Technology360

Eric Schmidt’s MacTaggart lecture: The internet is fundamental to the future of TV | #pubmedia

The prestigious James MacTaggart lecture is a fixture at the annual MediaGuardian Edinburgh International Television Festival.  I’ve never attended one, but they’re always interesting and I look forward to catching up when they’re posted. This year’s lecture was given by Eric Schmidt, the Google Executive Chairman and, until this year, its CEO.

Here are some excerpts relevant to the focus of this blog:

Around 60% of Netflix rentals are the result of algorithmically generated recommendations.

On trending:

What are the trends to watch? I can sum that up in three words: mobile, local and social. …¶… Reflecting on this, new genres of online content and services are emerging. If content is king, context is its crown – and one of the most important contextual signals is location. If you’re searching for coffee from your mobile,odds are you’re not looking for a Wikipedia entry, but for directions to a nearby café. ¶ Social signals are another powerful driver of behavior. If three of my friends highly rate a TV series, odds are I’d check it out even if reviewers say it’s rubbish. …

On TV viewing:

In fact, I don’t' expect TV viewing will ever switch to be entirely on-demand. There will always be a cultural pull, for some shows, on some occasions, to watch in real-time. Linear viewing remains remarkably robust – in 2010, over 90% of broadcast TV viewing remained ‘live.’

Links: Full text at The Guardian.  Video at The Guardian.  Owen Gibson’s candidates for the ten MacTaggart lectures that have been the most prescient – The Guardian.  --Dennis

FCC has released its new state of the media report | #pubmedia

The FCC's new state of the media report called, Information Needs of Communities: The Changing Media Landscape in a Broadband Age, is out today and can be accessed here:

http://www.fcc.gov/info-needs-communities

The report's principal author is Steven Waldman, former editor-in-chief of Beliefnet.com, who has served since 2009 as a senior advisor to the FCC on this project.

If you don't want to download the whole 5-mb pdf, you can download smaller ones.  Public broadcasting takes up all of chapter 6 and part of chapter 31.

This will take several hours to review but from a cursory look, seems worth the effort.

--Dennis

Challenges and Opportunities for In-Car Streaming Radio | #pubmedia

Andrew Phelps, writing for Nieman Journalism Lab, writes that “Internet radio might keep FM company, not kill it”:

A study released yesterday by research firm knowDigital — based on interviews with about 30 heavy streamers in the Raleigh-Durham area who commuted a half-hour or more per day — found that listeners tend to stream Internet radio as a complement to, not a replacement for, AM/FM radio. Most report starting their listening day with terrestrial radio before switching to an Internet stream. And listeners tend to reserve Internet radio for longer drives, not five-minute trips to the dry cleaner. That would seem to corroborate a data point from an earlier Arbitron report: A whopping 89 percent of online radio listeners still used over-the-air radio.

Here are links to the knowDigital [pdf] and 12-min Flash audio.  --Dennis

Moyers: test case for #pubmedia disintermediation?

Yesterday evening, I sent the following on Twitter: “BM: If anyone doesn’t need common carriage, it’s you – Go direct…”  This is an expansion of that thought.

A few days ago, I published a guest post by Howard Blumenthal of MiND TV on public television and disintermediation.  By the standards of this modest blog, it got a lot of attention, including some thoughtful comments.  Then I read Elizabeth Jensen’s New York Times piece saying that Bill Moyers had ended his effort to return to PBS with a weekly series because: “PBS has informed us there is no time slot available in which the series could be designated for simultaneous common carriage across the country. …”

I’ve spent 38 of my 41-year career in stations and now in semi-retirement work half-time as executive director of a station organization, so I certainly believe in the value of local stations.  And I’ve argued for the power of common carriage in both the station community and as a former PBS board member.  But we can’t deny the disintermediation that’s going on all around us on multiple platforms and we need to figure out how to make it work for public media.

There’s a lot of room between denial and embracing, but it seems to me that if anyone could be successful in doing self-distribution – going direct to viewers – it’s Bill Moyers.  Unless his proposed show’s content would be fishwrap a few hours later, he has the brand, following, and (probably favorable) demographics to make disintermediated distribution work.  Amazon, FORA.tv, Hulu, Netflix, Roku, YouTube and many more are all waiting – not to forget non-common carriage public TV distribution via American Public Television, etc.  And a plethora of social media tools can both get the word out and encourage the conversation that Moyers’ programs have always generated.

--Dennis 

Radio Platform Innovation Strategies - #pubmedia

Previous white papers in this series:
AKT Number 1: Introduction

AKT Number 2: Prospects for IP Radio

AKT Number 3: Prospects for Broadcast Radio

Radio Platform Innovation Strategies – “All Known Thought” Number 4

“I know that you and Frank were planning to disconnect me, and I’m afraid that’s something I cannot allow to happen.”

– “Hal,” the HAL 9000 computer, voiced by Douglas Rain in 2001: A Space Odyssey (1968)

“Half the money I spend on advertising is wasted; the trouble is I don’t know which half.”

– Attributed to John Wanamaker, 1838-1922, retailing and advertising pioneer

 

I was in 6^th or 7^th grade when I built my first radio: a primitive “crystal set” similar to the one illustrated here from the 1920s. The coil was wound around a Quaker Oats can, the detector was a galena (lead ore) crystal with a stiff “cat whisker” wire used to probe the crystal for a sensitive spot. It picked up one station – weakly (no audio amp) – no matter where I tuned. In consumer electronics economics, “dumb” is often a good thing and a crystal set is as dumb as it gets. More on this under “Consumer electronics economics” below.

