Category Archives: Technology

Measuring Journalistic Success

GigaOM’s Mathew Ingram on Andrew Sullivan’s decision to erect a paywall and start charing for content:

Despite his following, however, it is far from clear that Sullivan will be able to make the transition work — yet if he does, he could become the first real success story of the post-industrial journalism era. [emphasis added]

I have a problem with this quote.  I don’t mean to pick on Ingram, who I think is a good writer, but rather an overall trend I see when people write about paywalls. There’s an implicit assumption that financial solvency equals success.

The truth, I think, is more complicated.  One of the great things about the Internet when it comes to journalism is that it has explicitly decoupled financial renumeration from success. There are plenty of people blogging and writing for no money whatsoever, either because they think it will further their professional career or they just plain like having a venue in which to spout off (like me).

Even if we assume that being able to make a living as a writer/journalist is the end goal, then it’s far from clear that all financial models are created equal.  Presumably many writer/blogger/journalists are in the game to influence the debate.  They want to have an impact, have their ideas get out there and change the world. If that’s the case, then, an ad-supported model is clearly superior, all else being equal.  If you can make $100k/year from a member-supported site that reaches 10,000 readers or an ad-supported site that reaches 1,000,000 readers, doesn’t the latter have the potential to make a greater impact?

Of course, there are all sorts of caveats here.  For one, it may not actually be possible to make an equal amount of revenue. Ad-supported journalism online is a really hard game to play.  HuffPo does it better than most and they need millions of pageviews and lots of crappy content to make it pencil. Alternatively, those 10,000 readers could theoretically be incredibly influential policymakers and therefore have a lot more impact than a million schmucks like me reading your writing.  Finally, I don’t know the particulars of Sullivan’s paywall but I imagine it will have some porousness for allowing links from social media or other sources so as to maintain some broad appeal.

All of which is to say it’s not a clear cut case one way or the other.  So when we talk about “success” in journalism, there’s a lot more at work these days than whether you can pay the rent.

The Five Companies You Meet in Technology

Nick Bilton’s article on Twitter CEO Dick Costolo:

Mr. McCue of Flipboard says Twitter and its C.E.O. are pushing the traditional limits of what it means to be a company that delivers media to people. It is neither a technology company nor a media company, he says.

“Twitter is an entirely new thing,” Mr. McCue says. “I like the fact that Twitter is unapologetic to that. Dick is the same way with his decisions.”

When something new comes on the scene, we often have a hard time classifying it.  You see this all the time with music.  Or TV: when “reality TV” first appeared a dozen or so years ago, we didn’t quite know what to make of it.  A couple of years ago, the Emmys started a “reality” category.  But what we call “reality” is really two distinct genres: prime-time game shows like Survivor, The Bachelor and Top Chef, where contestants win prizes (true, sometimes the “prize” is a husband, which is a little creepy), and trashy documentaries (Housewives, Jersey Shore, etc.).

There’s something similar afoot in “technology.”  With due respect to Mr. McCue of Flipboard, I’m fairly certain Twitter is a media company.  The way I see it, there are really only five types of companies that tend to get branded with the “technology” label. Or rather, there are five business models typically associated with technology.

  • Hardware companies, who sell people physical devices in exchange for money, like Apple, HP, and Dell.
  • Software companies, who sell bits.  Microsoft, Adobe, and Salesforce are all in this camp.  I’d argue even web companies like PayPal fall into this camp – PayPal business is selling its credit card processing software in exchange for a fee.
  • Media companies, who aggregate eyeballs and sell them to advertisers (Twitter, Google, Facebook, Yahoo!).
  • Marketplaces, like eBay, Amazon, and Craigslist, who facilitate the sale of goods, often in exchange for a fee.
  • And finally, consultants, like agencies and IT shops, who build software or service hardware in exchange for money.

Obviously some companies engage in more than one of the above lines of business.  The fact that Apple sells software or Google gives software away does not change the fundamentals of their business model.  LinkedIn sells both advertising and software (professional recruiter tools).  Calling them all “tech” companies illuminates very little.


Transforming the TV Industry as the 800-lb Gorilla

MG Siegler says of course Apple is talking to the TV content companies and of course they won’t be able to transform that industry overnight:

It seems that the shock of this news is more around the fact that Apple may not actually completely transform the industry overnight. No shit. You know what other industry they didn’t transform overnight? The mobile industry.

John Gruber links and comments:

Or the music industry for that matter. The iTunes Music Store wasn’t some all-new thing that obviated the existing music industry — it was built on top of the existing music industry.

