CTO Connection: the audio edition launches!

Several weeks ago, both I and Mike Dunn hinted at something we had in the works, and the time has come to launch a new venture I am really excited about: CTO Connection, the audio edition. The first installment is an interview with Mike, and you can download the audio here, or subscribe to my new podcast feed (more on how I built the feed later). The fact that I still consider this project to be "in beta" with a few jagged edges might obscure the hours and hours I put into creating it, but in the end, I decided to accept some of the nagging imperfections and surface this experiment to get the feedback loop going.

I explain the thinking behind my new podcast series in this week's InfoWorld column (quoted in full below for your RSS full-text feed-reading pleasure):

When I think about my early introduction to Linux more than 10 years ago, the specific steps required to compile the kernel on an old 486 escape me, but I very clearly remember spirited late-night talks over coffee with the helpful co-worker who introduced me to Linux. In my current role at InfoWorld, my regular conversations with other CTOs continue to mold my thinking, both in terms of strategic technology and personal growth. Aside from small nuggets of those conversations in my weekly column and in my Weblog, much of the content of those conversations has been lost to unrecorded history, even though I'm certain you'd find many of these dialogues every bit as enlightening as I have. But I've decided to change all that.

Thanks to the relatively new phenomenon of podcasting, I'm now able to capture and distribute those conversations with other CTOs and share them with you on a regular basis. For the uninitiated, Wikipedia defines podcasting as "a way of publishing sound files to the Internet, allowing users to subscribe to a feed and receive new audio files automatically."

After four years of writing this column for the magazine and the Web, I'm pleased to announce that I'm launching an audio edition of CTO Connection, an interview series where I sit down with a working CTO and discuss the business of running and managing IT, career issues, and the mindset required to be a successful CTO. I'm hoping that this audio series will give you a peek into the kinds of behind-the-scenes conversations I have with CTOs so you can learn from them in the same way I have during the past four years. I'm looking forward to sharing the thoughts of CTOs in their own words with you.

We won't be talking about technology alone, because I don't think the path to being a successful CTO is paved purely through technical proficiency -- a balanced point-of-view is essential. CTOs are multidimensional people with passions and interests that extend beyond technology, so expect to hear a little about what CTOs do for fun.

For the first installment of the audio edition of CTO Connection, I spoke with Mike Dunn, vice president of interactive media at Hearst Interactive, a founding member of InfoWorld's CTO Advisory Council and an enthusiastic blogger. Mike's career includes stints as the CTO of Time Warner and as the first CTO of Dell's (Profile, Products, Articles) category-leading online business.

As a fellow media CTO, Mike has given me excellent advice on a number of occasions about everything from content management to collaboration technologies and techniques. As an avid consumer and sometimes-producer of podcasts, Mike inspired me to make the leap into audio and diligently worked with me to get this series off the ground.

In our conversation, Mike and I talk about how to make the transition from hands-on technologist to CTO, what Mike is learning from the ways that his kids use technology, why podcasting is important, and a little about snowboarding -- one of Mike's passions (and a personal failure of mine). You can download the 40-minute interview as an MP3 file from my Weblog. Thanks for reading -- and now, listening.

Here is some more background on the podcasting process, at least as it developed for me.

My prior experience with audio

First, a little background on my experience with audio production. While I don't consider myself a pro by any means, I have probably dabbled in audio production a bit more than the average person. Going into this process, I had some high-end headphones (Sennheiser HD280 Pro), a couple of Shure SM57 mics with all the assorted cables and adapters, mic stands, a USB audio interface (the M-Audio MobilePre), a basic sense of setting line levels, and some experience with multi-track recording, first with the Roland VS-880 (a hardware unit with which I have struggled mightily over the years) and more recently with Logic Express on OS X. For summer vacation last year, I spent a few days fooling around with recording equipment -- and I considered it an absolutely superb vacation. This is what I consider "fun."

Interview setup

When I started the process, I initially decided that I wanted to go for the "easy" and "simple" approach so that when I released the podcast, my effort would serve as a model for quick-and-dirty podcasting. As I got deeper, I found myself growing more dissatisfied with the production quality. Since the core of my podcast was going to be the interview with Mike, I decided to capture the interview first and build around it, adding elements like background music and voice-overs later on. I knew that Mike used Skype and I had a cheap Telex USB headset, so I decided to use that to conduct the interview, following Glenn Fleishman's very thorough and excellent explanation of how to capture Skype audio with OS X (also, be sure to take a look at Hugh Schotman for more detail on this kind of setup). As you will hear, the interview audio, though a bit "tinny" at times, is generally not that bad considering the inexpensive setup I was using. There are occasional bits of Skype-related distortion and the levels of my and Mike's voices aren't quite what I would like always. Occasionally, you'll hear me unconsciously exhale a bit harshly into my cheap Telex mic, which can be annoying to the listener (especially to the listener who is editing the audio, i.e. me.) I am hoping that the hours spent listening to and editing the interview audio have mostly cured me of that tendency (the second interview I did -- to be announced later -- does sound better in that regard).