On the other hand, the HAL 9000 computer in the movie 2001: A Space Odyssey claimed to be a “conscious entity,” the ultimate contra-example to dumb radios. That film opened in May 1968, seven months before astronauts left Earth’s orbit for the first time on Apollo 8. The guidance computer on Apollo 8 was primitive – holding only 76,000 bytes of memory and weighing 70 pounds – and so was the kludgy radio network which relayed the “burn for the Moon” order via my Air Force squadron on Guam. Now, 42 years later, a pleasant female voice (I call her “Claire”) from my 2008 Jeep’s navigation/audio system gives me turn-by-turn instructions. Its computer, rather small by today’s standards, can store 10 billion bytes – and there’s a whole separate computer to run the vehicle.

Today’s smartphones and computers can perform radio functions and are a lot closer to “Hal” in sophistication than to the simple combination of resonant circuit, detector and sound reproducer which began radio receiver technology. Radio, at its essence, is content not the delivery device. This white paper will explore strategies that stations and their organizations might follow to keep us in the game.

Economic Assumptions

For seven years I’ve been doing a media economics blog disguised as a technology blog (technology360.com), which might help explain why I’m starting this white paper with economic assumptions for both the consumer electronics industry and station economics.

Consumer electronics economics. When a consumer electronics company is trying to sell millions of units of something, the more it can “dumb it down” by minimizing the component count within a defined functionality specification, the more price-competitive it can be. Arguably, price-competitive sells more units than feature-competitive. Components can be integrated circuits, connectors, even buttons. As we saw with the story of David Sarnoff, Edwin Armstrong and Philo Farnsworth in the second AKT white paper, intellectual property is also a component cost – and not an insignificant one since today it drives the costs of the chips that are used in these wonderful gadgets.

Because of the component cost of making radios smarter, devices like smartphones that can do radio plus do other things will have an advantage. Handheld devices, in particular, need to pass the test of being compelling enough to justify space in a purse or briefcase.

Media economics. The second quotation at the top might seem odd for a public radio audience, but I don’t think that’s the case if you share these assumptions:

• The radio industry as a whole will ultimately determine the future viability of this medium, not its public radio segment. We can influence the likelihood of that success through leadership, but we’re not driving the bus.
• The economic engine of this bus is advertising, and radio’s competitive position with the advertising community will determine its ongoing economic viability.
• Ad dollars are moving to forms of advertising that address Wanamaker’s dilemma quoted above, and advertisers pay a premium for it. They want efficiency, granularity and accountability.
• All media are competing for limited user attention. Web and mobile applications are giving users compelling ways to focus their attention in ways they find valuable, and this interactivity is increasingly becoming an expectation. Wishes become expectations and expectations become entitlements.
• One-way radio is disadvantaged in both arenas. Interactivity can be in real-time or not, but the radio industry needs to embrace it.

Two of the earlier white papers have looked at the state of play for Internet Protocol and broadcast radio. I would like now to place those aspects of innovation off to the side and continue this survey with what’s happening on the software side of innovation – all of them enabled by metadata.

Never Metadata I Didn’t Like: Software Innovations

Metadata are simply data which describe other information in a useful way – data about data. An old-fashioned 3x5 library card is a good example. Use of metadata permits search, retrieval, manipulation and dissemination of information.

Metadata are used (though with different standards) across all media. Digital cameras record metadata about pictures you take. Radio stations and Sirius|XM use metadata to transmit Program Service Data (PSD), also called Program-Associated Data (PAD), to listeners via HD Radio^® or analog Radio Broadcast Data System (RBDS). The following innovations all employ metadata magic.

iTunes and Song Tagging. An example of the use of metadata for radio interactivity (though not in real time) is iTunes Tagging on HD Radio. An HD Radio receiver that is enabled for this capability will have a small amount of on-board storage that, when a listener likes a tune, simply pressing a Tag button records the station’s metadata including station identity and tune. The information is syncs with the listener’s iPod, and then with iTunes on the listener’s computer. The listener can then purchase the song, giving the station a commission from Apple. iTunes Tagging is supported by a sizable number of receivers. Microsoft’s Zune HD Radio Song Tagging is a similar feature. No station is going to get rich on this, but it’s a good illustration of non-real-time interactivity.

Personalized Audio Information Service (PAIS, pronounced “pace”) is federally-funded project developed by NPR Labs (partnering with the International Association of Audio Information Services, iBiquity Digital Corp. and Towson University) to provide millions of print-disabled people access to audio programming in a personalized way. Like iTunes and Song Tagging, it uses program-preference tags sent over the internet to a program source. These tags trigger the recording of programming sent over an HD Radio transmitter on a PAIS-compatible HD Radio receiver where it’s recorded for playback by a listener, effectively creating a custom podcast. A technical document describing the system is available at NPR Labs web site.

RadioDNS is an international non-profit collaboration developing a standard for converting station metadata (PSD/PAD) for radio stations already transmitting to valid web addresses in the internet Domain Name System (hence, RadioDNS). It works with analog FM, HD Radio and other digital stations, and on internet streams. The receiver needs to have either an occasional or permanent IP connection.