All true. But two things are different now: Apple is the most valuable company in the world and Steve Jobs isn’t around to cajole the TV industry into playing by his terms. The latter is less important (although Jobs was a master salesman), but it’s easy to forget that back in 2003 the music industry was willing to take a chance on Apple because the iPod only worked if you had a Mac (!) and thus was limited in its potential impact.  As for the iPhone, clearly AT&T’s made a lot of money off it, but if you asked them to do it all over again and take the kind of terms that Apple’s offering (no cut of app sales, marginalized SMS fees, huge data usage and infrastructure costs), they’d probably ask to renegotiate.

Apple’s been so successful at redefining industries on their own terms that it’s no surprise the content companies are wary this time around. They’ll eventually give in, because they’re terrible at building compelling customer-facing digital experiences and Apple is fantastic at it, but that doesn’t mean they won’t spend some time trying to go it alone first.



Designing Services

Service Blueprints

CP+B CEO Andrew Keller writes about the need for a “fundamentally different way to make a brand”:

We need a fundamentally different way to create a brand: the way it talks, behaves and thinks and the way to leverage ALL touch points–especially those beyond the obvious “social” touch points like TV. The opportunity is to build a brand that is relevant and explosive within the world’s new social contract and to socialize old media points, uncover/activate new touch points and bring it all together to create a system that connects people to a brand.

Keller’s argument boils down to a few key points, which are:

  • People experience a brand through certain touch points
  • These touch points are how the brand comes to be perceived
  • Conversations with customers means having good content for sharing social media (another touch point)
  • All touch points need to be designed in a way that reflects the brand

Put another way, it doesn’t matter how good Delta’s TV commercials are if they lose your bags or you get a rude flight attendant. The touch points are the brand.  Keller concludes:

So whatever you want to call it–marketing, advertising, Hollywood meets Silicon Valley, branding, inventing–it’s never been more exciting. But brands will need to be made ready for this world, this new social contract, and then activated.

There actually is a word for this idea: service design.  It means designing customer experiences wherever they occur.  It’s great to see Keller getting the memo.  The question is, can those of us in user experience design who’ve been doing this kind of work for years can convince our clients to help us tinker with their core operations in this way?  I hope so.

Making it Better

Jonathan Ive on Apple’s design process:

Q: What are your goals when setting out to build a new product?


A: Our goals are very simple – to design and make better products. If we can’t make something that is better, we won’t do it.


Q: Why has Apple’s competition struggled to do that?


A: That’s quite unusual, most of our competitors are interesting in doing something different, or want to appear new – I think those are completely the wrong goals. A product has to be genuinely better. This requires real discipline, and that’s what drives us – a sincere, genuine appetite to do something that is better. Committees just don’t work, and it’s not about price, schedule or a bizarre marketing goal to appear different – they are corporate goals with scant regard for people who use the product.


That’s actually a fairly radical statement if you think about it, one that’s at odds with how the company is viewed in the popular imagination.  People who don’t understand Apple tend to applaud the company for being new and different, but Ive says that “new” and “different” is specifically not what Apple’s about.  Apple’s about being better.  Better than whom? The implication is that someone else is doing something poorly already, and Apple’s poised to swoop in and improve upon it.

  • Apple didn’t invent the PC, they made it better.
  • Apple didn’t invent the MP3 player, they made it better.
  • Apple didn’t invent the smartphone, they made it better.
  • Apple didn’t invent the tablet, they made it better.

Ironically, Apple’s critics get this better than many of its fans. The critics, though, mean it as an insult: “Microsoft invented tablets back in 2002!” “It’s all just marketing,” etc. But real success comes when you drill down to the essence of a product or an idea and make it better.  Why do you think Apple’s R&D budget is so small? They’re not wasting money inventing “the living room of tomorrow” or whatnot. They’re just taking the living room of today and removing all the stuff that sucks about it.

All of this makes sense if you view apple as a conservative company.  I don’t mean that in the American political sense.  I mean small-“c” conservative, in the sense of resisting change for change’s sake.  This manifests itself in various ways, from the giant cash hoard to the schmaltzy, throwback design style of iCal and Game Center to the 70s-era suburban mega campus.  Perhaps this originates with Steve Jobs’ back-to-the-farm hippie roots, or maybe it’s the hard-won knowledge that people can only process so much change at once.  Whatever the reason, Apple clearly values moving slowly when they enter new markets.  Better to come in later after everyone’s made all the mistakes.

Apple, Amazon and the Wicked Problem

Note: I cross-posted a version of this post at Seattle Transit Blog

The Atlantic/City of Cupertino

By now you’ve probably heard that Amazon is planning three new office towers in downtown Seattle.  Plans call for more than 3 million square feet of office space, in addition to the 2.7 million square feet that Amazon already leases in the Downtown – South Lake Union area.  Somewhat intriguingly, Clise Properties is “hoping and expecting” that Amazon will eventually buy all 13-acres Clise owns in the Denny Triangle (wishful thinking?).

Looking at the plans, it’s hard not to contrast Amazon’s project with the proposed Apple campus in Cupertino. At 2.65 million square feet, it’s roughly equal to what Amazon already leases between Denny Triangle and South Lake Union. Several blogs have already commented on how Apple’s proposal violates all sorts of urbanist principles, especially when compared to Google’s New York City office or the new Twitter offices in San Francisco.  To be fair, Apple’s proposal is no worse than in urbanist terms than the headquarters of most Silicon Valley titans of its vintage, such as HP, Yahoo!, and Oracle.  Still, when the opportunity arises for one of the world’s most admired companies to completely reconfigure a huge swath of one of the world’s wealthiest ZIP codes, it’s reasonable to expect something more than a re-hash of the classic suburban office park, beautiful as it may be.

If you’re a design nerd like me, you may be familiar with Gary Hustwit’s trilogy of films, HelveticaObjectified, and Urbanized.  One way to think about these films is that they move outward in considering increasingly complex forms of design: graphic design, product design, and finally urban design.  (By “complex,” I mean in terms of the sheer number of people involved.)  Apple products are featured prominently in Objectified, and Apple’s new campus is clearly an effort to create a building that could be featured in that film. But it’s also a paean to a bygone era of suburban sprawl.

Steve Jobs may have been a genius, but he was also – to quote The Big Lebowski – a man for his time and place.  He was born and raised in the postwar Bay Area suburbs, and even became obsessed with the specific style of suburban tract home he grew up in. Thus, the Apple campus represents the apotheosis of a time when we thought of buildings as distinct objects, like iPads. Now we recognize them as part of a complex ecosystem involving streets, sidewalks and transit.  For example, smart building design these days often involves letting the outside world in as much as possible to minimize energy costs. By contrast, Apple’s campus won’t let employees open the windows.  Then again, Jobs was never much interested in complex ecosystems he couldn’t completely control.

It’s no coincidence that Hustwit ended his trilogy with Urbanized.  Urban design is a wicked problem.  This may be why all transit thinkers inevitably come face-to-face with urban design, building design, and even complex systems like education and health care delivery.  It’s impossible to fully consider the city without doing so.

In the meantime, props to Amazon for building densely, in the city, a few blocks from Westlake Station. I won’t assume that they did it for urbanist reasons.  Perhaps the famously frugal Jeff Bezos simply wanted to build fewer expensive underground parking spaces than the 10,000 subterranean spots that Apple has planned for Cupertino.  More likely, Amazon knows that attracting good workers means being close to amenities. Whatever the motive, it represents an encouraging trend in high-tech corporate headquarters away from the sprawl of yore.

Dept. of Pre-Crime

Three news stories appeared in my Twitter feed over the last month.  First: Google Maps Help Predict Meth Labs Before They Open.

Using the data, the pair were able to successfully prove meth manufacture was creeping slowly through more and more middle-class neighborhoods in Colorado Springs. Map data analyzed over time successfully demonstrated the spread of meth labs throughout a metropolitan area–and even predicted where they would pop up next.

Second: Misfortune Teller:  A statistics professor says he can predict crime before it occurs:

Drawing from criminal databases dating to the 1960s, Berk initially modeled the Philadelphia algorithm on more than 100,000 old cases, relying on three dozen predictors, including the perpetrator’s age, gender, neighborhood, and number of prior crimes. To develop an algorithm that forecasts a particular outcome—someone committing murder, for example—Berk applied a subset of the data to “train” the computer on which qualities are associated with that outcome.

Third: Pre-Crime Detection System Now Being Tested in the U.S.

The FAST system has the capability to monitor physiological and behavioral cues without contact. That means capturing data like the heart rate and steadiness of gaze of passengers about to board a plane. The cues are then run through algorithms in real-time to compute the probability that an individual is planning to commit a crime.

The Department of Pre-Crime prophesied in Minority Report is well on its way, and it won’t require three mutant sisters submerged in water.  Just add big data, behavioral psychology, and ubiquitous monitoring.  Companies like Palantir will make a fortune.  Hopefully some of us will stand up and ask, “is this what we want?”

“User-Led Innovation” Does Not Exist

This piece in Fast Company has been getting a lot of traffic lately. In it, authors Jens Martin Skibsted and Rasmus Bech Hansen argue that “user-led innovation can’t create “breakthroughs.” The authors offer IKEA and Apple as proof, two successful product companies who apparently “don’t waste [their] time asking users,” and instead just build great products.

The biggest problem with the article is the misleading phrase “user-led innovation.” No designer in their right mind would ask users to “innovate.” As Henry Ford said, if he asked people what they wanted, they would have said a faster horse. You can’t just survey people and give them what they say they want and expect to succeed (well, unless you’re a politician or a Hollywood action movie producer). There’s a huge difference between user-centered design, which is an attempt to create empathy with the user and so better understand their perspective, and “user-led innovation,” which I’d never heard of until I read this article.

