CSTA 2019

# COMMENTS

So, I got back from CSTA2019 and promptly got a summer cold which laid me up for the past couple of days.

Phoenix was beautiful even with it being unbearably hot at times and it's a place I think I'd like to visit again when it's a bit cooler.

I said in an earlier post that for me, CSTA is typically more of a person conference and less of a session conference and that held true again. I missed a few sessions that I originally wanted to attend because I was involved in really good conversations in the exhibit hall and before I knew it, the session was over and done. I can't say that I regret sticking with the conversations.

So, for the sessions, I don't have much to say. I do wish it was easier to glean information from the session descriptions. Even sessions with lengthy descriptions left me with some question and given that sessions were an hour long I wish some had more clarity. Specifically, there were a good number of sessions on AI and ML but often was left wondering if the sessions were more about doing cool problems with some AI engine or api and real world data or if it was really about learning about AI and ML and how to teach them.

The last session I attended was "Nifty Assignments" and that was an interesting contrast to the SIGCSE session of the same name. Most notably at SIGCSE the assignments are generally for a college level CS1 or CS2 but these spanned K12. You can check out all of the assignments here. The one that I found particularly interesting was Dale Reed and Joe Hummel's lesson which, from a CS point of view is similar to a set of lessons I've been doing for a few years (post). I think it's worth looking at both assignments to contrast how you can have very different approaches and emphases with similar core material.

I did a session with my partner in crime, JonAlf on using GitHub as a web presence. I think this approach could be a good way to introduce students and teachers to Git and GitHub in a low friction manner. It's a good first step to using GitHub for program development. You can use the GitHub web page to update a web page time and again and each time leave a little comment about what you've done. It turns out this is just a commit message - just like "real" programming with git. I can't speak to how good the session was - you'll have to ask the attendees.

I'm thinking of proposing a full blown workshop for next year's CSTA conference that goes through both how I use GitHub and also models the process I use to bring my students from novice to comfortable user.

Moving from the sessions, this year CSTA added a poster session. Unfortunately, the posters went up shortly before the session and came down right after and since I couldn't get to the posters during the official session I couldn't really take them in beyond a passing glance. I'm hoping that next year the posters go up early and stay up for the duration.

Some notes on the vendors area - wow - it was a far cry from two years ago where there were just a handful in a hall. I think I heard that there were around 60. There were the usual suspects - Google and Microsoft, curriculum providers like CodeHS, BJC, and PLTW, lots of robotics stuff and organizations like NCWIT and the College Board. I was there with the GitHub contingent so I spent most of my time at the booth. The nice thing is that while I was there to talk to people about GitHub I was free to give the real deal - my experience - the good parts and the pain points. Also had plenty of time to talk CS and CS Education in general.

I'd also like to mention Devorah's Blanket Statement project. Crowdsourcing knitting a blanket for a child in need.

Somewhere between 20 and 40 attendees spent time with Devorah at the GitHub booth putting idle hands to good use. Devorah and I are greatful to GitHub for hosting us and for the CSTA conference organizers for allowing this added activity. We're hoping it can become a regular activity at the CSTA conference.

So, that's my conference summary. I'm already looking forward to next year in Arlington Virginia.