Second: Got creative. I wrote a UNIX shell command and ported it surreptitiously to the wifi router on the plane. From there, it propagated to the other routers, and then on to all wifi connected devices.

Third. Instruction time! Unix commands tell all the coach-connected devices to turn over 30% of their processing power to the “new hive mind” (i.e. botnet, my wife is Russian, I know this shit), which thereby launches a denial-of-service attack on the Business class devices, who – can you believe it! – have their own business class IP cluster. Which presently all stop working, while all remaining bandwidth is prioritized to the devices in coach.

Fourth: having accomplished this, I drink some bad red wine (this is coach class, after all), reflecting that I have just connected the Occupy Wall Street and NetNeutrality movements! Brilliant!

And better than trying my hand at a stuxnet worm and then [xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx]

(the rest edited by TSA censors…)

Friday was my last day, and inevitably, over the past two weeks, I’ve been reflecting on the people and brands I worked with, and what we got done together.

It doesn’t actually seem like yesterday, but a long time ago that Howard Applebaum hired me — ostensibly to revamp Billboard.com but in truth, as it turned out, because I possessed an extremely large (mostly useless) vocabulary and had a knack for fixing computers. At any rate, these past three years have been wonderfully productive, a time in which we accomplished a great deal:

In July 2009, we unleashed a new Billboard.com on the world — fully consumer, highly social and interactive, with music playlists and the full Hot 100 and Top 200 charts. Since then, total site visits have grown 200%, from 6.5 mil per month to 19 million. And we won a coveted ASME award for best website design in the process. This work wouldn’t have been possible without my team, as well as Bill Werde, Howard, Lila Gerson, and Jess Letkemann.

Also that year, we launched the first Billboard Chart API, which was later used to great effect by the New York Times, in the wake of Michael Jackson’s death, to create an interactive graphic detailing MJ’s chart rankings.

We made the unprecedented decision to digitize the entire Billboard archives, get Google to pay for it, and make them available to the world. Special thanks to the Billboard Charts team for helping make this happen.

We launched the first Billboard iPhone application — the Chart app — which has been downloaded over 300,000 times and made #5 on the iTunes top music apps list within its first week. Andrew Min brought this deal, Lisa Howard gave it crucial support (as she has done for so much over the past 18 months), and George White got it done.

In October, 2010, we finished in 3.5 months what normally takes 6 months – the complete redevelopment and relaunch of THR.com. In the ten months since then, total visits to the site have increased from 4 mil per month to 10 million, while pageviews grew 200%, from 7 million to 23 million. The vast majority of this growth came from search, google news, and social, spurred, in turn, by design and platform changes — proving that no matter how good your content is, it doesn’t matter unless Google and Facebook users can see it. 

For helping get THR.com done on time I thank Richard Beckman (who pushed me to hit a goal which no one believed possible), Gautam Gulliani, Alex Boyce, Reed Hallstrom, John Marchesini and Lindsay Powers.

Also in October, we launched the company’s first iPad app, offering individual issues of the beautifully redesigned Hollywood Reporter, as well as breaking Web content.

Most recently, in April of this year, we relaunched Adweek.com, probably the smoothest project I’ve worked on here, and great fun as well. Despite deep-sixing two other sites in the process (Mediaweek.com and Brandweek.com), traffic to Adweek.com in August 2011 was 101% higher than traffic to the three sites combined in August 2010.  Special thanks to Michael Wolff, Doug Ferguson, Nick Eckhart, Hillary Frey, and Jim Cooper for maintaining a sense of humor and focus.

In summary – to use a Billboard metaphor – you could say we brought these brands into the digital age. With a bullet.

None of this would have been possible without my team, my colleagues, and the support of the owners. For that, and for the laughs and the adventures, they have my enduring gratitude and respect.

I look forward to watching how the brands evolve digitally over this year and next, as I am sure they will.

The latest online music sensation, Turntable.fm, has reportedly raised a $7 million round of financing at a $37 million valuation.

No doubt this is the most interesting online music service to come around in a couple years; it is hugely engaging and highly addictive. I am a big fan of the service myself and use it nearly every day. But still — $38 million seems a bit rich for a company with 300k users, no clear business model, and likely legal challenges. No doubt the company will grow its audience, and figure out some kind of revenue model, but the resolution of that last issue — the legal challenges — could make or break the company.

Turntable.fm operates as a statutory licensee of SoundExchange, just like Pandora, Slacker and other online radio services, all of which operate under the legal umbrella of the DMCA and pay royalty rates set by the Copyright Royalty Board (as well as fees to ASCAP, BMI, and SESAC). Although the DMCA is a notoriously imperfect document, it does spell out some clear rules for Webcasters, particularly around what they cannot do. Failure to comply with these rules usually means you’re in the “interactive” (on-demand) bucket, where individual label deals are required and the fees are much higher. You don’t want to be in that bucket unless you have no choice.

