I’m : a programmer, writer, podcaster, geek, and coffee enthusiast.

The Daily shutting down

The Daily didn’t fail because the iPad isn’t a viable publishing platform.

It didn’t fail because apps are inherently worse than websites for publications.

It didn’t fail because too few people want to pay for good content.

It didn’t fail for any technical implementation or design decision they made.

The Daily failed because what they chose to make, with its huge staffing costs, required far more than their 100,000 subscribers to be financially sustainable. And it didn’t attract more subscribers because what they chose to make was, itself, deeply flawed.

We all wrote this article when it launched nearly two years ago. Here was mine:

First, it’s weird to me, as a long-time internet-only news reader, to pay money for a bunch of content I don’t care about. More than half of each issue is sports news, entertainment gossip, ads, and little newspaper games (crosswords, Sudoku, horoscopes), and I need to buy all of that to get the news, editorials, and app reviews that I care about.

Bundling a bunch of stuff I don’t care about with the few pieces I want to read is the old-world model, when custom-targeted or on-demand news for each reader was infeasible. But in this century, I can go to a handful of websites whenever I want news, view the handful of stories that interest me, then move on. Flipping through a bunch of uninteresting-to-us content and ads was an annoyance of the old world, like blow-ins, that we tolerated because we had to — but now, we don’t.

These old-world annoyances would be easier to ignore if the content was great, but it’s not. It’s acceptable, for what it is: a very lightweight rundown of the previous day’s most mass-marketable news, with one or two editorials that usually leave me wanting more depth.

There’s no room in today’s market for publications like The Daily, and their heyday ended long before the iPad launched.

Well-established news sites are much better for news. Editorials and feature articles need to either be free, like most blogs, or consistently great and worth paying for, as in magazines such as The New Yorker or The Atlantic. But The Daily offered an overreaching mix of ineffective news coverage and unmemorable editorials and features. I’ve never seen anyone share a link to something in The Daily saying that we had to go read this great article that would make us want to subscribe. (In fact, I’ve simply never seen anyone post a link to anything in The Daily.)

The Daily required an extremely large staff to produce. And even with supposedly over 100,000 subscribers, netting them at least about $3 million per year plus ad revenue, that’s simply not enough to pay for a staff that large. (Not even close.)

Some writers are citing traits of my new iOS publication, The Magazine, as what The Daily should have been, but it’s not a fair comparison: we’re not doing the same things at all. The Magazine works because I chose a niche that can be served by a very small staff, I applied a very small staff to it, and it appeals to enough people to support the cost of a very small staff.

As The Magazine’s subscriber base has increased beyond the threshold of profitability, I’ve been reinvesting in it: I’ve raised the writer payment rate, hired an editor, and added more articles. This week’s issue will include the first illustration. Soon, I want to be able to fund more in-depth articles and even investigative pieces.

The Magazine costs less than The Daily and has far fewer subscribers (so far), but that’s fine: I can’t even imagine how I’d spend $3 million per year on it. But I’m also not trying to make an all-purpose news and editorial publication for everyone, every day.

A long time ago

When The Simpsons first aired in December 1989, I was 7 years old and my sister was 10. It was so good that my mother let us stay up “late” to watch it at 8 PM. Every week, we’d watch The Simpsons as a cherished family activity.

Remarkably, the show is still running, and in last night’s episode, “The Day The Earth Stood Cool”, The Simpsons made a Tumblr joke. The site that David and I started when I was 24 was referenced in the show I started watching when I was 7.

I haven’t watched it in over a decade, but this feels like a major accomplishment.

My master plan for revolutionizing the future of publishing and saving tablet-native journalism

I have a lot of respect for the publishing industry and the futurists who try to predict where it’s heading. Nobody in that business has it particularly easy. But with The Daily dying so soon after I launched The Magazine, many writers have gotten the wrong idea: that The Magazine was made to show the industry “how it’s done”, or to “save journalism”, or as part of a grand plan of mine that somehow involves both Instapaper and The Magazine. None are true.

The Magazine and Instapaper operate independently. There’s no master plan. I wanted Instapaper to exist, so I made it. Five years later, I wanted The Magazine to exist, so I made that.

I don’t know how to save journalism, but I’m also not qualified to. I’m not a journalist and I don’t know much about that industry.

