• Digital-first audio

    Two interesting stories from the world of American public radio last week.

    First, the creation of Radiotopia, a new podcast network:

    With $200,000 in funding from the Knight Foundation, PRX will attempt to create a new model for what they call “digital-first audio programming.” The seven shows will make up a collective, helping to market one another, providing guidance on technical issues, and sharing lessons on growing audiences. Radiotopia will provide the framework for raising money through grants or Kickstarter campaigns, selling sponsorships across the network, and sharing revenue between the shows.

    “Digital-first” is an interesting focus, with the potential to shake up the way stories are told in public radio. There’s a huge difference between “podcasts on the radio” vs. “radio shows that happen to be available as podcasts, too.”

    Second, a nice piece by Eric Athas, summarizing NPR’s attempts to make audio “go viral”:

    Audio doesn’t translate to the Internet the same way an image, video, or text article does. It’s not produced for the social web, which means it doesn’t have the look and feel of the things we’re used to sharing and seeing on Facebook and Twitter. That’s why NPR and member stations often “webify” radio stories. This process turns an audio story into a text story.

    Particularly resonant is NPR’s focus on finding “must-listen” audio. That is, radio stories that contain some sonically interesting element (i.e. not just an interview) that can be excerpted and used as a hook. Examples:

    Indeed, Spark found some success with this when my colleague Kent produced an interview with Julian Treasure about the most beautiful sound in the world.

  • Podcasts are based on loyalty

    Speaking of audio (not) going viral, in episode 76 of UnprofessionalMarco Arment makes an important point about podcast listenership:

    [47:46] Podcasts are a very different kind of medium than most other things on the web. Most “new media” is all about getting this flood of hits, and things “going viral” (or whatever that means), and just getting tons of drive-by hit traffic.

    Podcasts don’t work that way. Podcasts are based on loyalty, and getting people subscribed who are going to listen to every episode. So, there’s your core audience of people who you should cater to. That’s what the advertisers pay for. That’s what people tune in for. That core audience is the backbone of the show.

    And then there’s the people who will listen to one episode and complain about it. And so you have to prioritize what you do and the way you do things to please that core audience, because they’re the ones who matter a lot more.

    If you measure the success of long-form audio content using the same yardstick as a listicle, you’re doing it wrong.

    While we’re on the subject, Andrew Kurjata makes some more good points about audio virality in his post on the subject.

  • Our former selves

    As anyone who’s been to Grownups Read Things They Wrote as Kids probably knows, I’m fascinated by the relationships we have with our former selves. Most often, when a person gets up on stage to read juvenilia at GRTTWaK, there’s a distance. A distance between who they were, and who they are now. At a really basic level, I think that’s the tension that makes a GRTTWaK reading funny, or uncomfortable, or bittersweet.

    And it’s a big part of what I love about Chino Otsuka’s photo series Imagine Finding Me. In it, Otsuka takes old family snapshots, and pastes in present-day versions of herself:

    1976 + 2005, Japan

    1976 + 2005, Japan

    I first saw Otsuka’s work in Glenn Fleishman’s Glenn’s Gumbo newsletter.

    Speaking of Glenn’s newsletter and nostalgia, here’s a lovely excerpt from earlier this month:

    My friend Leah used the word saudade from the Portuguese recently. Wikipedia has a surprisingly lyrical definition:

    It describes a deep emotional state of nostalgic or deeply melancholic longing for an absent something or someone that one loves. Moreover, it often carries a repressed knowledge that the object of longing will never return.

    I don’t know Portuguese, but I wonder if it’s possible to feel saudade for a younger version of yourself. Because deep down, I suspect there’s a fair amount of that blowing around at GRTTWaK, too.

  • Vitality, not virality (or, a Few Thoughts About Online Audio, CBC Radio, and “Going Viral”)

    If you work in or care about public radio, podcasting, or online audio, I strongly recommend Stan Alcorn’s terrific piece Is This Thing On? Why Audio Never Goes Viral.

    In it, Alcorn digs into some of the reasons audio is a second-class citizen online: the web is a primarily visual medium, audio content is difficult to index, search, and skim through, etc.

    But he also drops some telling data points about just how non-viral audio really is. For instance (emphasis mine):

    In June 2013, the list of the 100 most-shared news articles on Facebook included three from NPR, but none included audio.

    And this:

    “Radiolab” and “This American Life” — public radio shows that are among the most popular podcasts and the aesthetic guiding lights for young public radio producers — are both approaching a million digital listens for each new episode. For these shows, the occasional episode will get shared more than others, but that “viral” bump is on the order of 10 to 20 percent, and even that seems driven less by social media than old-fashioned word of mouth.

    Needless to say, Alcorn got me thinking a lot about the way we approach online audio and social media at CBC Radio, which at the moment seems to be placing an increasing level of importance on online metrics like pageviews and audio streaming.

    If virality is indeed a worthy goal for a public broadcaster (and I’m not convinced it is, but more on that later), what are some things that CBC Radio could do to help make that happen? If we want more of what we do to (pause for cringe) “go viral,” how do we create the best possible conditions for that to happen?

    Here are a few thoughts.

    More pupose-built audio tools

    When CBC Radio producers make radio shows, we use tools purpose-built for audio. In studio, we use pro-level audio gear. We mix and edit sound using software built specifically for radio.

    But online, we use tools that were never really intended for audio.

    Most CBC Radio websites run on a blogging engine built for text. Our audio content lives at a CDN that describes itself as an “online video platform.”

    Should we be surprised when tools built for sharing text and video don’t work particularly well for a completely different medium with its own set of strengths, weaknesses, and modalities?

    If you want me to mash potatoes, don’t hand me a whisk.

    So then, it comes as absolutely no suprise when Alcorn writes, “If there is any company attempting to create a modern web alternative to the podcast, it’s SoundCloud.”

    Why? Because SoundCloud was built from the ground up to be an audio platform.

    Speaking of tools…

    Build audio players with sharing and spreadability in mind

    The existing embedded CBC audio players are terrible. I’d embed one here, but I don’t want to put you through that. How terrible are these players? Let me count the ways:

    They’re built in Flash. They don’t work on iOS. You can’t embed them on Facebook. By default, they auto-play. They’re wasteful, because they start downloading and playing an audio file regardless of whether the user clicks play or not. They don’t link back nicely to related non-audio content. And finally, the CBC has an agressive expiration policy for archived audio. The player enforces this, even if the underlying audio is still available.

    Luc Latulippe summed it up nicely:

    Trying to embed the CBC’s audio player is like trying to nail Jell-O to a tree

    Public broadcasters can’t realistically expect listeners to share and spread radio stories if the experience of doing so is absolultely miserable.

    Slice it up

    As I wrote back in 2011, segmentation matters:

    People want what they want when they want it. Offering easy access to individual stories (like TAL does, or how Quirks and Quarks offers a segmented podcast) helps people share and spread the stories they love. It helps create the conditions for virality.

    Things are getting better at the CBC. We’re slicing up audio, and making individual stories available. But this kind of thing is still far too common:

    “Fast forward to 10:34” or “Skip ahead to 23:30” is a terrible user experience.

    The good news is that there are tools and specs built to address this very problem. Audio deep-linking is possible. There’s a whole W3C recommendation for media fragments. Groups like Podlove are actively building tools that address these issues.

    Pay attention.

    Vitality, not virality

    The subtitle of Alcorn’s piece comes from this quote:

    “Audio never goes viral,” writes radio and podcast producer Nate DiMeo. “If you posted the most incredible story?—?literally, the most incredible story that has ever been told since people have had the ability to tell stories, it will never, ever get as many hits as a video of a cat with a moustache.”

    The thing is, so many people in public radio world really want it to. They want so very badly for audio to be able to (again, pause for cringe) “go viral.”

    When you stop and think about it, if you swap out the word “audio” for something else, the whole question seems a bit silly.

    People eat food all the time. Why doesn’t taste go viral? Almost everyone has a nose. Why don’t smells go viral? Is it because hip Berlin startup-types have yet to build SmellCloud? I doubt it.

    There’s something more basic going on here, and Alchorn gets at it when he asks:

    is a hit-machine for audio possible? And is it something anyone even wants?

    Other than radio execs, I’m not sure anyone does. And I’m not sure that virality should be a goal for public radio. I’m not sure it’s achievable. Past performance suggests it’s not, at least when we measure success using the metrics of other media.

    The last thing public radio needs is a cargo cult approach of aping what works for other media, then acting all disappointed when it doesn’t work out so well for us.

    A very savvy CBC executive once advised the Spark team to “fish where the fishes are.” I think that applies to CBC Radio more broadly.

    Radio excels in certain contexts. For instance, commuting. In Canada, 82% of commuters travel to work by car. Audio is great in cars, and tech-wise, lots of exciting things are happening in cars: iOS in the car and the Open Automotive Alliance to name two. So then, what is public radio’s car strategy? What is CBC Radio’s car strategy? Analog transmission?

    Part of CBC’s mandate is to ensure our programming is “made available throughout Canada by the most appropriate and efficient means and as resources become available for the purpose.”

    I would much rather CBC Radio’s digital efforts go towards figuring out how to do a better job of what we’re already good at, rather than trying to get good at making empty-calory web content (listicles, memes, GIFs, etc.) that simply cannot compete with the stuff coming out of Upworthy, BuzzFeed, and others.

    I don’t know exactly what digital success for public radio looks like. But I do know getting it right means embracing the best qualities of the medium itself, not shoving radio stories into listicle-sized holes.

  • How I ditched my voice plan and went data-only

    One year ago, I decided to take my iPhone data-only. No voice plan. No texting plan. Just sweet, sweet data.

    I don’t like paying for things that I don’t use, and I rarely use the telephone part of my iPhone. That is, I don’t make many voice calls over the cell network. So I figured I could save some money.