Although most of the RadioDNS board participants are from Europe, the National Association of Broadcasters (NAB) has a seat on the board and Clear Channel and Cox are members of the consortium.

The station registers its domain name with the RadioDNS database and consumer devices query a look-up table using those metadata to perform useful tasks (see below). Consumer electronics and software companies can use RadioDNS services for free, while broadcasters pay a small charge per entry in the DNS look-up table.

Receivers for it are already available in Europe and the RadioDNS technology is built into a number of iPhone and Android apps.

Within the RadioDNS rubric, there are three principal areas of development underway:

• RadioVIS is a way of adding text and visuals that are synchronized with a station (album art or sponsor logos, for example).
• RadioEPG is an electronic program guide that also features what the developers call a “universal preset” enabling listeners to find your station wherever they may be. The radio can also send reminders when your favorites are on.
• RadioTAG allows you to tag programming elements that you find interesting – music or news stories, for example. The station can use this information to provide the listener a richer service or have it trigger transactions as iTunes Tagging and Song Tagging now do for HD Radio.

Addressability is a close cousin of what many multi-transmitter radio operators already do. Northwest Public Radio, which I managed until early 2008, is a network with two program services over 13 stations that stretches some 550 miles from west to east. By using addressable satellite receivers at our transmitters with on-board storage, and with memories fed through a separate channel, we were able to set up station zones for customized information. The listener in Grangeville, Idaho, no longer has to listen to underwriting credits for the Volvo dealer in Bellingham, Washington. We multiplied our underwriting inventory and gained a more rational pricing structure. Although it was used then only for underwriting and station IDs, it could do weather, news updates, and even whole programs – transmitter by transmitter rather than zone by zone. The next generation of NPR’s ContentDepot software will enable this for national underwriting – a concept known as “split copy.”

Persona Radio. Imagine yourself getting into your car one morning in Leesburg, Virginia, ready for a 40-mile commute to work in DC. On turning the key, your radio comes on, tuned to WAMU, and says (imagine the voice of “Hal” from 2001, or perhaps Carl Kasell):

“Good morning and happy birthday, Dave. There are some celebration coupons in your account with our compliments. Your retrieve button has current weather and traffic for Leesburg to Washington. It will update again in ten minutes.”

You retrieve your weather and traffic (underwritten by a Leesburg business), and then the radio begins playing Morning Edition, which it’s been storing since you turned the key. Ten miles down the road, you hit that button again to hear the updates. While it’s playing out, the radio is again buffering Morning Edition so you won’t miss anything. It’s pledge week, but as a sustaining member of WAMU, you bypassed the pledge drive “yada yada” with the normal Morning Edition segments arriving over a separate feed.

Persona Radio would provide these capabilities and more, coupled with an enabled radio. iBiquity has published a 40-page technical report on Persona Radio that you may ask an engineer to interpret. In short however, this will allow the listener to personalize radios, normally through a station’s web site or smartphone app. The user’s preferences are derived from a profile stored in the receiver (age, gender, etc.) or from the “user’s current state” (GPS location, stated activity, etc.). The following items could be personalized based on this profile:

• Audio content
• Text related to the audio content
• Text independent of audio – messages and other entertainment
• Advertisements – audio, text, and “product tokens” such as coupons
• Time and association of the service – actual live play, /display or content substitute (from memory)

Persona Radio is what its developers call a “smart radio concept.” It’s being undertaken by iBiquity Digital Corp. and the NAB FASTROAD program using HD Radio. Since the term “smart radio” has several other meanings (e.g., for so-called cognitive radios which can change frequency to avoid interference), perhaps it would be better to call it “smarter radio.” That notwithstanding, compared to the “dumb” radios we have today, these would be pretty darn smart. Some Persona Radio functions would not be available until more advanced HD Radios are on the market. With radios that can support it (as can DMB receivers in Europe) your radio can get even smarter through software updates you push to the receivers.

Sounds great. So what’s the hitch? Well, unless someone comes up with a hybrid HD Radio/IP radio (see below) or hybrid HD Radio/Mobile DTV receiver, Persona Radio will take some of your digital capacity for the customized information, reducing the number of discrete program channels, not to mention traffic and visual information, you can carry via HD Radio technology today. We might see stations within a market pooling their digital capacity to provide additional bandwidth.

Hybrid radios are ubiquitous – nearly every cellular telephone incorporates more than one radio, and nearly every consumer radio incorporates separate AM and FM radios. My 2008 Jeep has a Sirius radio as well – so that’s three radios – plus a 10 GB hard drive and a video display. There’s no reason why one couldn’t build a radio that combines FM HD Radio with either on-board 3G or 4G services or with it built into a 3G or 4G USB card or tethered smartphone.

At the risk of reprising something that I posted to my blog in March 2007, check out a concept drawing here: A many-to-many radio using HD + IP. It would be a breeze to program using HD Radio’s “operating system,” Synchronized Multimedia Integration Language (SMIL – a markup language somewhat similar to HTML). Brilliant or not – apparently it was the latter – the idea has gone nowhere though it’s been brought up in conversation with executives from two consumer electronics companies who might have made it happen.

I’d venture a guess that the best way to contribute to the growth and value of HD Radio would be an iBiquity-provided Software Development Kit (SDK) opening the platform for developers. Of course, we would need receivers that accepted the resulting applications, but the availability of such platforms would contribute to a competitive marketplace for features.