I also have no interest in showing the periodical industry “how it’s done”. I set out to create this magazine, not make a template for other publishers to follow.

But it’s working (for this magazine, at least), so naturally, many publishers and platforms have already started similar-looking efforts. I get daily inquiries from people who want to license The Magazine’s platform to get a head start, a business that I don’t think I want to be in.

In fact, I really don’t want a bunch of other Newsstand magazines to launch that look just like The Magazine.

I want The Magazine to be the only publication that looks like The Magazine. People should recognize the style as uniquely The Magazine’s. Cloning it to death would only dilute what I’m trying to build.

More importantly, I want The Magazine to have a reputation for high quality, and not to have a reputation for “just” being a bunch of blog posts behind a paywall. It was made to have magazine-quality articles: that’s why I hired a great editor, why we’re paying print-competitive rates to attract print-quality writers, and why we’re starting to integrate illustrations and photos where they can be beneficial. We’re still finding our way in some areas, but we’re on a strong path.

The last thing I’d want is for a bunch of The Magazine lookalikes to flood the App Store with mediocre articles that haven’t passed through an editor and should just be (or already are) someone’s mediocre blog posts, just so they can easily charge for a subscription. There’s a time and a place for less-formal, less-polished blog writing — here and now, for instance. But there are plenty of reasons why The Magazine isn’t just Magazine.

If the App Store gets spammed with hundreds of bad clones, The Magazine itself will lose credibility and potential subscribers as people make incorrect assumptions about its article quality.

That’s why The Magazine isn’t part of a bigger strategy: I don’t want everyone to rush into this model, making an app that looks and works just like mine. I don’t want to make “the WordPress of Newsstand”, because I don’t want it to be that easy to copy The Magazine.

A publication’s app should be designed and built with purpose and consideration. The Magazine works because I based decisions not on what everyone else was doing, but on what would be best for this magazine. Every publication has its own unique needs, audience, economics, and style, so their apps should reflect that.

In the past, publications had a harder time differentiating themselves. Magazines and newspapers all needed to be the same sizes and shapes, working the same ways with the same business models and the same limitations. Today, we can all tailor our publications to our needs much more closely.

“Tablet-native” publishing shouldn’t mean any particular multimedia features or structures. True tablet-native publishing should mean using the freedom of modern platforms to break out of the idea that publications need to follow a universal mold. They’re all just software now, and a unified platform would only limit the possibilities.

Simply cloning a few successful formulas would be a tragic waste of this potential.

Loading iOS fonts dynamically

I recently mentioned on the podcast that some of Instapaper’s font licenses require me to encrypt the font data, so I need to decrypt and load them dynamically instead of the usual method of putting unencrypted OTF filenames in the UIAppFonts key in Info.plist. Well, it Turns Out™ that almost noboody knows you can do this. I didn’t, either, until a programmer from one of the font foundries told me about CTFontManagerRegisterGraphicsFont.

Here’s a usage example:

NSData *inData = /* your decrypted font-file data */;
CFErrorRef error;
CGDataProviderRef provider = CGDataProviderCreateWithCFData((CFDataRef)inData);
CGFontRef font = CGFontCreateWithDataProvider(provider);
if (! CTFontManagerRegisterGraphicsFont(font, &error)) {
    CFStringRef errorDescription = CFErrorCopyDescription(error)
    NSLog(@"Failed to load font: %@", errorDescription);

You can load a font this way at any time during your app’s execution, and it immediately becomes available in UIFont and UIWebView (via regular font-family CSS declarations, no @font-face required) just as if it had been declared in UIAppFonts.

How you encrypt the font data in the bundle is up to you, but since you may need to decrypt hundreds of kilobytes of font data, you should choose a method with fast decryption.

Also, be careful what you agree to in your licensing contracts. Like any other method of app-resource encryption and obfuscation, you can’t guarantee that nobody will ever be able to copy the unencrypted data out. You can only make it harder, make it take much more effort, and reduce the number of people who can do it.

The Leica M9, as a pro-hobbyist photographer

As a gift to my wife, I rented a Leica M9 with the Summicron-M 35mm f/2 ASPH lens, a setup that would cost nearly $10,000 to own.

Most fans describe Leica’s benefits in vague, unverifiable terms, much like a wine aficionado describing the taste of a $200 bottle of Pinot Noir.

