broadband, and wireless broadband is the future of mobile, operators are moving to VoLTE.  The “future” in all these cases is seen as being revenue-positive.  It remains to be seen how fast IP voice will penetrate the rest of the TDM world.

A couple of talks at MWC may be signs of important future trends in mobile and online.  AT&T CEO Stephenson said that the difficulties in moving content between devices, was hampering mobile content opportunity.  He also commented that AT&T believed that apps should run across devices, not be linked to a single gadget.  While at least the second of these could be a sour-grapes reaction to Apple’s App Store concept, no longer an AT&T private lake, it does seem likely that Stephenson is on to something regarding content portability.

Google’s outgoing CEO Eric Schmidt also spoke, and his often-disconnected talk did showcase something important; online mobile services are reshaping our lives in ways nobody predicted.  If you add this to Stephenson’s talk, you get a picture of an industry that’s grappling with a combination of enormous opportunity and influence and enormous structural change.

We can find out what friends are doing almost in real time, so at least for a time we do.  We can grab video clips to laugh over, check the weather, and so forth.  But all of this is like a new TV show in a way; the first couple episodes are fresh and exciting, then the writers just run out of ideas.  What Stephenson is worried about is that a focus on mobile social indulgence is going to cover up the need to establish something real and useful at the core of the mobile revolution, particularly in content.

My research has long shown that the majority of online content consumption takes place when and where traditional content can’t be consumed, meaning that it tends to be incremental to rather than to replace the old forms.  It also suggests (though this is a hard point to survey or model) that a decent piece of the consumption is either linked to youth behavior that passes with age, or to novelty.  But there’s no question that a smart multi-screen strategy would go a long way to creating a truly useful and enduring model because people often have to skip out on something, move during a showing, etc.

Deeper into Stephenson’s talk you find the app reference, and if you think about it there’s a linkage here.  Apps that are truly compelling need to be fully pervasive.  We know that the reduction in price of consumer electronics is going to breed a multiplication of devices, and so we need to ask whether we can become dependent on future apps without having those apps available to us when we think we need them.  What Schmidt is suggesting is that mobility and online services will change our behavior, and that’s most likely to be true when we can learn to depend on the combination.  Every time we don’t have access, every time we have to stop and start, every annoying glitch, makes dependency harder to establish.

In many ways, what both Stephenson and Schmidt are suggesting is that we need more network-centric services for mobility, which is ironic given that for voice service that’s what we’ve had from the first.  The broadband revolution in the mobile world bypassed the network operators almost totally.  They jumped on smartphones to pull through higher ARPU and let the appliance vendors and their stores add non-voice (and now even voice) features.  That could have been a fatal step, and still could be.

Handset players, especially those like RIM and now Microsoft/Nokia, seem to be courting operators to dodge the problems of feature disconnect by partnering with them.  That, of course, only limits the number of people stealing from you in the perception of the operator.  Somehow they’ve got to be a part of the picture, and they’ve been so far unable to wrestle the initiative from the handset players.