Communications

Gaping hole in models for using spectrum efficiently

In February, the FCC’s Office of Strategic Planning and Policy Analysis published three studies (1, 2, 3) on spectrum licensing and spectrum utilization.  Thanks to Nick Ruark for pointing them out.

Since the papers discuss tradeoffs between licensed and various “unlicensed” approaches to spectrum utilization, I printed them out and threw them in my luggage, finally reading them today.  These are theoretical studies that assume you are familiar with Nash equilibrium (named after John Nash, the mathematician made famous by the movie A Beautiful Mind) and mechanism design (a part of game theory that is directly applicable to designing auctions).  Heavy stuff indeed!

Reading on I was struck by a gaping hole in their assumptions.  Of course an academic study can start from any assusmptions the authors prefer, but it helps if there is some relationship to reality.

What’s missing in these papers is any thought of future innovations.  The key feature of the currently available slivers of “license exempt” spectrum (in bands just above 900 MHz, 2.4 GHz and 5 GHz) is that they provide options for future innovation.  None of what we know of as WiFi or Bluetooth, either technology or business models, was in existence when the FCC originally opened up the 2.4 GHz band — a band that was previously devoted to (and is still used by) microwave ovens and industrial heating equipment.  Indeed, the principal reason why the FCC is studying spectrum allocation models today, is the innovation that resulted from opening up license exempt spectrum.

So the goal isn’t to figure out how to make current users of unlicensed spectrum use it more efficiently with today’s technology.  The goal is to motivate people to find new ways to get more utility out of the spectrum in question.  How do we foster innovation?  How do we increase the opportunities for innovators to find new and improved ways of using wireless spectrum?

Of course, I shouldn’t be too hard on the authors.  After all, the economics profession only began to model innovation in the past 20 years or so.  For the previous 200 years, innovation was lumped in with other “exogenous” factors, i.e. factors that didn’t fit the model.

Zippy interviews me on Mobile TV

Zippy (actually Richard Grigonis) got my views on Mobile TV for a series on TMCnet. 

I'm negative on broadcasting :-) but very positive on mobile video on demand and on the profusion of other video-related things that people are finding to do with their mobile handsets.

Off for three days in Loma Linda California

I'm leaving (very) early in the morning for Gordon Cook's meeting on the Architecture and Economics of IP Networks, a.k.a. the Cook-In.  It looks like a full schedule, some people I know and many others I know only via email.  This should be very good.

Broadband penetration vs. broadband capabilities

I'm listening to the first session in the VON pre-conference workshop on Competition Policy and getting increasingly depressed.  Everyone on the panel is entrenched in traditional policy infrastructure — no thinking out of the box here.  What's more, at least two of the five people on stage seem to think the US is doing well, among OECD countries, with our deployment of fixed line broadband access.

The problem is definitional.  They are comparing broadband penetration rates.  I am comparing the relative performance of readily available broadband services.

The FCC's defines "broadband" as at least 200 Kbps in at least one direction.  The OECD uses 256 Kbps, while the ITU definition is “transmission capacity that is faster than primary rate Integrated Services Digital Network (ISDN) at 1.5 or 2.0 Megabits per second (Mbits).”  Using the OECD's definition, the US comes in 15th for broadband penetration by country.

But that's not the point!

Speeds of available broadband services

In Stockholm, dark fiber reach to every city block and an enormous number of buildings.  This dark fiber is available for lease by anyone.  As a result, there are many ISPs offerring services in greater Stockholm (and elsewhere in Sweden).  Just about anyone in Stockholm can get 100 Mbps symmetric Internet connectivity for 98 Kroners pe rmonth — that's $16 per month.

Similar service (100 Mbps) in Tokyo costs ~$26 per month and in Seoul it was closer to $30 the last time I checked with friends there.

In the Boston suburbs I have Verizon FiOS Internet access.  Downstream capacity is 20 Mbps most of the time and upstream measures out at 2.7 Mbps.  This costs $50/month.  I'm extremely fortunately to be able to get service this fast.

So my friends in suburban Stockholm have 5 time the downstream capacity and 30 times the upstream capacity for 1/3rd the cost.

Why don't people talk about broadband performance?

Most cell sites have only one or two T1/E1 links, or less

One of the things that surprised me at eComm was the audience reaction to hearing that most cell sites have only one, or perhaps two, T1 or E1 links going to them.  eComm was a sophisticated audience but they weren't familiar with this kind of operational detail.

Of course, I've seen a lot of data on this particular subject as, until our recent sale of the AccessGate product line, NMS Communications was in the backhaul optimization business.

Net-net:  As spectral efficiency keeps improving with HSPA and so forth, backhaul is rapidly becoming the bottleneck. 

Cell sites are fixed locations.  The vast majority are in urban or suburban neighborhoods.  Their bandwidth requirements will continue to grow, indefinitely.  In any rational world, one would purchase dark fiber to most of these sites.  In the irrational real world that is the US today, dark fiber is available on long haul routes but is extremely hard to come by in populated areas where most cell sites are located.

