No, I'm not walking back anything from my last post. I still think we have to stop using professional development as the primary means for preparing CS teachers but I wanted to take a minute to make it clear that while there's plenty of lousy PD out there, there's also plenty of good stuff.

For bad PD, it's frequently mandated and frequently neither useful or relevant or even if its potentially relevant, but the time it becomes useful you don't remember it.

For good, first we look to our fellow teachers. A couple of years ago at SIGCSE, they released the results of a survey of CS educators - I think it was limited to colleges but the results were both expected and interesting. It turned out that the majority of respondents considered the best resource to improve their teaching practice to be techniques learned from other teachers. I believe first it was in department and then in external meetings but I could be wrong. "The research," on the other hand, was way down the list.

PD delivered by colleages is the best and can happen at all levels. A teacher might not be a CS ace nor may they be very experienced but they can run a session on something they tried and how it worked (or didnt') and I'll take that any day over something from a "thought leader."

Of course, back to yesterday's post, you can't prepare a CS teacher that way - you can take a teacher and add trick, a technique, a practice, or a topic.

Great stuff for what it is when used appropriately.

Every time we would have those school wide required PDs, we'd all look at the offerings and seek out the good stuff being delivered by our fellow teachers.

There's another class of PD that can also be excellent although, again, it's not a replacement for full teacher preparation. That's when some group goes in and isn't trying to "train the teachers from scratch to proficient" but rather understands that they're delivering a limited experience. This might be to expose a new audience to CS on an introductory level or a non specialist on how they can deliver a starter set of CS experiences or it could be something deeper but it'll be focussed on growing the teachers tool set and knowledge base, not on delivering some set scripted experience.

This type of PD seems to frequently be delivered by researchers and can be excellent stuff. It can also be delivered by content providers.

Finally, content providers can and do deliver worthwhile PD as long as its taken in the right context. If you're a knowledgeable CS teacher you might love some provider's curriculum and their training can be the best way to go. Nothing inherently wrong here as long as the training is not the actual teacher prep.

One of the best places for all of these types of PDs are at conferences like the CSTA conference and that conference has historically had the best of all versions of PD.

So, there it is, I'm not down on PD but I recognize that there's a lot of garbage PD to go with a lot of good stuff. The problem and the danger is that it's too often used as a replacement for real teacher prep.