Attack of the Pod Penguins 6: MP3 Basics

We've passed the halfway point in this series of articles on podcasting with open source tools. From the value of podcasts, to a summary of the available open source podcasting-friendly tools, to a detailed look at both Audacity and normalize, we've covered a fair amount of ground (to see a list of stories from the series, see related links box, right). We've picked our way through some of the minefields of the tools, and exposed some hidden process gotchas. We've also talked about the importance of sound quality, even in a podcast intended to emphasize its DIY roots.

In this installment, we'll focus on encoding your final edit to a format that's easily transportable across the Web. But, we're only going to do that in an indirect way. This isn't so much a "how-to" installment as it is an important background installment. There's some deep water ahead, as we troll the waters of podcasting with open source tools. We've reached a point that requires a discussion of licensing, for example - a discussion that lies at the heart of open source. We also need some stronger background in audio theory in order to understand how to use the limitations of human perception to effectively minimize file size. We'll move back into practical application in the next installment. This installment, in more ways than one, is about your brain.

Let's start with licensing. Licensing is everything in open source. Licensing is the place where the concept at the heart of our favorite free and open applications and operating systems comes home to roost. Not only is it a vital thread in the open source tapestry, it's the loom and the weaver, as well.

That necessarily places podcasters intent on sticking to the critical principles of open source squarely at a crossroads. Simply put, the most popular Web-portable audio file format is not open source. The most popular format for audio on the Web is proprietary, its license held (in some dispute) by large corporations.

Therein lies the crossroads. Free and open source codecs for audio have been completely eclipsed in popularity by the MP3 format, licensed jointly by Thomson and Fraunhofer. It's the format the vast majority of podcast listeners expect. It's also the format most universally read by audio players. There's a synergy there - listeners expecting MP3 and players providing MP3 capability - that has pushed adoption of the MP3 format to the top of the heap. In other words, if you're true to your open source roots, choosing ogg or Free Lossless Audio Codec (FLAC) formats, for example, you can be proud of your idealism, but it's unlikely your podcast will be heard by more than just a very few like-minded listeners.

So, is there room for compromise? Can we find a way to deliver a podcast to the most listeners while avoiding the fear of licensing costs imposed by a for-profit licensor? Let's look at some critical elements of the MP3 license.

Thomson has chosen to split their MP3 license into several sections based upon the use of the mp3 encoder or decoder: PC Software, Hardware, chips including DSPs, Games and Music Services. Each license type requires payment of a royalty. PC Software, Hardware, Games, and Music Services all require a minimum annual royalty payment. In the case of music, that payment is due in January of each calendar year, and runs $2,000. This fee is, in practice, an advance payment against creditable royalties, defined by Thomson as a running royalty. The rate of that running royalty is strictly income-based and fixed at 2% of revenue generated from the use of the MP3 format. The licensee pays the two grand each and every year, regardless of whether the running royalty ever reaches that amount. If it doesn't, they won't pay more. They just won't ever pay less. And, if the income derived from the MP3-related product exceeds $100,000, the licensee is on the hook for any running royalty in excess of the minimum paid in January.