Month: January 2014

For the first time since the iTunes Store was opened, digital music downloads decreased in 2013. Most analysts and industry publications, like Billboard, attribute this to a huge increase in the use of streaming music services. But Horace Dediu of Asymco sees it differently:

Consumers have a fixed time budget, a more rigid constraint than their spending budget. Competition for a slice of a consumer’s time budget is far tougher than competition for a slice of a consumer’s wallet. […]

Downtime or “boredom” was filled with app interaction. This includes some social media consumption. These are not immersive experiences. They are “casual”, inconsequential and trivial. At first anyway. And that’s the rub. As apps enter a consumer’s world they initially take on non-consumption, which is easy to beat. But as the experiences become increasingly compelling they “move upmarket” and compete more aggressively with existing media consumption patterns.

In a nut, Dediu claims that apps took over the time previously occupied listening to music; i.e. he believes that less time overall is being spent listening to music. I don’t buy it, and Dediu’s own words illustrate why:

Music would appear less vulnerable to this app-based substitution because it’s an eyes-free experience and can be easily consumed concurrent with another activity. There is however, a subtle de-emphasis and hence devaluation of the experience as it fades into the background. The “job it’s hired to do” changes.

I don’t see how this doesn’t contradict what Dediu claims earlier in the article. Even if music is being played in the background, it’s still being played. People still want new music, even if it’s to be listened to in the background. As he points out, music is playable concurrently with just about anything else (aside from, say, a movie), so this paragraph seems to directly contradict his premise that people are listening to less music.

In fact, I’d wager that more music was played in 2013, if anything. The rise of streaming services means that it’s possible to discover more music for the same price, and it can all be done in the background. Services like Spotify Radio and Pandora ensure users are hearing a constant steam of music.

Dediu’s analysis makes little sense to me. I’d be very surprised if people were listening to less music; I’m not surprised that people are buying fewer songs when streaming services make it easy for many people to never purchase another song again.

The Times released their almost entirely overhauled website today, and it’s pretty impressive. They updated the home page, section pages, and article pages; Times Topics and some slideshows remain unchanged, however.

As this is one of the websites millions of people — including me — visit many times every day, any design changes need to be executed with elevated care. The design needs to reflect the Times’ status as a newspaper of record, too, so it has to be “right”. A redesign is a tall order, but I think they’ve mostly succeeded.

The home page retains much of former Times design director Khoi Vinh’s grid obsession, but everything gets a little more space. As a result, it scans much easier than the old, cramped page. In addition to the sections in the header, there’s a nice new flyout menu on the left which, during testing, featured a customizable “shortcuts” section. The ability to customize shortcuts has been removed in the version which went live today, though I hope it returns. This menu replaces the previous static column on the left; it hasn’t been universally praised, but I think the additional click required to open the flyout is a small sacrifice to open up the home page’s content area.

Top: existing typography; bottom: my bold-only version.
NY Times comparison
The typography has generally been improved, too. Menus, section titles, and other accent elements are set in Franklin Gothic, while headlines are set in Cheltenham. I’m not keen on the extensive use of the bold italic weight of the latter: bold or italic text should be sufficient, but using both seems like overkill.

The section pages are also new. They look much brighter than the outgoing version (noticing a theme?). The brightness and space makes it much, much easier to read, and the revised typography looks spectacular on the Retina display of my iPad, though the book-weight variant of Cheltenham is a bit hard to read at smaller sizes on my non-Retina Mac. Despite the additional space elsewhere, though, the title in the header feels comparatively cramped.

The new individual article pages are perhaps the most significant beneficiaries of the redesign. Gone are multipage articles and floating images, replaced with a simple, inline column of text and graphics which are either centred, or which sit in the column of white space to the right. I’m very pleased to see that the issues I raised in the preview stage have largely been rectified. Scrollbars have returned, and the strange friction coefficient has been eliminated. I didn’t save a copy of the prototype’s source code, so I’m not sure if they simply eliminated their odd overflow div implementation, or if they massively improved the Javascript used to run it. Also nice: comments are hidden by default.

I’m very impressed with this redesign. It’s comprehensive and contemporary, but it doesn’t feel trendy. I’m looking forward to seeing it every day.

Semil Shah (full disclosure: I have an ongoing business relationship with Shah, but he didn’t pay me in any way, or make any requests for this post):

Jelly harnesses existing social networks to more efficiently “route” questions to those who have a great likelihood of responding and/or knowing the answer. Quora does a great job of this on the web, and it’s a very hard problem to get right. Takes years. The app has a nice touch of pulling the avatars of people you recognize, which creates more of an incentive to engage and respond.

This, I think, is what defines Jelly, and what separates it from asking a question on Twitter, for instance. It’s not just your friends, but your friends’ friends. This huge network of people gives you a vast pool of potential answers. Time will tell whether coupling this to a fun interface is enough to, uh, preserve the attention of users.

So far, my least favourite aspect of the app is that everything remains locked into it. There’s no way for me to export all the answers, or even a permalink for a question (if you “forward” a question via email, the link is only temporary).

Kara Swisher, Recode (the slash can bite my ass):

In an interesting new study that measures consumer brand strength, Microsoft reigned supreme while perennial winner Apple lagged behind. The reason? The Seattle area software giant was deemed just-us-folks, while the Silicon Valley style icon was unapproachable.

Huh? Let’s let UTA Brand Studio’s Larry Vincent explain (emphasis mine):

“Brand Dependence is a distinct measure of brand equity that determines how much people see a brand as being like themselves, sharing their values, and being an indispensable part of their lives. It is driven by brand-self connection (how much a brand is like me) and brand prominence (how automatically thoughts and feelings about a brand come to mind). It has been shown in numerous studies to be a distinct and better predictor of behavior than typical measures of brand attitudes (‘I like that brand,’ ‘That brand is cool,’ etc.)”

What studies, where?

Also, it seems funny that neither UTA nor Recode disclosed that Larry Vincent used to work on Microsoft’s brand strategy. Perhaps they simply forgot, but it seems pertinent.

It’s worthwhile to consider John Moltz’ lessons from surveys (in this case, one conducted by CouponCodes4U, which was as stupid as it sounds):

Consider the source. In many of these cases the source is not pushing science, they’re pushing themselves. CouponCodes4U was interested in one thing: getting its name in publications like the LA Times. And mission accomplished.

Bingo.

Back to Larry Vincent:

“Among the ways to interpret this index, for example, is that Microsoft isn’t dead yet, and Apple is vulnerable,” he said. “Apple is vulnerable because of how polarized people were in their views about the brand. And Microsoft is still seen by many as being more like them.”

I’m sure Apple’s quaking with fear at the prospect of announcing their Q1 2014 earnings. Or giddy with excitement. It’s real hard to tell.

Biz Stone’s mysterious startup Jelly isn’t vapourware:

Say you’re walking along and you spot something unusual. You want to know what it is so you launch Jelly, take a picture, circle it with your finger, and type, “What’s this?” That query is submitted to some people in your network who also have Jelly. Jelly notifies you when you have answers.

I’m intrigued, but I’m also a little confused; how is this significantly different than simply asking that question on Twitter? What makes Jelly more compelling?

Update: Like every single person opening a Jelly account today, I asked why Jelly is compelling… on Jelly. And I got some really great responses. From “Matt”:

Perhaps because we’re presented with one question at a time vs. a fast-moving feed of tweets?

“Jesse”:

Content on Twitter gets lost in the stream. This is explicitly about questions looking for answers.

I’m skeptical, but I think this feels a little special. The app is really fun to use, and feels instantaneous. I’m hopeful.

Randall Sullivan, Wired:

At 4’11” and just over 100 pounds, Michelle Gomez doesn’t look like the sort of person you’d hire to retrieve earthmoving equipment stolen by a Peruvian crime family. But in the summer of 2013, that’s exactly what she was doing.

A truly remarkable story.

I really hate CES. Half of the show seems like the kind of bargain-basement crap you’d find at Memory Express, or Micro Center for my American friends. About a quarter of the show is full of “revolutionary” upgrades to things which have resisted revolutionizing for decades (a computer controlled oven, anyone?). The remaining quarter is loaded with the kind of crazy technology you’ll never be able to buy.

But CES is somewhat redeemed by moments like this one. Watch as Michael Bay has a “fuck it, we’ll do it live” kind of moment (language warning, obviously), although much tamer than Bill O’Reilly’s. Couldn’t have said it any better than Gerry Duggan:

I didn’t realize scripts meant so much to Michael Bay.

I’m a little surprised that Bay doesn’t leave venues via explosion.

They haven’t solved the utility issue yet — it’s still a second notification screen for those whose idea of arduous, back-breaking work is reaching into their pocket. But at least it doesn’t look like something cobbled together from an old K’nex kit any more.

Peter Cohen, iMore:

So Apple has blessed the use of the Mac Pro on its side, but the question is why? In short, Apple is guiding for the use of the Mac Pro in server settings.

Another really great use for a side-mounted Mac Pro: the biggest live concerts are at least partially run off a Mac Pro. The computer is usually mounted on its side in a flight case, and the old Mac Pro was perfect for that. This new one will require a much more comprehensive custom solution for securing one in a flight case.

Of the rumours that 9to5Mac reported last year, the ones which were wholly original content were nearly entirely accurate. However, they continued to report third-party rumours alongside their own scoops, and these were largely inaccurate.

Here’s what I don’t understand: with their incredibly accurate original reporting, why does 9to5Mac feel compelled to share things they cannot verify, risking their cachet as a result? I suppose that’s a somewhat unfair question: they probably report unverified third-party rumours because otherwise they might not have anything to post on a given day. But does the increase in page views from these posts offset the discomfort Mark Gurman and Seth Weintraub must feel allowing this mediocre content to parasitically feed off their cracking original reporting, and the damaged reputation Weintraub’s site suffers as a result?

Update: An earlier version of this implied that Mark Gurman owns 9to5Mac. This has been corrected and clarified.

Remember .Mail? Designer Tobias van Schneider:

Sometimes you have to let go, to make room for new things & new thoughts. I like to take the start of the new year as an opportunity to let go and clean up my desk. Sadly, there are no plans to continue working on the development of .Mail app – I say this with a heavy heart, as I know that so many of you looked forward to this app the same way I did & I’m very thankful for all your support.

Bummer. I was really looking forward to this re-envisioning of the desktop mail experience.

I came across this article by way of Jim Dalrymple. I don’t mean to single out Michael Mulvey, but this kind of post gets under my nose; chiefly, I suspect, because of these two diagrams (via iDownloadBlog):

The different display resolutions of iOS on the left, and Android on the right.

Using this to illustrate fragmentation is pretty silly, given that we’ve been building websites which work at virtually any display resolution from really tiny to really huge. Granted, there are different requirements for the design of UIs for mobile apps, but a good designer is able to define how things should look at varying display resolutions. There are tools for this built into both Android and iOS. It’s more difficult, granted, but I think this is among the smallest of issues for designing on Android. As Mulvey himself points out:

The iconography lacks sophistication, the typography is derivative and there’s an overall lack of cohesion to the experience of the operating system. Android clearly feels like a system built and designed by engineers, not designers.

Then there is genuine fragmentation on Android, by way of the wide variety of versions in active use. According to Google’s stats, there are seven different API versions in widespread use — eight, if you count Honeycomb’s 0.1% share. A full 24% of users accessing the Play Store as of December 2013 were using Gingerbread, a version of the OS which is three years old. That’s the real fragmentation issue: developers must build for ancient versions of the OS. iOS developers have the luxury of developing for the most recent version without having to worry about older versions of the OS..

My favourite pet topic is back in the news. Janko Roettgers, GigaOm:

YouTube will be demonstrating 4K video at CES in Las Vegas next week, with a twist: The Google-owned video service will be showing off ultra high-definition streaming based on VP9, a new royalty-free codec that Google has been developing as an alternative to the H.265 video codec that’s at the core of many other 4K implementations.

There are two things to unpack in this: YouTube streaming in 4K, and the use of the new VP9 codec Google is developing.