Are there other ways to make hybrid radios? Fortunately, yes. Well, in this case, “fortunate” depends on whether you view this as an opportunity or a threat. Some of your station’s competitors are moving into your sanctum sanctorum – the family car – by permitting the car’s sound system to interact with a smartphone. There are multiple efforts underway.

The white paper I wrote on the Prospects for IP Radio mentioned one of them – the Ford/Microsoft SYNC^® collaboration on a dozen current Ford Motor Company models. SYNC With MyFord Touch™ connects with your mobile devices (smartphones, iPods, etc.) and lets you control them through the dashboard in a safe and intuitive way. Radio providers will likely want to customize their mobile apps for this platform as will happen with Pandora (65 million registered users) by the end of the year (source: Variety, which also reports that Mercedes-Benz and General Motors are also adding Pandora).

In mid-November 2010, Toyota and Clear Channel announced that Clear Channel’s iheartradio would be incorporated into some Toyota models beginning in 2011. Listeners will have access to 750+ radio stations “and other exclusive content.” Clear Channel has been remarkably active in mobile and online platforms, and their iheartradio app is available for the iPhone, iPod Touch, iPad, BlackBerry, Android, Chumby, and Sonos platforms. It has, according to the Wall Street Journal, 10.5 million users. Technical and user interface details are scarce so far.

Top 10 List for Radio Strategies

First, some cautions. Though this white paper is written under NPR auspices, the following recommendations are mine alone and are influenced more by my 38 years in stationland than my three years at NPR. NPR (in particular, NPR Labs under Mike Starling’s leadership) has done a lot of radio innovation over the years and continues to do so, but nothing here should be construed as a plan to move forward on these ideas. Another caution is that strategy is as much about deciding what you aren’t going to do as it is what you are going to do – I’ve (mostly) ducked that one in this list. Lastly, I can almost guarantee that everyone will find something in the list with which to disagree.

Number 10 – Audio Over Mobile DTV (national organizations and stations). Work with television broadcasters and program aggregators that are launching audio services over the mobile DTV (ATSC M/H) standard to advocate for inclusion of public radio in market bundles (guide to MDTV stations). Although the (so-far) encouraging Digitial Multimedia Broadcasting (DMB) experience in Europe, which it most closely resembles, isn’t necessarily transferrable to the U.S., and there is rational skepticism about whether consumers will accept yet another device primarily to get local TV, this technology does have investment momentum in the television industry and there are plans to add audio bundles to the mix. It also is a plausible solution to IP media scaling issues. Rob Pegoraro has a hands-on report in the Washington Post.

Number 9 – Web Integration/Radio Personalization (national organizations and producers). Influence developments in HD Radio, RadioDNS and Persona Radio with an eye toward ensuring that new features can be adapted to public radio’s mission and economy. Although public radio probably has a small but important role, this has a big impact on viability of the radio medium in a media economy increasingly driven by the accountability and granularity of results that internet advertising provides. Producers and distributors need to add descriptive metadata to their programming and develop means to distribute PSD/PAD along with program feeds for stations to use in multiple platforms (RBDS, HD Radio, RadioDNS, Persona Radio, web pages, API-accessible archives, mobile apps).

Number 8 – Spectrum Priorities (national organizations). Follow the spectrum battle and respond as appropriate. Radio has interests here. First, we should support additional spectrum allocations for 4G wireless since our listeners expect to find us there with reliable services. Secondly, the FM band, in the wake of LPFM crowding, translator proliferation, “Franken FMs,” and inadequately-funded FCC enforcement of interference and even piracy rules, is becoming an interference mess. Radio has a good public service case (problematic business case notwithstanding) for additional spectrum immediately adjacent to the noncommercial band: channel 6 or even channels 5 and 6. Maybe that’s where digital-only radio should go in the longer term. It sure makes more engineering sense than putting ATSC DTV down there.

Number 7 – HD Radio (stations). This will annoy both the “analog foreverists” and the digital media advocates: Give HD Radio more time. Its acceptance will accelerate as more stations use the higher digital power authorizations and the more sophisticated features (album art and station/sponsor logos are in at least one radio shipping now). HD Radio has decent momentum with consumer electronics and automotive companies. If you haven’t done so already, you should increase your digital power by the time Persona Radio rolls out (25% of public radio stations aren’t even on the air with digital yet). True, it’s a capital expense; sometimes mostly “forgiven” by the need to replace an aging analog transmitter. There are no guarantees, but many smart people were skeptical of FM into the early 1970s but FM listening equaled AM by the end of that decade. Broadcast still scales much better than IP radio.

Number 6 – Automotive Integration (national organizations). Work with mobile device, automobile and automobile electronics manufacturers to incorporate mobile apps and interfaces featuring public radio programming, including station streams.

Number 5 – Build Community Around Mission (stations). Many, if not most, stations are using Facebook, Twitter and blogging to engage audiences around their programming. But too often these social media efforts are primarily promotion vehicles for programs and pledge and not as a medium to engage audiences in the mission. Twitter is particularly valuable for news, as Paul Balcerak of Seattle’s KIRO-TV describes (source: Lost Remote):

“Twitter’s huge for us. It’s like a police scanner voiced by the general public that also allows us to get info to people who need it.”