Some of the Webcasting rules include: 1) defining a radio “station” as at least three hours of music, 2) during which the same artist cannot be played more than three times; 3) not allowing users to play songs of their choosing on demand; and 4) and not publishing to users the order in which songs will be played in the station, or even the randomized list of all songs in the station.

Although Turntable.fm does have some safeguards in place, it is easy to “break” many of the rules above, often by making use of the site’s social features. For example, I could create a room called the Billboard Hot 100, invite four of my friends to it, and we could all agree to play the songs of the current Hot 100, in order, and nothing but the Hot 100. This would break rules 3 and 4 above, since we’d be colluding to play a specific playlist of songs of our choosing. Or I could start a Madonna room, in which people play nothing but songs by Madonna, breaking rule 2 above. It’s easy to think of other examples.

The likely outcome of this is that Turntable.fm will be forced to either a) follow the rules of the DMCA strictly, disallowing the above examples; or b) do deals with the labels, at rates higher than those of SoundExhange.

Both scenarios will cost them users. Scenario A will have a negative effect on the user experience, making the service less exciting and addictive. This is a fundamental law of online music services — user experience usually suffers around the time rights holders get involved (partly this is the fault of the user who expects digital content to always be free and unrestricted).

Scenario B will force Turntable to become a subscription service, since the revenue from advertising is never ever enough to cover the costs of on-demand licenses — a second fundamental law of online music services. This will make the service available only to those willing to pay, which we know from experience is not a large number in the US, somewhere around 1 – 2 million.

Personally, I would pay something to use Turntable.fm, but will the masses?

This redesign is long overdue — since 2006, the company has made almost zero improvements to the design or behavior of its basic Web product. Instead, they’ve focused on a) improving their song matching intelligence (an investment in technology IP) and b) building and launching their mobile app, which has driven massive user growth. They didn’t invest in improving their user experience because they didn’t have to — they were gaining users anyway.

Now, with a bunch of things happening at once in this space — the incompatibility of Flash with many Apple devices (and its decline as a technology), the rise of compelling social music services like Mog & Rdio, the launch of Spotify in the US, the threat of iheartradio, more VC funding for digital music startups, their IPO — Pandora realizes it’s no longer the upstart but the incumbent.

Interesting how the new design looks like the offspring of Spotify, iTunes and Rhapsody Beta.

• An excellent user experience for all of their products; in the early years of the company Google avoided taking advertising on search pages because they strongly believed in growing their user base first. Like Facebook, they take an extremely long view and – almost as a rule — do not chase short-term monetization. In fact, many of their finest products make little or no revenue on their own (e.g. Google Maps, Google Earth), but must be seen within the constellation of other Google products.

• Being world-dominant in certain areas, including:

1. Advertising: by forcing it to be more metrics-based, accountable and efficient;
   
2. The digitization and distribution of information;
3. The distribution of content: through platforms such as YouTube, Google TV, the impending eBook store, the pending Music Store (not to be confused with its current iteration), Android-based mobile phones;
4. The collection of user data (though they don’t admit this): via searches typed into Google, the use of Gmail accounts and Google Documents, the near-ubiquity of Google Maps, a huge push into the mobile operating system market through Android, the launch of the Chrome web browser, etc. In triangulating between all of these interactions by a user, Google ends up possessing a great deal of data about that particular user. This allows them to continue to improve the user experience of their products, as well as serve more targeted advertising.

Google is not interested in (based on observations of their behavior):

• short term revenue initiatives;
• exclusive deals with any particular content partner;
• selling hardware devices;
• traditional media business models.

The Google Music potential:

Give users a music store, with songs and albums tied into search results, and allow them to be bought and ported effortlessly to any Android device (on a cost per song or subscription-basis). Other than securing the licensing deals with the labels, quick and easy. No need to use downloaded software like iTunes or physically sync a device to a computer.

Google + Android + Music = a virtuous cycle of influence and revenue

1. Google’s primary revenue stream is advertising;
2. The more devices running Android, the better it is for Google’s advertising initiatives, for a couple of reasons:
   1. Devices running Android are more “open” than devices running Apple, Microsoft or Blackberry systems, and therefore not subject to control by those companies;
   2. Android natively works with other Google products – some of which are advertising-based, all of which collect data – in a seamless way;
   3. In 2009, Google acquired one of the largest mobile ad platforms in the world, AdMob, allowing them to better control ad serving on Android (and iPhone) devices; between ’07 and ’09 alone, mobile search queries increased 500% — all of these queries are advertising opportunities;
   4. Consumer adoption of Android devices is driven by better and better content being available to the users — music, TV shows, apps – and seamless interaction across other Google mobile services, such as Maps, Gmail, YouTube;
   5. Go back to step #1, and repeat.

The viewpoints expressed herein are my own, and not necessarily those of my employer.

Brief and entertaining.

The barcode scanner makes the Amazon app pretty much the best thing you can have with you while shopping in a physical store. Imagine you’re in Best Buy looking at a 46″ Samsung TV – is it cheaper on Amazon, you wonder? Twenty seconds later, you have your answer (it wasn’t).