The root problem is we have given Verizon and AT&T privileged access to the public right-of-way, without requiring them to sell dark fiber, or lease dark fiber, or offer any connectivity other than T1 service.

What would it cost to construct dark fiber to 95% of these sites, if it were done as part of a community-wide dark fiber build like that being planned in Singapore and already completed throughout greater Stockholm (and most of Sweden)?  Construction costs for point-to-point fiber are 20% or so above those of passive optical networks (PON), but the cost of the dark fiber alone (no electronics) is perhaps 80% of the total construction cost, so they're in the same ball park.  Verizon has directly and indirectly reported various costs for their FiOS construction (FiOS is a PON network), but all current estimates are below $1000 per home passed. Even at $5K per cellsite, the payback for a mobile operator would be measured in months.

A few Notes from eComm

Three days of rapid pace content, mostly 15 minute segments with no overruns!  Presentation were mostly high level, focused on industry insight and direction.  Almost all were good or very good and some were excellent.  Almost no pure product pitches.  And, most important, tons of interesting people.  In fact it appears there were ~220 people and perhaps as many as 250 who attended at least part of the event.

In short, one of the best events I've attended.

I was completely immersed, and besides the WiFi was misconfigured the first day, so I didn't attempt blogging.  Others have done a great job.  Technorati has a good list.  Andy Abramson has a rundown of early media coverage although Google News has more of course.

The credit for making it all happen goes to Lee Dryburgh.  Thank you!

Above photo thanks to James Duncan Davidson.

Other pictures here and here and here and here and then there are some videos here.

Two weeks in California — eComm, VON and other meetings

I'm leaving for northern California early tomorrow (Tuesday) to participate in eComm 2008 in Mountain View March 12-14 and then Spring VON.x in San Jose March 17-20.  I'll be back in the office Friday March 20th.

At eComm 2008, I'm speaking on Thursday and moderating on Friday:

• Thursday, 10:20am:  Own the Network — A Radical Approach to Internet Connectivity
• Friday, 2:00pm:   Panel: What Will Drive Wireless Innovation?

The Weekend

Most of my weekend is uncommitted, as yet...  suggestions?

Then at VON, I'm speaking on Tuesday and Wednesday and running a session in the "Unconference."

• Tuesday, time uncertain...:  Unconference session: Rethinking Broadband Internet Connectivity
• Tuesday, 2:50pm:  Panelist in Content Delivery over Alternative Wireless Architectures
• Wednesday, 4:00pm:  Speaking on Wireless Mobility Trends

If you are attending either of these events or live in or are otherwise in the area and interested in meeting, please send me an email using "rbt", i.e. my initials, at nmss.com.

Woman in High Tech — Is it getting better?

Woman's Radio News has a great interview with Dr. Dawn Nafus about her upcoming appearance at eComm 2008.

Dawn Nafus, Ph.D., an anthropologist at Intel, discusses why a technology company would have an anthropologist on staff, and exactly what she does for them. Dawn will be speaking at the eComm Conference being held March 12 - 14, 2008 at the Computer History Museum in Mountain View, CA. Her topic will be “Context Aware Technologies” and how they can assist different cultures and countries around the world.

About 4:30 minutes into the interview, Pat Lynch asks Dr. Nafus why there are only a few women on the program at eComm and indeed at most high tech conferences.  She doesn't have a simple answer but she does point out it's a myth that women's position in high tech is getting better gradually over time, at least in Silicon Valley.  Silicon Valley is very young in the grand scheme of things, it has little or no history.  And yet, it has reproduced the male dominated culture that was a characteristic of older industries.  Now older industries are improving at a greater rate than high tech.

I just recently read Herman Goldstine's classic history of the early days of computing, The Computer from Pascal to von Neumann.  Interestingly, some women play key roles, not just Ada Byron (Lady Lovelace), but multiple women during and after WWII.  And when I think back to the Association for Computing Machinery (ACM) that I joined as a student in the early 1970s, there were more women involved in programming, and the mathematical side of the computer industry than in traditional industries.  Also my first part time job was with a small company where 25% of the software staff (1 out of 4) were women.  :-)

Recent counts by VC Christine Herron of women at high tech conferences range from 13% to 18%.  Many old line industries have changed.  But not high tech?

I look forward to hearing Dr. Nafus speak on "Context Aware Technologies" at eComm this coming week.  Hopefully I'll also get a chance to talk with her, as she mentioned some references to recent literature on women in high tech.

Broadband Access — What should we regulate?