From my wife’s photography business, we already own a pair of Canon 5D Mark IIs and a lot of great Canon glass. The 5D Mark II and the M9 both have full-frame sensors, and the M9 is only one year newer. But the M9 costs more than three times as much as the 5D Mark II today, and the Summicron-M 35mm f/2 ASPH costs almost four times as much as Canon’s best comparable lens, the new 35mm f/2 IS USM. So what’s the draw of the Leica?

We were curious to try it. We’ve seen a lot of amazing photos from amazing photographers using the M9, and surely, there has to be some reason why Leica can command such a premium price. Our family Christmas vacation in the mountains was a perfect opportunity: outdoor scenery, indoor family moments, and lots of free time.

The biggest M9 difference is that it’s a rangefinder rather than SLR: instead of looking through the lens when composing and focusing, the viewfinder on the M9 looks through its own fixed lens diagonally above the “real” lens. The viewfinder includes approximations and indicators to guide the photographer into properly focusing and composing the frame, but it’s nothing like composing a photo by looking through the lens as with any other camera type on the market today.

There’s also no autofocus, and since you aren’t looking through the lens while composing, the only ways to know if you’re focusing on the subject are to estimate distance or line up a ghosted clone of the image in a portion of the viewfinder.

It’s a very manual experience. While there are some other advantages to the rangefinder design, I think this is most of their modern-day appeal: nostalgia, retro fashion, and “slow” photography. As so many will say, even after I became proficient at focusing, the Leica does indeed force me to spend more time setting up each shot.

But that also means that it’s very difficult to capture fleeting moments with it. For spontaneous cuteness and family memories, the Leica is barely usable unless you have a very patient family.

It’s also just poorly suited to indoor use. The sensor is mediocre at high ISOs, especially compared to other full-frame sensors (even in 2009). To meet my standards for noise and detail, the highest sensitivity I could use on the M9 was ISO 1600 (I generally stop at ISO 2500 on the 5D Mark II). By keeping the lens wide open at f/2, I was able to get just enough light to hand-hold indoor evening shots, but the narrow depth of field also made focus errors much less forgiving. With a rangefinder, you need as much focus forgiveness as you can get.

The rangefinder design also hurts short-distance framing: since you’re not looking through the lens to compose, it’s easy to think you’re framing a shot differently than the resulting photo shows. The closer the subject, the harder it is to frame the shot as intended. (This lens also can’t focus closer than 0.7 meters. By comparison, that Canon 35mm IS USM can focus to 0.24 meters.)

The M9 is obtuse in other ways, too. It has almost no controls and zero convenience features. Many of its buyers probably demand that simplicity, so I can’t fault Leica for that. But the M9 also has slow shutter response, a slow image processor, slow shot-to-shot time, poor battery life, and a small, low-resolution screen, even by 2009’s standards. The 5D Mark II is luxurious, fast, and extremely responsive by comparison.

The M9, then, is hard to recommend. It certainly doesn’t feel like a $6,500 camera.

The new Leica M240 adds live view and a larger screen, which will be a huge improvement in some areas. But given the M9’s unremarkable sensor, poor performance, and other shortcomings, I don’t think it will feel like its $7,000 price either.

But great photographers probably don’t buy the M9 because it’s a perfect camera body: the much bigger draw for Leica is the glass.

Optically, this lens is extraordinary, which seems common among most Leica M-mount lenses.

Every time I use the M9, I want to throw it out the window. My wife, who this rental was for, barely used it at all before getting frustrated and going back to the 5D Mark II. (“I think this is too hip for me.”)

But when I nail the focus, the timing, and the composition — which isn’t often — I say, Holy shit. This lens is incredible.

I tried to nearly replicate one of my Leica landscape shots with the 5D Mark II and the Canon 40mm pancake — a $3,200 lens against an (unusually good) $150 lens — so I could somewhat objectively evaluate the Leica optics. The Canon pancake performed surprisingly closely. The Leica did have a slight edge in sharpness and CA, but I wouldn’t call it a $3,050 edge.

Like the M9, it’s hard to justify the cost of this lens. But unlike the M9, it really is extremely good.

Our Leica rental experiment is nearly over now, but my conclusions were clear after the first day:

I think it’s time to rent a Sony RX1.