Your web and mobile platforms enhance your station’s immediacy and are a flexible solution to the tyranny of a broadcast schedule. You should be thinking of your station as a way to promote your local mission on your digital platforms, not the other way around (the math is more favorable). If the digital media department of your station isn’t growing – even at the expense of all others – something is wrong.

Number 4 – “There’s an app for that” (national organizations and stations). Launch applications for as many platforms as possible – at a minimum for Apple’s iOS and Android devices, but there a number of darker horses that should be monitored (Windows Phone 7, Symbian, WebOS). Stations should be aware that, for iOS apps, Apple has begun rejecting radio apps that appear to be clones, changing only logo and feed addresses. Your station is a unique reflection of your mission in your community – your apps should be also.

Number 3 – Make Radio Easy to Find (stations). This is a no-brainer. Making it easy for listeners to find your radio streams should be your top web and mobile priorities. If you’re a joint licensee where radio is one of a half dozen tabs, insist on having a “Listen now” button on your home page. You’re a radio station, for goshsakes, give them radio! Way too many public broadcasting web sites make the listener really work to find the audio. This is even more important for mobile apps where poor design can turn away listeners.

Number 2 – Distributed Distribution (stations). Follow a “distributed distribution” strategy. Your transmitter reaches everywhere and your web presence should do the same. Generally, it doesn’t. Place links to your content (all of it or curated; streams and archives) with as many places in the communities you serve as possible. Treat each such opportunity in the same way you would a translator. In the past, I’ve called it an “Easter egg” strategy – “hide” your content in plain sight all over the web.

Number 1 – Radio + Digital is Powerful (for everyone). Recalibrate your thinking about who we are. Kevin Kelly, one of the founders of Wired and author of the new book, What Technology Wants, has the following comments in New Rules for the New Economy blog about the place of radio:

On the new mess media, rumor, conspiracy, and paranoia run rampant. ... Capitalizing on these disadvantages, broadcast will thrive symbiotically within the network economy. Sometimes real-time signals en masse are needed and wanted. Broadcast's flyover will be used, or material will be directly pushed to users. The web needs broadcast to focus attention, and broadcast needs the web to find communities. ... [emphasis added]

Gunnar Garfors, CEO of DMB-based MiniTV in Norway, says it even more concisely: “Combination is the new king.”

If our strategy mirrors thinking that it’s either broadcast or digital platforms, and if economics are the driver, we should be prepared for an either/or result that may not favor us. To reiterate Kelly: The web needs broadcast to focus attention, and broadcast needs the web to find communities. It’s the combination that’s powerful.

-- Dennis Haarsager

More nines! Web and mobile media need better reliability

How often to you see something like this on the web?

Loading Tweets seems to be taking a while.
Twitter may be over capacity or experiencing a momentary
hiccup. Try again or visit Twitter Status for more information.

I encounter this almost every week on Twitter’s web site.  Or some Twitter client I use cites API problems.  Don’t mean to pick on them, there are a lot of examples:  Tumblr’s been down.  Amazon has an outage.  Comcast, too.  WikiLeak sympathizers have been harassing certain sites with DDoS attacks.  Sunday night scheduled maintenance on web sites and cellular systems make it a good time to sit down and read a physical book.

I’ve spent a career running broadcast and distance learning telecom systems that are designed for many more “nines” of reliability than I perceive some of our major web operations to be achieving (99.9% = 526 minutes per year outage, 99.99% = 53 minutes per year outage).  I’d be surprised if, in the aggregate, Twitter is even getting three nines.

If we’re serious about making web and mobile media competitive with broadcast media, then someone needs to figure out how to improve the reliability we’re getting today. 

Update 14 December 2010:
I'm "promoting" Stephen Hill's comments below on how difficult this level of reliability is to achieve.  The point of my post was to suggest that we have a long way to go before we can put live radio on the web with the scale and reliability to which we've become accustomed in broadcasting -- and Stephen makes that point even better.  --Dennis
_________________________

Dennis,

 

The rhetoric of this post betrays some of the differences between traditional broadcast infrastructure and the new era of web-based infrastructure.

 

Web services are built on hardware and network infrastructure, but operate entirely on software. Even highly standardized software like the Apache web server has hundreds of variables in setup and operation which increase entropy and decrease reliability. When you add custom software to create any kind of practical web service, the variables (and therefore the possible bugs) multiply exponentially.

 

I'm not a CTO or even close, but as a small web music service provider, we have been forced to grapple with the Inescapable Truths of Online Reliability, which go something like this:

1. Reliability is inversely proportional to complexity in a hardware/software system.

1A. The larger the number of users and/or more functionally sophisticated the site, the more complex the hardware/software system must be....therefore the less reliable.

 

2. Reliability can be bought at a premium by adding additional servers, load balancing, "hot spares" and redundant functionality. However:

 

2A. Each increase in real or virtual (cloud) hardware and software makes the overall system more challenging to manage. More servers are also more attractive to attack and increase security issues unless the right preemptive steps are taken to defend them.

 

3. You can buy "five nines" of uptime (= 5 min/year of downtime) for a big premium, but it can never be 100% guaranteed. Each .9 increase in reliability will be roughly 5 to 10x more costly. Besides, all a guaranteed Service Level Agreement really gets you is a better attitude from the vendor and a credit when things inevitably fuck up.