    I looked around for data-only smartphone plans in Canada, but couldn’t find any. The carriers don’t seem to offer data without voice.

    However, there’s an exception: tablets.

    Most of the big carriers are happy to sell data-only plans for use with a data-only device, like a tablet, hotspot, or wireless modem. But can you sign up for a data-only plan intended for a tablet and use it with a smartphone?

    Yes. And that’s exactly what I’ve been doing for the past year. I use a month-to-month tablet Flex plan from Bell with my iPhone 5. Most months, I pay ~$40/month (tax in) for 5GB of data, and that suits me just fine.

    Sound good? Here’s how I got it all set up, and some of the workarounds I’ve found that make living data-only a little easier:

    Getting a SIM card

    Bell Nano SIM

    To pull this off, I first needed a Nano SIM card for my iPhone. Bell sells Nano SIMs off-contract for $10, but I’d heard that if you ask nicely, the Apple store gives them away for free.

    Sure enough, when I visited my local Apple store and asked for a Bell Nano SIM, they gave me one. I didn’t have to sign up for a contract in-store, or provide any personal information. Easy.

    Setting up a new account

    To sign up for Bell’s tablet flex plan, I had to set up a new account. Honestly, I would have preferred a prepaid option, but the flex plan (with its desirable tiered pricing) was only available on a post-paid, month-to-month basis.

    Setting up a new account was simple. I telephoned Bell (on Jenna’s phone), asked for the tablet Flex plan, and read them the ICCID from my newly aquired Nano SIM.

    Because Bell was running a promotion at the time, they waived the usual $35 setup fee, and offered me “worry-free” (no overages) data for $5/month for the first two months. Bonus.

    Shortly after hanging up with Bell, I popped the Nano SIM into my iPhone, and was happy to see that everything worked with zero on-device configuration. Nice.

    Secret Bell phone number

    A side note: despite signing up for a data-only plan, my Bell Nano SIM does have a Toronto-area 647 phone number associated with it. I can place and receive calls, and send texts to and from this number. But I don’t, because calls and texts are billed by Bell at high rates ($0.40/min and $0.20/text in 2013), and I have better, cheaper workarounds (see below). That said, it’s nice to know that in an emergency, I could make a good old-fashioned telephone call.

    Speedy data

    In my everyday use, Bell’s LTE is fast. This is typical:


    Bell’s top flex plan tier gives me up to 5GB/month — way more than I’ve ever actually used (outside of the first two “unlimited” months, when I went crazy). Data costs $10/GB if I ever go over.

    Yes, tethering works just fine, with no additional charges. This is especially handy when Via Rail’s on-board WiFi just isn’t cutting it.

    Workarounds: Voice Calls

    Like I said, I could make calls over the Bell network, using the number associated with my SIM. But that would defeat the purpose of going data-only, wouldn’t it? Instead, I make voice calls from my iPhone using a handful of alternatives.

    For calling “regular” phones, I use Acrobits Softphone for iOS, configured to use a SIP account through VoIP.ms. I ported my local Toronto phone number to VoIP.ms in late 2011, and haven’t looked back since. This SIP setup lets me place and receive voice calls easily, and never uses Bell’s voice network.

    I also have a Google Voice account, which I sometimes use for longer outgoing calls. Google’s Hangouts app for iOS lets me dial regular phone numbers directly.

    Finally, for calling other iPhone owners, I use Facetime Audio. It’s baked into iOS at the system level, the audio quality is very good, and it works well on both WiFi and cell data networks. I agree that Facetime Audio is Apple’s biggest little feature addition in iOS 7, and I look forward to its debut on the Mac in 10.9.2.

    A small Facetime Audio wrinkle: Since Facetime on the iPhone is associated with your iPhone’s telephone number, and since my iPhone has a Bell-issued number that I don’t use (and don’t want anyone to know about), it’s important that I set my Facetime Caller ID to something other than that “secret” number. I use my email address instead.

    I know this sounds overly complicated. Believe me, it’s not. Like I said, I don’t make a lot of voice calls on my iPhone, and for anyone who dials me, my phone number is the same as it’s been for years.

    Workarounds: Texting

    iMessage has almost completely replaced texting for me. All data, all the time.

    When I need to text someone who’s not on an iPhone, I use the Google Voice iOS app, and make sure to preface my message with “It’s Dan,” because my Google Voice number is US-based.

    A texting wrinkle: Weird things happen when people text my local Toronto (416) number. Because my SIP provider doesn’t support incoming SMS messages, texts often turn into voicemail messages featuring a text-to-speech robot. In the future, I may switch away from VoIP.ms to another SIP provider that supports SMS, like Anveo.


    It took a bit of work to get set up, but a year in, living data-only works well, and saves me money.

    My ideal wireless carrier would act like a dumb data pipe, and let me pick and choose the services I run on top of it. With a data-only tablet plan, I get as close to that as possible.

    As always, your mileage may vary.