May 8th, 2008

YellowLeaves

Deep in the heart of Florida

I have YouTube on to keep me company, so my flat isn't so lonely while I eat my dinner. Thanks to poh I have a preselected bunch of tunes to listen to.

I've eaten dinner by myself every day since I've been here. This is not something I enjoy doing. At least I have people to talk to at lunch at the conference, since they feed us all there.

Tonight I am having a personal pan pizza from the Pizza Hut in the hotel and a salad from the deli.

*shadowdaddy calls to check in*

Well! That was nice. I will end my whinging now and get to work on my paper.

PS: I think I'm doing pretty well all things considered even though I really think that spikeylady should be here so she could enjoy the pool with the waterfall. (And I dreamed about her and ergotia and lilithmagna and some_fox last night, so I guess I've got home on my mind, even though I did used to dream about them when I lived in Seattle.)
Tiara

StarEAST day 4: morning session

Well, I've just finished up two talks, one "The ROI of Testing" (Shaun Bradshaw) and the second "Learning from th Past: Leveraging defect data" (Brian Robinson). The first one was emphasizing showing the financial benefits to be made from doing testing earlier in the process, rather than when Dev "throws it over the wall." He included several formulas that I think would be useful - defect injection rate, defect repair cost, test effectiveness - but unfortunately he went over it so fast I neither got the formula nor the usefulness. I guess I'll have to look at the slides and see if I can figure it out from there.

The second talk was dreamy, kind of a vision of what I would like to be doing for a living. It was a presentation of a big project this guy worked on, to analyze where the defects were coming in for a very large company's software product and then figure out, not just what could be done to reduce the defect introduction rate, but how to sell it to the people who needed to do it and the people who needed to pay for it. The bugs were broken down by criticality and origin area, and they came up with two easy ways to attack the most glaring bugs at the source: unit testing and code reviews. To be honest, the description of how he dealt with it politically was just as interesting to me as how he "fixed" the problem, because you can come up with all of the brilliant programs you want to and if you don't get buy in you're just blowing smoke out your **s. I want to figure out what kind of PhD this guy has and see if I can get one just like it.

Now it's time for lunch and I'm ravenous. Check in later after *gulp* my talk! (New theme: "She Blinded Me with Science.")