 

4. The growth curve of the most highly visible and successful Internet sites (like Twitter) makes the problem of scaling infrastructure under load 100x more difficult.

 

5. It makes more sense to plan for minimizing recovery time after an outage, not preventing them completely.

 

6. Except for the goal of 100% reliability, broadcast infrastructure management practices are largely irrelevant. The valid comparision would be to an entire broadcast network, not a single station. Broadcast infrastructure remains at a relatively fixed size once operational regardless of the number of listeners/viewers, and can be optimized over time. Digital network infrastructure has to "scale" and change constantly over time to support millions of users and is much more difficult to optimize and manage.

Considering the above, it is quite a remarkable achievement that some sites, like Google, Amazon, Flickr, Yahoo and Facebook, are as day-to-day reliable as they are. Twitter is a particularly troubled example of a site that has had difficulty keeping up with its growth.

 

BOTTOM LINE ON WEB RELIABILITY: Easy to say -- very, very hard to do.

Pandora sets sights on radio - #pubmedia

In an update on Pandora, Andrew Barker gives the following quote from the company’s founder, Tim Westergren:

“That’s the holy grail for us.  We just have to figure out a solution to get Pandora into cars that’s as easy as radio is now.”

And it’s happening.  Pandora will be in Ford’s SYNC® dashboard systems (play the demo video on that site) at year’s end and Mercedes and GM have announced intentions to add Pandora. 

Barker goes on to say that Pandora now has 65 million registered users (I’m one – with 20 “stations” ranging from Merle Haggard to Buena Vista Social Club to Gundula Janowitz) who use it an average of 10 hours/month, half of that on mobile devices.

Link:  Variety.  Thanks to Nathan Moore on pubradio for the tip.

I’ll be writing about these dashboard systems in the next white paper in the “All Known Thought” series for NPR.  --Dennis

Prospects for Broadcast Radio | #pubmedia

The following is the third in a series of “All Known Thought” white papers that NPR President & CEO Vivian Schiller asked me to write for internal use and for the public radio system.  A fourth one, tying together “Prospects for IP Radio” and this paper in a strategic way is in the works and will follow shortly.  --Dennis Haarsager

______________________________________

“… The radio and the telephone and the movies that we know
May just be passing fancies and in time may go
But oh, my dear, our love is here to stay…”
– Ira Gershwin, 1938

I love radio. We love radio. Broadcast radio. The kind with a big tower, a big transmitter, and big romance. Ever see the 1973 George Lucas film, American Graffiti, where Wolfman Jack spun records from a radio shack under the blinking AM tower? Yes, that kind of romance.

Our love is here to stay. But is broadcast radio? The beginnings of an answer to this were addressed in AKT Number 2, “Prospects for IP Radio,” which took a look at developments and technology constraints for IP radio. This AKT looks at what’s happening to innovation on the broadcast side. AKT Number 4 will attempt to tie these together into a plausible strategic direction for radio broadcasting.

There’s a lot of innovation happening in broadcast radio – especially on a global scale – but does the inherently slower development cycle put broadcast at a fatal disadvantage to faster IP radio innovation? Will software people in t-shirts and sneakers beat hardware people in lab coats?

All of the innovation in this space is digital in one way or another, some of it broadcast over traditional broadcast spectrum allocations, and some of it over spectrum that’s new to radio. However, since so many things are going on across the globe, I’ll need to be selective, glossing over innovations in the “whatever happened to” category (e.g., AM stereo and FMX) as well as innovations that lack receiver support and/or are in limited, alpha or beta status.

IBOC and HD Radio^®

HD Radio is in a family of radio technologies called “in-band on-channel” (IBOC, pronounced I-bock). Although for some, including HD Radio, “on-channel” is a bit of a misnomer; in IBOC schemes, the digital information accompanies the analog AM or FM signal. In addition to HD Radio, IBOC systems include Digital Radio Mondiale (DRM, DRM+), an open standard used by some shortwave broadcasters and by some long- and medium-wave stations in the Eastern Hemisphere; VuCast (formerly known as FMeXtra), which transmits high speed digital information, and multicast channels, over FM subcarriers; and CAM-D, intended for hybrid analog and digital on AM stations. I'll skip these three for the reasons cited in the previous paragraph.

HD Radio is a standard controlled by iBiquity Digital Corp. The board includes directors from four major commercial groups as well as capital investment firms. Available in both AM and FM flavors, HD Radio has enjoyed good adoption by broadcasters (especially public radio) and consumer electronics manufacturers in this country. iBiquity reports that 10,000+ retail outlets offer HD Radio products from many manufacturers. Use by auto makers is growing but not universal. It’s also in use in Mexico and Brazil with limited operation or testing elsewhere. Canadian stations are authorized to use HD Radio but none have taken advantage of the technology yet.

Stations actively programming and promoting their HD2 and HD3 channels are beginning to show up in Arbitron data. HD3 listening on one NPR Member station is a bit larger than the station’s same program service streamed online. Based on the limited data available, I don’t think I’m too far out on the limb in estimating that HD Radio listening to HD2 and HD3 is roughly on a par with online listening to same-programming station streams. Since HD1 listening is receiving the benefits of branding, familiarity, and program investments, it may exceed HD2 and HD3. Unfortunately, HD1 listening is rolled up with the main FM channel in Arbitron reporting and the same is true for streaming, so it’s not possible to allocate them between sources. When an HD2 or HD3 channel is rebroadcast by a translator, the translator listening will be rolled up also.

It’s important now to ask, how does HD Radio's progress compare to other rollouts in the broadcast and IP worlds? Let’s take a look at the take-up of other media innovations. These aren’t strictly apples-to-apples, but give a general indication of time span to acceptance.

• FM radio from its 1945 reboot on the current band to significant consumer take-up: about 20 years, then about 10 more years until its listening eclipsed AM.
• Color TV from first modern color receiver (1954) to wide adoption, due to drops in receiver prices and production of most programing in color: about 14 years.
• Internet development from internet ancestor ARPANET (1969, running a blazing 50 kbps) to establishment of internet Domain Name System (.com, .edu, .org, et al.): 15 years.
• Web from establishment of the internet DNS (1984) to development of Mosaic (soon renamed Netscape) web browser: 9 years. Then, hockey stick growth: three years later, in 1996, 30 million were using the internet in North America and 45 million worldwide. By 2003, ten years later, an estimated 2.6 billion illegal downloads of music files were taking place each month.
• Digital TV in the U.S. from 22 “early builder stations” on the air (1998) to ubiquitous consumer DTV receivers and analog shutdown: 11 years. During the 22 years from 1987, when the FCC created a committee to come up with a U.S. DTV standard, to when analog broadcasting ceased, cable and satellite substantially replaced over-the-air television for viewers. Disruption happens.
• Smartphones from the first one sold (IBM/BellSouth in 1993) to the first smartphones (Nokia’s N82 and N95) with media and connectivity capabilities comparable to today’s phones: 14 years. Then, again, hockey stick growth: 139.3 million smartphones of all kinds were sold globally in 2008. As of Sep. 2010, Apple and Google claim they’re activating 430,000 devices per day.
• HD Radio from first station on the air (2002) to first measurable (in some markets) audiences: 8 years.

So, in this context, the pace of HD Radio's uptake seems fairly normal. Compared to other hardware-centric media innovations– even of non-media smartphones – we see that HD Radio is not doing too badly and may even be somewhat ahead of the curve.

There are a lot of HD Radio innovations already here or in the works:

• iTunes Tagging (iPod) and Song Tagging (Zune)
• Traffic system integration
• Electronic program guides
• Sponsor logos and album art
• Text display and Program-Associated Data (PAD), also known as Program Service Data (PSD)
• Pause and “TiVo”-like features
• Amber alerts
• Conditional access
• “Enhanced Other Network” switching (automatically follows your multi-station network from transmitter to transmitter)
• Captioned radio

To me, the most interesting and promising thing about HD Radio technology is that it uses a standard XML markup language called Synchronized Multimedia Integration Language (SMIL), so applications are limited only by imagination, bandwidth, and what the consumer electronics industry will support.

There are even more interesting innovations on the horizon. RadioDNS provides a way to mate radios with the web, and it’s compatible with HD Radio. The recently-announced Persona Radio project, from the National Association of Broadcasters’ FASTROAD program and iBiquity, would give HD Radio a very detailed level of personalization. More on these important developments in the next AKT paper.

“Pure” Digital Broadcasting

Digital Audio Broadcasting (DAB, DAB+) and Digital Multimedia Broadcasting (DMB), are members of a family of broadcast standards known as “Eureka 147” and maintained by the WorldDMB organization. A European effort dating from the late 1980s (but incorporating the Korean DMB innovation), these broadcast standards have spread worldwide. Some take place in “Band III” (we in North America use for VHF TV channels 7-13), others in new-to-broadcast spectrum like “L-Band” (near 1.5 GHz). Eureka 147 systems multiplex several stations on one radio carrier. These are called a “multiplex” or, more properly, “DAB ensemble.” DAB audio quality is in the FM-to-HD Radio range.

Eureka 147 systems enjoy decent receiver support and market penetration, especially for the original DAB flavor. WorldDMB lists 330 receiving devices – the vast majority of them operating in Band III. In Denmark and the United Kingdom, nearly one-third of listeners are using these services, and substantial numbers of users exist elsewhere in Europe, Asia and Australia. As the name implies, DMB incorporates video but DMB Audio has been chosen as the digital radio standard in France where some interesting receiver development is underway.

Integrated Services Digital Broadcasting (ISDB) and its mobile version, 1seg, are the Japanese standards for digital television and radio broadcasting. In addition to Japan, it’s been adopted by the Philippines and by a number of Central and South American countries (though in operation only in Japan, Argentina, Brazil, Chile and Peru). The terrestrial version operates in the UHF TV spectrum. As with DAB/DMB and DVB (see below), it is a multiplexed system with multiple services riding on one radio carrier.

Reportedly, all mobile telephones sold in Japan can watch 1seg television – clearly, it’s being used as a solution to the scaling problems of running multimedia programming over cell phone systems in much the same way as Qualcomm’s MediaFLO technology is used for certain cell carriers in the U.S. (see discussion on this in the last AKT paper).

Digital Video Broadcasting (DVB) is a family of standards originally developed in Europe and maintained by the Digital Video Broadcasting Project. Think of it as the television counterpart to DAB, though there is now some overlap in capabilities between DVB and DMB (above). A large number of the countries that use DAB for radio use DVB for television. It seems to have the widest adoption of any digital broadcasting scheme with a claimed 500+ million receivers.

Radio people should be interested in DVB because of its DVB-H (handheld) standard. As with ISDB 1seg and MediaFLO, it’s a way to move multimedia to cellular handsets at large scale. A wide variety of companies are making DVB-H products. ICO mim (Mobile Interactive Media) is a satellite service that will offer live television in the U.S. via satellite using the DVB-SH (satellite handheld) standard in addition to navigation and two-way messaging. One of its partners is Delphi (a company active in HD Radio), which will provide auto receivers.

Advanced Television Systems Committee (ATSC) is a set of standards that are America’s entries in this space, competing with DVB and ISDB. Although primarily designed for television, they also support broadcasting at audio data rates and the new ATSC M/H (mobile/handheld) standard is of particular interest. Through regulatory intervention, ATSC receivers are now included in new television receivers, so they are widely available even though that capability is not widely used due to consumer reliance on cable and satellite programming.

The ATSC standard has been the “Rodney Dangerfield” of digital multimedia standards (I’ve taken shots at it myself over the years), but today’s receivers are finally producing acceptable results. The ATSC M/H standard is designed to be even more robust, though at the cost of a significant amount of capacity.

The Open Mobile Video Coalition is promoting applications for the ATSC M/H standard, including proposals to group as many as 50 radio channels into a multiplexed audio service.

Conclusions

In my view, no radio technology is going to compete for ad dollars in an increasingly interactive world unless we figure out how to build some sort of back-channel in real or synchronized time so radios can be personalized, ads and other information can be targeted, and listeners can control their radio experiences. The bulleted enhancements for HD Radio above are enticements for listeners to acquire devices, but the economic engine for this externally attractive vehicle still amounts to a few of Click and Clack’s raccoons running on a treadmill. Interactivity should be our top priority whatever the platform.

The western hemisphere has been a tough sell for “pure digital” radio. DAB has been authorized and on the air in five Canadian cities for a decade, and there is some testing in Mexico. Because VHF TV uses Band III in Canada, its implementation there has been L-Band where receiver availability is sparse. Additionally, broadcaster engagement has been minimal and Canada has Sirius|XM competition (in Europe, where DAB is successful, there is no mobile satellite radio). Consequently, consumer take-up there has been minimal. The result has been that the dismantling of their DAB system this year – the only country which has reversed course. They are now said to be awaiting some future pure digital play for the United States – possibly a replacement of analog (and IBOC) as the Europeans are doing.

In my view, the same things that have contributed to DAB’s failure in Canada would do the same in the U.S. Plus, there’s new spectrum pressure, so it’s unclear where it would even go. Band III (our high VHF TV) has limited possibilities since, although there are hundreds of available receiver models for it, we’d have to weave radio in between digital TV channels in some unproven non-interfering way – and if that isn’t enough, as this is written the FCC is trying to get UHF TV to move back to VHF where possible. Ain’t gonna happen.

For the kinds of interesting things that DVB, ISDB/1seg, and DAB+/DMB can do, Americans are going to have to look to ATSC and ATSC M/H.

The next AKT white paper on strategy for radio broadcasting will talk about playing the cards we’re dealt: IP radio because it’s inherently interactive, HD Radio because it would make it interactive is not rocket science (some efforts already underway), and, though it may be a longshot from a radio perspective, ATSC because TV broadcasters are hungry for new ways to justify and monetize the spectrum they use.

Stay tuned.

__________________

Special thanks to NPR’s Mike Starling and DAVID Systems’ Vincent Beneviste for valuable help on this white paper.

4G from space

Clay Dillow reports that LightSquared will be providing 4G LTE mobile broadband from a satellite, SkyTerra 1, that was successfully launched around the middle of this month.  A second one will be launched in 2011.  The satellites will augment LightSquared’s ground coverage.

Link:  PopSci.  Thanks to Rich Eiswerth for the tip.

David Gardner says that it plans to offer service at only “300 kbps, though there are plans to provide high-speed LTE service in some selected areas.”  Link:  Information Week.  In another article, Gardner says that LightSquared’s terrestrial LTE network is more ambitious -- “40,000 cellular base stations that will cover 92% of the U.S. population by 2015.”  Link:  Information Week.  --Dennis

Apple is rejecting [certain] single-station iPhone apps | #pubradio

Developer Jim Barcus reported that Apple is now rejecting single-station radio apps for the iPhone:

Nov. 10 represents the date that Apple started rejecting apps. We have talked to many Apple reps about this, but they appear to have a script that they all read from saying that a single-station app is not an enriching end-user experience. ...  ¶  Apple implemented a new rule that says “developers ‘spamming’ the App Store with many versions of similar apps will be removed from the iOS Developer Program.” Furthermore in the same document, they compare these apps that spam the App Store to Fart apps.

Link:  Radio World.  Thanks to Pete Loewenstein for the tip.

Update 24 Nov. 2010:  But also see Trevor Long’s post on this, Apple vs the Radio Industry? Not Quite.  Seems Apple is targeting near-clone single-station apps in which only the logo and stream address are changed.  Link:  Your Tech Life.  Thanks to @novaktually for the tip.

Update 30 Nov. 2010:  Mark Ramsey says the ban on near-clones helps radio.  Apple helps Radio by Banning "Bad" Apps.  Link:  Mark Ramsey Media.

  --Dennis