This is not a brand new update to their app (October last year), but it shows how good digital products tend to be highly iterative, with small updates sometimes driving huge increases in utility value. The lesson is that almost no one gets it exactly right the very first time.

Here is my rough attempt at a “social graph taxonomy”, a phrase that would seem pretentious were it not already highly in vogue. This is based on my personal and professional experience:

Foundational:

• The core graph upon which much else – key relationships, data, accessibility to other services – rests;
• Semi-permanent & denoted by high repeat usage;
• Focuses on established relationships;
• Allows a presence for 3rd parties – most notably, brands and app developers;
• Facebook is the obvious example.

Broadcast & Dialogue:

• Main purpose is to allow the quick broadcasting of short, headline-style messages designed for easy pickup and further transmission;
• Two-way dialogue is possible, but more than that becomes unwieldy;
• Often used as a means of referral to other content, whether text, video, audio, or image;
• In the case of brands, affords the opportunity to: i) get direct feedback from customers, ii) put a human face on things, iii) establish a style;
• Relationships are i) more anonymous than personal, ii) ever-shifting, and iii) usually dominated by a small % at the top with large followings;
• Like in high school, power is based on popularity: the more followers one has, the greater the influence and status (if you detect an edge of bitterness here, it’s completely accidental);
• Yes, I am talking about Twitter.

Closed conversational:

• Closed spaces for group texting, generally invite-only;
• Most present on mobile devices, although don’t have to rely on native SMS;
• Augment the textual experience by making use of public APIs to bring in Location services (Google Maps, Foursquare) and allow the sharing of media, such as photos;
• Key examples are GroupMe and Beluga.

Location-based:

• An especially interesting, mostly mobile-based, group — these apps enable a kind of augmented reality by allowing data and media to be layered onto real places, making those places more “meaningful” by connecting the participant to the experience of others at or near that same space;
• One of the most shifting, transitory social graphs possible, since participants are connected mainly by geographical locations, which are by definition highly variable and numerous;
• There may be no overlap whatsoever with the Foundational social graph, nor is any necessary for the experience to be interesting or rewarding, although importing contacts through FB or Twitter authentication can add another dimension;
• Important examples are Foursquare, Color (for images) and Soundtracking (for music).

Taste-based:

• Main connective tissue is a shared interest in a particular cultural artefact, such as a song, photo, or video;
• The social connections can be transitory, but also very meaningful, since people with similar tastes have a higher liklihood of getting along than those with divergent tastes. This is the essence of homophily (“love of the same”) which is at the core of all strong social networks;
• Main examples of taste-based graphs are Jango (the other people you see in your music player listening to similar music), Last.fm (the concept of “neighbor radio”) and YouTube.

Creative & Generative

• A graph based on a shared creative endeavor – whether pursued professionally or just passionately – such as photography, music, writing, or video;
• The bond is based on the creation of original content, as well as the user’s need for i) an audience response to his/her work — a need shared by all creative people, irrespective of monetary considerations; ii) constructive feedback on the work itself by other professionals;
• There is much overlap with the Foundational and Broadcast social graphs, since both serve as important promotional outlets for the works created. Indeed, it is hard to imagine SoundCloud without the ability to embed; or Flickr or Vimeo without the ability to share;
• Key examples: Flickr, Vimeo, SoundCloud.

(I didn’t bother to include LinkedIn anywhere above, because it seems to have evolved into a job site / DIY Hoovers hybrid).

My next task, once I’ve gotten some feedback on this, will be to represent the above taxonomy visually.

As Glenn Peoples notes in his good piece on Pandora, people tend to focus on a couple of key things when thinking about the company’s IPO; the most interesting one to me is that, even with $90 million in revenue (for the nine months ended 10/31/10), Pandora is barely break-even. In their S-1, the company even states “we expect to continue to incur operating losses on an annual basis through at least the end of fiscal 2012.”

Since Pandora’s growth is driven by mobile, but display advertising on mobile devices is in such infancy that it will not cover the expenses of that streaming content, Pandora will have to do two things to get profitable, in my opinion:

1) derive much more revenue from locally targeted audio advertising than they do now (in this regard, it’s helpful that most mobile devices are GPS-enabled)

2) increase their revenue from subscriptions – either through a more compelling freemium model, or by charging some kind of fees for their mobile app.

________________

Pandora filed its S-1 in February, and the document makes some interesting reading, especially the P & L page. Some quick takeaways:

• The company had an operating loss of $15 million in 2010
• 58% of its total revenue gets spent paying content licensing fees
• 9% of its $55 million in 2010 came from subscriptions, the rest came from advertising

Among the risks they list to their business, the following two will be acutely familiar to anyone who’s been in the Web radio:

• our ability to more effectively monetize mobile listener hours, particularly as the number of listener hours on mobile devices grow;
• our ability to continue to operate under the statutory licenses set forth in the Copyright Act;

The whole document can be found here.