Music

I'm an obsessive music fan, and when I first started thinking about this podcast, theme music was surprisingly high up on the list of problems to solve. I was driving around one day listening to Deee-Lite's World Clique album (known mostly for the cut "Groove is in the Heart," which is indeed a song for the ages) and decided that a sample from the song "E.S.P" would be perfect (the section between 0:15 to 0:30), though I knew that licensing would be difficult. I decided to check out licensing through traditional channels like ASCAP. Whoa. Since InfoWorld.com is a revenue-producing entity, I was looking at a license fee in the mid five-figures (or higher) which was too high a price to pay for what amounted to an experiment. Not satisfied to have no music, I began some Google digging and came across Doug Kaye's posts about the subject (here and here) and that nudged me towards Magnatune, an innovative Internet-based record label (read founder John Buckman's "Why I Created Magnatune Records" for the full scoop). For those of you not familiar with Magnatune, their tagline is "we are not evil," and when I sent an e-mail to John inquiring about licensing Magnatune music for my "commercial" podcast ("commercial" in all ways but the critical "making money" part), I got a decidedly non-evil response with an extremely reasonable offer to license the music, which I accepted (be sure to listen to the IT Conversations interview with John Buckman here). The deal with John and Magnatune set off a binge of music-listening that eventually led me to sample a bit of Electric Frankenstein's song "New Rage" for the lead-in music to my podcast. Ultimately, I would have liked to have spent more time listening to the eclectic Magnatune catalog and working music into my podcast more fluidly, but it was starting to distract me from my original intention: producing a compelling interview show. In any case, I came away from my dealings with Magnatune utterly impressed -- I encourage any music fan out there to support them. Not only did John deal with my request quickly and thoughtfully, he freely gave me a lot of detail about a presentation he is giving at RedHat Summit 2005 on June 1st entitled "Surviving Slashdot" in which he describes some the limitations of the LAMP platform from his experiences with Magnatune (John is also CEO of Lyris, probably the oldest and best-known commercial provider of high-performance e-mail list management software and services). [Update: John Buckman wrote to let me know that he sold Lyris on Monday. Congrats!]

Editing, voice-overs, and mixing

I wanted to do some simple editing of the audio I had gathered in the interview, and a few searches led me to Amadeus, a simple waveform editor for OS X. Amadeus turned out to be a nice tool for quickly snipping out bits of audio. (I found the sample editor in Logic to be a little too slow for my tastes.) Unfortunately, Amadeus only accepts AIFF files for editing, so I had to convert my lossy mp3 file from my Skype session to AIFF, which is far less than ideal. Live and learn, right? In any case, by the time I had completed the interview portion and got deep into editing it, I decided that I was no longer satisfied with the sound provided by my cheap Telex USB headset, so I recorded the parts before and after the interview with the Shure SM57 mics running through my M-Audio MobilePre going into Logic Express. I could go on endlessly about mics (try this if you're interested), but I have always heard over the years that the SM57 is a good all-purpose mic, and I just happened to have a couple on hand. There are better mics, of course, and I'm actively evaluating a few options. I'll save that discussion for later. In any case, the voice-overs with the SM57 setup are superior to what you will hear with the Telex USB headset setup I used for the interview. I dumped the edited AIFF file of the interview with Mike into one track on Logic Express.

Overall, I ended up with six tracks of audio: 1) the intro music, 2) my voice-over with the intro music saying "You're listening to CTO Connection. . . ", 3) my voice intro about the interview with Mike, 4) the interview itself (both my and Mike's voice were on the same track because of the Skype setup, 5) the outro music, and 6) my closing comments on the interview. I did some minimal mixing on the tracks and bounced it to the stereo 96kbps mp3 you will hear as the final podcast.

Posting the final product

As luck would have it, we're still running Movable Type 2.63 here at InfoWorld, which doesn't natively support RSS enclosures out of the box (check out MT-Enclosures for one option). While upgrading to 3.x is on our list, I decided it would be simpler to run my current RSS feed through Feedburner's SmartCast service, which converts the first link to an audio file into an enclosure. My Feedburner SmartCast podcast feed is therefore at http://feeds.feedburner.com/infoworld/dickerson/podcasts.

Looking ahead

Serendipitously, I had the pleasure of being invited to a lunch with Doug Kaye of IT Conversations a few weeks ago, right after I had done the first interview with Mike Dunn (Kevin McKean, our CEO, had been talking to Doug about helping us produce InfoWorld-related audio content). I explained to Doug some of the trouble I had with my Skype setup and he suggested that I take a look at a "digital hybrid," a piece of hardware that separates the audio from two sides of a phone conversation into separate tracks. Digital hybrids are used by call-in radio shows with either regular analog lines or PBX systems. As I understand it, when you separate the two sides of the conversations into separate tracks, it makes it much easier to clean up either side of the conversation, adjust the levels, clean up noise, etc. Also, one side of the conversation (usually the person on the other end of the phone) is usually quieter, and digital hybrids help you adjust for that. I'll be checking out at least one digital hybrid very soon, so look for another post on that subject. Doug also gave me some advice on microphones and other equipment and pointed me to the public IT Conversations wiki. As I get deeper into production quality issues, I'll post my thoughts as I get my hands dirtier and dirtier.

Finally, I'm tagging interesting podcasting items in del.icio.us with the tag podcasting. Many of these items point to particular pieces of gear, and some point to content, but all have been useful to me in some way as I've been working on this project.

I never thought I would say this in my role at InfoWorld, but "enjoy the show" -- and let me know what you think.

Posted by Chad Dickerson at May 10, 2005 01:10 AM