The first isn’t really anything new — you can already find loads of videos on YouTube which stream in 4K. But most of the existing 4K videos on YouTube — indeed, most of any kind of video on YouTube — are dual-encoded in VP8 (WebM) and H.264. If you’ve seen YouTube’s HD offerings, you know that the video quality isn’t great: everything is extremely compressed so, while these videos are ostensibly “HD” resolution, they’re really murky. It’s the same story with 4K.

The question, then, is not only whether a different codec will make a substantial quality difference, but whether that codec will be playable at all. While VP8 has been around since 2008 and owned by Google since 2010, it is used almost exclusively by them. With VP9, though, they insist that it’s going to be better:

This time around, Google has lined up a whole list of hardware partners to kickstart VP9 deployment. YouTube will show off 4K streaming at the booths of LG, Panasonic and Sony. And on Thursday, YouTube released a list of 19 hardware partners that have pledged to support VP9, including chipset vendors like ARM, Intel, Broadcom and Marvell as well as consumer electronics heavyweights like Samsung, Sharp and Toshiba.

Roettgers makes it sound like this is a different approach than Google took with VP8. However, Mashable’s Google I/O liveblog from 2010 suggests differently:

Google is back on stage, discussing partners. Opera, Skype, Adobe, Nvidia, Logitech, Qualcomm, Texas Instruments, Theora, Brightcove, and others are part of the [WebM] project.

Despite those large, influential partners, VP8 never really caught on outside of the Google sphere. Skype is the only other major user of the codec, but they also encode in H.264. Based on what I’ve seen so far with VP9, and the support H.265 has received so far, I don’t see this playing out much better. VP9 may have the support of television manufacturers this time around, but there is no existing 4K spec which does not require H.264 and, eventually, H.265 support. Likewise, those two codecs support the Ultra HD colour space. It seems like the codec for 4K has already been decided.

I don’t dispute any of John Siracusa’s grades. Particularly this:

More broadly, iCloud still doesn’t have a good reputation for reliability, and debugging problems related to it remains difficult. If the only user-accessible control for a service is a single checkbox, it had better “just work.” iCloud has yet to earn that label.

Josh Bryant and Levi Nunnink, cofounders of Droplr:

Next week we will release some exciting new features to Droplr that we’ve been working on for a long time and that many of you have been asking us for. At that time, we will be discontinuing our free accounts. All current free accounts and new sign ups will be placed on a 30-day trial. At the end of 30 days, you’ll be asked to pay for a Droplr subscription if you’d like to continue using it. If you don’t want to pay, you won’t be able to upload any more files, but none of your existing data will be deleted, and all of your links will continue to work.

Sounds fair to me. It appears that existing customers will be grandfathered in at their current rate. I couldn’t be happier with my Droplr Pro subscription, and it appears that the company has much bigger plans than basic file hosting. It will be exciting to see that journey unfold, and I’m pleased to contribute a small part towards it.

Andrew Kim takes a look at the iBook, eMac, and white iMac (via Stephen Hackett):

I do sometimes miss Apple’s white age. The designs of this age were so calm, warm and pure, despite the brutally honest and analytical design. I especially love the way the clear layer interacts with the opaque white inner shell. However all this glossy plastic did make Apple’s products universally prone to scratches. The lineup of today stands the test of time much better.

The white iMac generations (G5 until the mid-2007 aluminum model) hold a special place of nostalgia in my heart. Right around the time of their release, I started getting into Apple and customizing OS X, and soon found the MacThemes forums. Buried in “The Lounge” was a thread for users to post photos of their desks. Setups like Glenn Wolsey’s caught my eye with their pure simplicity (though, it must be said, the keyboards and Mighty Mice of that era were abysmal).

Of course, this nostalgia doesn’t necessarily mean the white Apple era contained better products. I’d certainly prefer to have one of today’s iMacs over a white one. Aluminum and glass feel much more premium than an all-plastic body, and the displays of those iMacs were terrible. But a small part of me would like to have the guts of a 2013 iMac inside the body of a 2007 one. They’re just so nice.