Read on Twitter
"In 1999, at the startup I was at, it was an amazing realization to see how my leadership would pay almost anything if we could ship one day sooner. So this is a completely different dynamic. I'm no longer optimizing for cost, I'm now optimizing for time." -- @testobsessed
(This is reminiscent of how @chawklady described how DevOps journey at Nordstrom began: "at board of directors level, we realized to avoid the fate of Blockbusters, Borders, and Barnes & Noble, we had to stop optimizing for cost, and instead, optimize for speed.")
"Our leadership, back then in 1999, seemed almost willing to pay anything to get to market one day earlier." — @testobsessed. (Fascinating she notes long time fellow traveler @samguckenheimer is sitting in the front row, recounting a LAWST roundtable — what a great scenius!!)
"I remember STEMR-3 workshop started by Dr. Cem Kaner that explored the question: what should be the ideal ratio between developers and testers. Around that time, this was an important question. MSFT recently stated that ideal ratio was 1:1, which was a hot topic." — @testobsessed
"Why a hot topic? Because back then, every technology leader needed to know the ratio of dev to test should be, because that would drive the staffing plan. So in this workshop, facilitator asked for the ratio of your best project, and for your very worst project." — @testobsessed
"For everyone's best projects, there were no discernable patterns among the workshop participants.
"But for everyone's worst projects, they were all very high — they had the most testers." (laughter)
"This is very prone to misinterpretation..." — @testobsessed
"But for everyone's worst projects, they were all very high — they had the most testers." (laughter)
"This is very prone to misinterpretation..." — @testobsessed
"Fast forward to another startup. In what became one of the worst days in my career, I started to realize that the better job I did as the head of quality, the worse it was for the organization.
"Here's the system of effects diagram..." — @testobsessed
"Here's the system of effects diagram..." — @testobsessed
(It's fascinating to me how so many people in the DevOps community have made these system dynamics diagrams: we used them to help write The Phoenix Project, @damonedwards often used them, etc. To me, it shows to what degree DevOps community are systems thinkers.)
PS: I'm going through and taking notes from my Episode #4 of @testobsessed's fantastic DevOps Enterprise Summit 2015 and 2014 talks. https://itrevolution.com/the-idealcast-episode-4/
This is the paper that @testobsessed mentioned:
Managing Proportions of Testing to (Other) Developers by Dr. Cem Kaner, Elisabeth Hendrickson, and Jennifer Smith-Brock
http://www.testingeducation.org/BBST/foundations/Kaner_pnsqc_ratio_of_testers.pdf
Managing Proportions of Testing to (Other) Developers by Dr. Cem Kaner, Elisabeth Hendrickson, and Jennifer Smith-Brock
http://www.testingeducation.org/BBST/foundations/Kaner_pnsqc_ratio_of_testers.pdf
"...we had a team that had long-lived feature branches. They'd work for 2 months without merging into trunk. It would feel great, feeling productive, ignoring everyone else.
"...then it's time to merge.
"Four weeks to merge everything together again." @testobsessed
"...then it's time to merge.
"Four weeks to merge everything together again." @testobsessed
