It had already been a frustrating morning. Too little sleep, and too little coffee. Squabbling kids. Traffic. Parents who didn’t know how to use the carpool lane at school and just stopped in the middle of the road, holding up traffic for 5 minutes while Junior fetched his lunchbox from the depths of the giant SUV. By the time I got to work, I had my cranky pants on.
And so when I read through a series of messages in some online forum or another in which some poor newbie was asking questions about Agile that indicated that they Just Didn’t Get It, I had to bite my tongue. I was not going to take out my frustrations with my morning commute on some poor unsuspecting person asking a perfectly legitimate, if naive, question.
But I couldn’t just ignore it. The wrongness of the assumptions in the message nagged at me, distracted me, kept me from focusing on the stack of work I intended to do.
So I wrote the following on Twitter:
If the org has 1-1 map btwn old vocab & new (e.g. WBS==sprint backlog, PM==Scrum Mastr, status meeting==Daily Scrum) they’re doin’ it wrong.
Turns out that was just the beginning. It seems there’s no end of ideas about how NOT to do Agile. I soon found myself writing about misinterpretations of TDD:
If the org thinks TDD means the testers write test documents before coding starts, they’re doin’ it wrong.
If the devs think TDD means writing *all* the unit tests in advance, and only then writing the production code, they’re doin’ it wrong.
@qualityfrog picked up on that with his contribution:
If those without title ‘developer’ aren’t included as team members, they’re doing it wrong.
Then @woodybrood suggested I should tag them. Thus, the #notagile tag was born. My most recent #notagile entries:
If an Agile team’s typical cycle time for critically important PO-requested fixes is measured in months, they’re doin’ it wrong. #notagile
If the Agile process has a reqs sprint and a design sprint and a code sprint and a test sprint, they’re doin’ it wrong. #notagile
I’m finding that having a place to put How Not to do Agile ideas is helping me. It’s cathartic. I’m not finding myself distracted the urge to write long diatribes on my blog anymore. And I’m pretty sure it’s lowering my blood pressure.
Follow along or feel free to contribute your own to the Twitter stream (tag your tweet with #notagile).
Elizabeth,
I’ve never worked in an agile group, but I like to read about it because I have found that I can apply some of the ideas to my current job. This post helps me in a different kind of way, though. I wanted to share.
I have found it so difficult to explain to people what “good testing” is, because it is such an abstract thing. When I start going on my rant about “information service” and “customer orientation” I start to lose people. Perhaps it would help me to list some of the things that I think “good testing” isn’t. I could list what “good management”, “good requirements”, “readable defects”, etc. aren’t. In fact, this sounds like a fun group exercise. I’ll keep it in mind! Thanks for the inspiration!
Joe
If agile means all testers on the project “essentially” write code (unit tests) they’re doin’ it wrong. #notagile Right?
This is somewhat an offshoot of what QualityFrog (Ben Simo) mentioned in his #notagile contribution.
Here is another one
If every occurrence of word “test” is associated with word “code” (in thinking and practice) they’re doin’ it wrong. #notagile.
Shrini
@testobsessed — if not keeping tests versioned with prod code is #notagile, what do you do with FitNesse tests, which aren’t in subversion?
Elisabeth responds:
Uncle Bob created a video around that very topic: http://vimeo.com/2765514
Are you collecting all the #notagile submissions outside of twitter? I think they would make the kernel of some good agile/scrum smells.