Network Neutrality is a hot topic in the US.  The FCC held hearings in my neighborhood recently (while I was in Asia).  Now I see Professor Susan Crawford will be testifying next Tuesday at a House Judiciary Committee hearing on "Net Neutrality and the First Amendment."  I look forward to her remarks, but I worry that the whole discussion will be focused on "IP Pipes," that is connectivity at network layer 3. This distracts us from the fundamental problem.  Imposing network neutrality is patching an obsolete regulatory system. We must be careful what we ask for. Despite the best of intentions, laws and regulation can easily produce unexpected results, as I've discussed here & here & here.

More importantly, laws and regulation evolve slowly with major changes taking years, even decades, so it's critical to understand first principles and what makes sense for the long term.

Consider the useful life of various network elements involved in fixed broadband connectivity.

This diagram says a lot about what can be efficiently regulated and where one would prefer competition.  Notice the logical dividing point is NOT at layer 3 — IP Routers.  It's at layer zero — Dark Fiber.

This suggests a regulated dark fiber provider and multiple competitive providers of access, connectivity and other services, operating above the dark fiber layer.

Is there a natural monopoly, and if so, where?

Local rights-of-way (ROW) are a bottleneck.  Whether they are owned by the state, the municipality or a neighborhood association, or are deeded arrangements between adjacent property owners, local rights-of-way are a limited resource.  Installing poles and conduits in these rights of way results in visual blight and/or construction mess, of concern to the owners of the rights-of-way.  And a given set of conduits holds a limited number of cables, although each cable may have very many fibers.

Between major US cities, we find competing cables in competing rights-of-way and a vibrant market where dark fiber strands are available as well as every conceivable higher level service.  That's not the case in front of my house where there is just one city-owned street and one customer (me), or in the "middle mile" between my neighborhood and the nearest major communications hub where there are a limited number of routes and a limited market (me plus my neighbors).  So, at least for local and middle mile dark fiber, some kind of communal, condominium, municipal or regulated monopoly provider makes sense.  But above that level?

Certainly telephony is not a natural monopoly.  Just look at long distance and VoIP pricing.  The only place where telephony has significant per-minute cost is where it's been granted a statutory monopoly.

Lighting the fiber need not be a natural monopoly

A fiber is only useful once it's lite, but the person or company that lites it has the potential to control what goes over the fiber, a distinct problem if there's only one dark fiber per home.  There are at least two solutions:

1. regulate the services of the company that lites the fiber,
2. give control of the fiber to the homeowner, allowing them to pick which ISP lites their fiber.

With option 1, all homes get the same speed electronics and are likely to be locked at that speed until the whole neighborhood is upgraded, perhaps years after the technology becomes available.

With option 2, each home owner gets to select a service provider and the speeds and services they desire.

Which would you personally prefer?  Which would you like to see adopted in the US?

What's happening in the rest of the world

1.  Condominium fiber.  Developers, or contractors working with groups of interested parties, lay fibers and sell individual fibers on a condominium basis.  There's a good overview of the subject here and there are success stories like these from Quebec.

2.  Municipal dark fiber.  The government organizes fiber deployment, making dark fiber available to all comers.  Stokab AB in Stockholm is a very successful example of this model.

Dark fiber is no more complex for the homeowner or businessman.  They need only select an ISP to light their fiber. The point is to make it easy for many, many providers to compete to offer services to individuals and businesses using the latest and greatest electronics.

Is anything like this possible in the US?

We have an existing, highly obsolete, regulatory system in place and therefore many vested interests, but if we can articulate the desired end goal, the compelling reasons why it makes sense and proof points like broadband in Sweden, anything is possible.

-----------------------------------------------------------------------------------------

I'll be speaking on this subject at eComm 2008 next week (that's in addition to my panel on Wireless Innovation) and I expect to be discussing these subjects the week after in the Unconference track at Spring VON.  If you will be at either event, please join in.

T-Ring: Audio logos promote mobile operator brands

It's not a new idea.  Short audio "logos" have been used to identify radio shows, TV networks, long distance phone companies (during the long distance wars in the US in the 1990s) and, most recently, podcasts.   But mobile operators have missed this opportunity, at least so far.

Now SK Telecom has introduced the idea in Korea and a number of other mobile operators are scrambling to leverage the idea.  In Korea it's called T-Ring and it's both a branding exercise and a key part of a special pricing bundle.

When you call an SKT subscriber, the first thing you hear is a 1.5 second sound clip — the SKT audio logo.  Then you heard the called subscriber's ringback tone (assuming they are among the 55% of Korean subscribers who pay for ringback tone service) or you hear normal ringing signal, until the subscriber answers.

Everyone hears the T-Ring sound, letting them know they are calling an SKT subscriber.  Even better, SKT has introduced a pricing bundle where subscribers get 50% off on calls to other SKT numbers.  As a Korean friend puts it,

Of course it is nice to hear the sound before the connection is made because your brain is naturally set to read "Oh 50% discount."

This is also good for NMS Communications, as the equipment is similar to that used for ringback tones and we already provide components and platforms to most of the major ringback tone suppliers worldwide.  :-)