This post has nothing to do with administration other than an attempt to manage a set of loose ends I’ve been contemplating to post. This post’s title comes from Tom DeMarco‘s excellent project management book called The Deadline.

  • Isolation is not always bad; alignment is not always good.
  • Themes and rhythms (cross industry scenarios and product solutions, respectively)
  • If you want to know where your raise went, read this. Key point: Benefit costs are driving a wedge between the revenues that businesses collect from their products and the size of the paychecks that workers receive.
  • One of the articles that inspired my why isn’t collaboration free? question in my previous post was this op/ed piece in CRN.
  • Confidence (n.) – the sweet spot between arrogance and despair (from Confidence: How Winning Streaks and Losing Streaks Begin and End). Arrogance involves the failure to see any flaws; despair involves the failure to acknowledge any strengths.

I seem to have started two new workflows in the office: (1) using a digital camera to capture whiteboarding, etc. and (2) using a combination of tools to record meetings, knowledge transfers, etc. Processing the pictures I take for (1) is easy enough via Office 2003’s Picture Manager or in a few cases via Adobe Photoshop CS. However, although I have a pretty effective process for (2), it certainly would be nice to automate it.

I use Microsoft Office OneNote 2003 to capture the original voice recording in (WMA) format, which offers good quality and good file size (e.g. an hour meeting consumes 6~7 MB). The original recording is processed in Adobe Audition as follows: center wave, normalize (90%), amplify (varies on original), obtain then apply noise reduction profile to entire waveform, remove whitespace, remove um’s, hmm’s, and’s and other speaker extras, maximize amplification without distortion, and finally save a copy of the processed original to MP3 format. The final product can then be replayed faster than real-time (e.g. 1.5x) as time-shifted audio.

While effective, the above process is somewhat tedious and hands-on. It would be great to have a program that could be trained to recognize the speech patterns of individuals, categorize those profiles, and apply them to subsequent recordings.