I just took a 2-day Certified Scrum Master class from Innovel taught by Mark Pushinsky. I guess that makes me a certified Scrum Master, although I have to say it was the least rigorous certification I've obtained to date - all you had to do was attend class and participate.

We have been dabbling in Scrum for my current project, through my decision to utilize a Scrum-like approach, and with my boss's support (he has been pushing for iterative development since before I got there, so he's happy to let me guinea pig it). Now that we've been playing with it, and because the training class was available in Pittsburgh (no travel costs!), it seemed like a good time to move beyond learning from the books and web sites, and pretending, and get some actual training on the topic. I have to say, having sat through a lot of crummy training, and some good training, and having taught some classes myself (I used to be an MCT back in the VB 6-SQL 7.0 days), that Mark is a very good instructor, and I found the training to be valuable. At no point did I find my eyelids sinking.

Seriously, though, it was good to have been through the training. I found some things that we aren't doing that we should be doing, and some things that we are doing wrong. The one key take-away: if you are going to try out Scrum, even if you don't bother getting training (and you probably can fake it well enough just by following along with a book or web site (reviews of some of the resources I've used will be coming soon), you really should pick a project, and do ALL of Scrum. Not just a few things to get you going. Not everything but the 2 things you don't like. All of it. And then evaluate from there. Mark gave this advice, and having not done that on the project, I can definitely see how what we are doing isn't going to give us the same value as if we had done it right.