Apple and IKEA engage in what Jared Spool calls “self design.” The product designers are the users: Apple and IKEA employees are also avid consumers of computers and furniture, so the empathy is natural and unforced. For many popular consumer products (Twitter, Nike, Facebook), this strategy will often work well.

But what if you are not the target user of your product? What if you are, for example, designing a product to help elderly people navigate their kitchens? Self-design won’t work. Perhaps an enthusiastic group of 80-year-olds will form a company and start knocking on doors along Sand Hill Road looking VC money. More likely, someone who is not elderly will attempt to serve this market, and will need to find a way to empathize with their population. Listening to, observing, and immersing themselves in the world of an elderly person is a great way to gain insight on designing a product for this group. Or, as Fred Wilson says, of the companies he invests in, “domain expertise to the point of obsession is highly correlated with the most successful entrepeneurs in our portfolio.”

In short: understanding your users = good; asking them to do your job for you = bad.

I understand where Skibsted and Hansen are coming from. They’re making a case for visionary products instead of market researched-crap. In that, I couldn’t agree more. But the way you get there is the same way you’ve always gotten there: understand the users first, and then design an innovative product that meets their needs.

The Future of the Music Industry

Matt Yglesias on the death of recorded music:

This is one reason why I would discourage bands from trying to underprice tickets at their own shows as a reward to fans. Since digital copies of recordings are non-rival and basically free to make, any non-zero sale price entails some deadweight loss. And since concert tickets are necessarily scarce, any sub-market price entails some deadweight loss. The optimal strategy for a popular band that wants to do something nice is market pricing for concert tickets, plus free recordings. Or even better, you could release your records into the public domain.

This is in the context of Matt’s ongoing efforts to get bands to charge market rates for their concert tickets, rather than selling cheap tickets and letting scalpers extract the difference in value.

As recorded music collapses, I wonder if we’re going to see new efforts to maximize the efficiency of the live music experience. In the future, bands will make more of their money on the road. One way to do this would be to play more shows, as Duncan Black suggests. Another option would be to play larger venues. The problem with the former is that each night on the road costs you another night’s lodging, another day away from the family, etc. The issue with the latter is that there are only so many large venues. Perhaps over time, average venue size will increase. But in the short term, it seems like the problem is more that these venues are typically empty for 19 hours a day, and then booked from 8pm – 1am (and the headliner often doesn’t go on until midnight — even on a weeknight). A hugely inefficient allocation of venue space!

So here’s Frank’s solution: book two shows a night in the same venue! First, get rid of opening acts on most shows. Opening acts were a good idea when it was hard to introduce people to new bands, but that’s gotten so much easier with social media, Pandora, what have you. Instead, have an 8pm show for geezers like me who can’t stay up late on a weeknight (and maybe the under-21 set, too, if licensing laws permit), and then a 10 or 11pm show for the 20-somethings.

In this scenario, you’re making better use of your resources. You don’t have the increased overhead of the larger venue (more security guards, etc.), you’ve increased the potential audience (presumably by bringing in the under-21s and the geezers, who would have otherwise not attended), and you’re now booking two shows instead of one for each night on the road.

As society gets richer, and the costs of having a live performer perform for a fixed audience get higher, I imagine we’ll see this sort of thing not just in live music, but other performing arts (theater, dance, etc.) as well.

Agency Information

One of Jarrett Walker’s readers asks for examples of cross-agency integration:

Do you have some good examples where services and customer information is more regional ie all the options in the region whether public or private?

I can actually speak with a bit more than just a blogger’s opinion on this, because this was sort of the subject of my capstone project in grad school.  You can read more about the project here, but the basic gist is this: we were looking for a way to integrate traffic incident reports from different agencies so that a user could view a single map and see which roads were closed throughout King County, whether they were city roads, county roads, or state roads.

This turns out to be a trickier problem than you’d think.  Technologically, it’s pretty straightforward.  The tricky part is that agencies store their data in very different formats.  Some agency might define an open road as any road that isn’t closed, while another agency might have specific fields for open and closed.  One agency might use mile markers to identify a stretch of road, while another uses GIS coordinates.

There are standards for this data, and a large part of our project was figuring out how to use the standard to translate among the different agency sites.  The result is that you can now visit King County’s traffic map and see not just incidents on county roads, but WSDOT roads as well.  The incidents are populated dynamically, using the same data that drives WSDOT’s own incident map.

On a related note, Walker goes on to speculate that, in the long term, transit agencies will cease to provide trip planning services, and instead simply provide data that private companies (like Google) can then aggregate and present to users.  This is a compelling idea, popular among “Gov 2.0” evangelists including, to a certain extent, myself.  But it will take some time to get there.