M3

It’s been about six months since the release of iTunes Match and I want to check up on how it’s doing. I listen to a lot of music, use my Mac and iPhone a lot, and have been very excited about this feature. I also used my iOS Developer Account to test the feature before its public début.

iTunes Match has made noticeable improvements but is still far from flawless:

• When it first launched, listening to songs over 3G while changing the song more often than about once a minute would easily overtax available bandwidth and iOS did not handle this well. While there’s little iOS can do about available bandwidth, iOS 5.1 does seem to cancel unnecessary downloads quicker. Apple also seems to have made the ready-to-play buffering calculations a bit more aggressive.
• Metadata syncing did not always seem to work: often, I would notice low resolution artwork on my retina iPhone, update the artwork on my Mac, and despite waiting a few days, the artwork was often not pushed. While I notice metadata syncing issues less and less, iOS still makes no attempt to, say, sync all artwork while charging on WiFi overnight.
• Matching is still an imperfect science, and there’s no way to give iTunes Match any hints. One of my favorite albums was recently remastered and released on CD, but not on iTunes, leaving Apple’s store with the older version. iTunes Match seems not to detect this, even with metadata hints (e.g. adding ‘Remastered’ to the album name), making syncing this remastered version of the album to iOS devices impossible short of turning iTunes Match off.
• iTunes Match downloading still seems a bit unfinished: ‘streamed’ songs (not Apple’s terminology) download through the iTunes app, which you can see for yourself by opening the Downloads section of that app. Also, the circle animation Apple uses to show that a song is loading still gets stuck from time to time, although less so on iOS 5.1/5.1.1.
• Songs not downloaded/cached on your iOS device do not seem at all through iPod stereos. This may be a limitation of the old iPod dock connector API. Apple has indicated through their bug reporter that their engineers are aware of this issue.
• iOS 5 automatic application cache cleaning apparently implies clearing of iTunes Match downloaded cache. This can be more than a little annoying if you are using your iOS device through an old iPod stereo (e.g. a car stereo) and songs you explicitly downloaded disappear before a long road trip.

Apple is clearly developing this feature in an agile-fashion: Genius Playlists were flat-out unsupported on iTunes Match until iOS 5.1 came out a few months later – the feature simply wasn’t ready on the launch date.

I can’t help but feel Apple felt it better to get a mostly working version of iTunes Match out at the same time of the Google Music launch to make sure they were not significantly threatened by the competition, and would otherwise have waited until iTunes Match was a bit more mature.

It’s a great feature at a reasonable price, but all these little issues means I still can’t recommend it to friends and family. Maybe in another six months.

The innovation, imagination, research, and development which occurs at public universities begins new industries, new ideas, creates and improves technology, creates business and grows the economy. Their value is priceless. Until today (pdf):

University of California President Mark G. Yudof, California State University Chancellor Charles B. Reed and Community Colleges Chancellor Jack Scott met with lawmakers and aides to Gov. Jerry Brown today to urge them to begin reinvesting in one of California’s highest-performing assets – its three systems of public higher education.

…

“The University of California alone generates $46.3 billion in economic activity in California and supports one in 46 jobs. That’s quite a return on the state’s investment,” President Yudof said. “Yet nearly every year our state funding drops. We estimate that every $1 cut from UC’s budget could reduce California’s economic output by $2.”

…

“We made it perfectly clear to everyone we met today that California’s future relies on an educated and well- trained workforce,” Chancellor Scott said. “But every time we take a budget cut, the state denies or delays a student the chance to become a valuable member of that workforce. We need to remind those who serve us in the Capitol that for every $1 they invest in higher education, the state receives $4.50 in return.”

I still don’t know what every common job title in a movie’s credits does. Director is obvious, actors too. Key grip, though? Unit manager?

Regardless, their segmenting of work is important. I think this separation of concerns is what makes commercial software feel more polished at times. Yes, money plays a huge role in corporate software often developing faster and with higher quality assurances. But this segmenting of work allows for individual roles to grow in talent and strength.

I don’t think programmers make the best UX designers. They certainly often are not the best artists. They’re terrible producers too, often re-writing large amounts of code to make themselves, not necessarily the end result, feel better.

Open source projects should be developed like a movie: there should be a producer, a UX designer, a graphics artist, etc.

I know some projects do have these roles, but I think it’s important: let the lead programmer make the technical programming decisions, but don’t dare to let him decide what a button should look like, or where it should go. Let a producer, not a programmer, not a graphics artist, figure out when the best point of delivering software should be.

My intent is not to knock the usual open source development process, but rather to highlight the seriousness and difficulty of these other roles.

We see programmers proudly declaring themselves advocates of open source: why not UX folks? Why not graphics designers?

We need to let them in.

Oracle is suing Google over its alleged use of Java intellectual property in Android. The tech press is pretty sure this is about Oracle wanting a slice of Android handset profits. That may be true, but I think fees are the icing on the cake.

I think this is about Java, not Android.

When Oracle purchased Sun Microsystems in 2009, it was pretty clear what Oracle wanted was Java. Oracle’s own database has large swaths implemented in Java and Java EE developers needing enterprise-grade database solutions surely make for a good business for Oracle. But Java hasn’t been doing well lately.

Sure, it’s the world’s most widely used computer language, mostly by a large amount (C++ technically ties it almost neck-and-neck but the two languages main use cases are somewhat disparate). But Java’s development has been slow: the language isn’t gaining features as fast as it should in the face of competitors (C# namely, which is often paired with Microsoft’s SQL Server, not Oracle) and one could make a strong case that a healthy Java means a healthy Oracle.

So Oracle bought Java, and accelerated development. Java 7 was released in 2011, with Java 8 coming in 2013.

And while Google using Java for Android makes the future of Java look bright, it doesn’t make Oracle’s purchase of Java worth much more: Google re-implemented Java, including its major APIs and the runtime virtual machine itself.

Google is an honest threat to the control of Java. Google could come to develop and support a new Java paradigm of Google data sources and Google’s Java, making Oracle’s Java purchase, and existing business, much less valuable.

Oracle v. Google is about Java, not Android. I wish more in the tech press would realize this.

Samsung, Intel, the Linux Foundation, and others joined in 2011 to form a new mobile operating system called Tizen. Tizen succeeds Samsung’s Bada platform and Intel/Nokia’s Meego platform, and like Android, is a Linux-based open source mobile operating system.

Recently, we saw a handset actually running Tizen – admittedly an early look. John Gruber, among others, wonder why Samsung, with its very successful Android phones, and others are even bothering to develop Tizen.

Here’s my guess:

Tizen is markedly different from Android. While both can hold the labels “open source” and “Linux-based”, Tizen is community-developed. You can actually download up-to-the-second source code from developers as they work day-to-day. The same cannot be said for Android, which Google develops in secret at and then gives to the community, cutting out community involvement by design.

I do think independence from Google is important to Samsung and Intel, but the real reason is more subtle. These companies want control.

Take Linux (the kernel) itself. Major contributions (hundreds of millions of dollars over time) are made from dozens of very large companies: Intel, AMD, Red Hat, HP, and others. Over 75% of the code developed for the Linux kernel today comes from paid developers – an impressive feat when you realize the Linux kernel hires no one and is owned by no particular company or group.

The Linux Foundation strongly backs Tizen and surely wants it to grow into this kind of project. Smartphones, tablets, in-car computers and other post-PC devices have few or no credible open source alternatives. Android surely isn’t one.

Google just wants to ensure Google services are still heavily used on smartphones. Samsung doesn’t want to be behold to Google for an operating system. Intel wants to ensure a popular smartphone operating system runs on its upcoming mobile chips. These folks can all have their way. All it takes is a good, long look at kernel.org.

Neil Young issued a few potentially misleading statements this January at the D: Dive Into Media conference:

As Young himself cited, even CDs offer only 15 percent of the recording information from the master tracks. Pare that down to 256 kbps files, and you’ve lost a great deal of richness and complexity in audio.

Young seems to sell the AAC+ format a bit short. Fortunately, Ars Technica has published an excellent piece about Apple’s recent “Mastered for iTunes” program and praises the format:

Some musicians and record executives have recently bemoaned the fact that what ends up on a fan’s iPod or iPhone is of arguably much lower quality than what is laid down on tape or hard drives in the studio. While some players in the industry have pushed for higher resolution downloads, Apple’s current solution involves adhering to long-recognized—if not always followed—industry best practices, along with an improved compression toolchain that squeezes the most out of high-quality master recordings while still producing a standard 256kbps AAC iTunes Plus file.

…

We came away from the process learning that it absolutely is possible to improve the quality of compressed iTunes Plus tracks with a little bit of work, that Apple’s improved compression process does result in a better sound, and that 24/96 files aren’t a good format for consumers.

The article also cites an excellent essay I hadn’t come across before from Ogg Vorbis creator Christopher “Monty” Montgomery:

Articles last month revealed that musician Neil Young and Apple’s Steve Jobs discussed offering digital music downloads of ‘uncompromised studio quality’. Much of the press and user commentary was particularly enthusiastic about the prospect of uncompressed 24 bit 192kHz downloads. 24/192 featured prominently in my own conversations with Mr. Young’s group several months ago.

Unfortunately, there is no point to distributing music in 24-bit/192kHz format. Its playback fidelity is slightly inferior to 16/44.1 or 16/48, and it takes up 6 times the space.

There are a few real problems with the audio quality and ‘experience’ of digitally distributed music today. 24/192 solves none of them. While everyone fixates on 24/192 as a magic bullet, we’re not going to see any actual improvement.

I do not understand what Neil Young intends on contributing to this problem. It’s good that he wants to improve digital music, but irksome that his narrative detracts from the impressive work already done.