Friday, December 24, 2010

TWiST #25 and TWiST-Plus: Some Tech Talk for Christmas Eve!




TWiST turns 25 today :)!


Well, actually, TWiST has now posted its 25th episode, and with the fact that we had one week where a show was missed early on due to technical issues, that means that TWiST has reached the six month mark as a weekly podcast. It's been interesting to see how the show has progressed, and more to the point, to see how the technical details of producing the show have advanced.


Someone once asked me how much time it takes to do an episode, and I've always had to say "it depends", because the total audio of the shows varies from 15 to 30 minutes depending on the guests and the topics. I've finally come to the realization that every minute of audio on average takes ten minutes of production. Why? Because I go through and scrub the audio for various things; I remove unwanted sound artifacts like pops and screeches that occasionally show up when recording digital audio. I level and adjust audio to correct for balance and to remove sibilance in places. I fix around drop-outs. I remove dead air, and I clean out "extra words".


Earlier on, when I first posted my "tech of TWiST comments, I said that I tried to walk a fine line. I edit to remove what I call "pause words" like "um, er, and "you know" so as to not chop up the interview too much. I also try to not remove the natural flow of the conversation, but in truth, I do place a pretty heavy hand on this. It's a sort of OCD thing for me; I find it to be distracting, and I find I start counting stutters rather than listening to the conversation. So if you are one to notice that the interviewer and interviewees tend to be really crystal clear, yeah, there's a reason for that :). Also, if you are really on the ball and sense that there are some clicks that appear here and there inexplicably, you're not imagining it. I'm getting better at smoothing out those edit transitions, but in some environments (noisy ones in particular), it's not easy to make edits that are seamless, but each episode I get a little better.


This week's main twist episode is with Web Performance's Michael Cziezperger. Matt met Michael at STPCON in October of 2010, and said he gave one of the most highly rated talks there. The main topic was performance testing, and dealing with the insane challenges of people in perpetual crisis mode. He also talked a bit about the program that they use and offer (including a demo version that I talked about yesterday in my TESTHEAD BOOT CAMP post. If you’d like to hear this week's episode, please go to Episode 25.


This week, since it's Christmas Eve, I decided to throw in another of my poster paper interviews from the Pacific Northwest Software Quality Conference I recorded in October. This TWiST-Plus episode covers the idea of adding White-Box techniques to automated testing and was presented by Sushil Karwa and Sasmita Panda, both being test engineers with McAfee and working out of Bangalore, India. Since doing the interview, Sushil has accepted another job in London, UK, so my congratulations to him on his new endeavor. For the episode, you can listen to it here.


Standard disclaimer:


Each TWiST podcast is free for 30 days, but you have to be a basic member to access it. After 30 days, you have to have a Pro Membership to access it, so either head on over quickly (depending on when you see this) or consider upgrading to a Pro membership so that you can get to the podcasts and the entire library whenever you want to :). In addition, Pro membership allows you to access and download to the entire archive of Software Test and Quality Assurance Magazine, and its issues under its former name, Software Test and Performance.


TWiST-Plus is all extra material, and as such is not hosted behind STP’s site model. There is no limitation to accessing TWiST-Plus material, just click the link to download and listen.


Again, my thanks to STP for hosting the podcasts and storing the archive. We hope you enjoy listening to them as much as we enjoy making them :).

No comments: