I attended the Chicago Erlang 2015 event on Saturday Oct 10. I missed it last year, so I made sure to look for it this year. Well worthwhile I thought. Instead of a regular conference, it was billed as an intensive one day workshop, with two tracks:
1. Essentials: An introduction for programmers, who are new to Erlang and,
2. IoT App Buildout: A plunge into the deep end of Erlang by studying and modifying a backend Internet of Things app written in Erlang.
I thought it was very well attended. I estimated about 50 in the IoT Buildout class and maybe 30 in the essentials class. (Anyone who knows the actual numbers feel free to correct me in the comments).
Below are a few of my observations, with the caveat that eye-witness testimony is probably the most unreliable and error prone form of evidence. Another attendee will certainly have a different point of view, and will have noticed things that I missed.
A big draw at least for me, was the top notch team assembled. The IoT class instructors were Brian Troutwine and Garrett Smith. Fred Herbert author of Learn You Some Erlang for Great Good, was on tap for the Essentials class. A fourth instructor Martin Logan was unable to make it. Although not officially on the program, even Tristan Sloughter of rebar3 fame was wandering about. A little Bingling an you will find out these guys are very active in the Erlang community.
I chose to attend the IoT Buildout class because hey, I know what a gen_server is. The class was built around the Fahrradboten application (or Bicycle Messenger application for us Deutsche-challenged). Brian handled the code on the screen while Garrett served up color commentary.
We went through the app in the morning, with Brian encouraging the class to look for ways to improve it. In the afternoon, we were divided up into maybe a half dozen groups, worked on various changes, and submitted pull requests. Brian “carefully” reviewed each PR, and rebuilt the app. Which promptly failed with a mysterious error. Brian said he would fix it and blog about it later. This was, in a way comforting, to know that even Erlang gods have to debug programs like us mortals.
Some other odds and ends in no particular order:
This was an overwhelmingly Apple/Mac crowd. I think I was maybe one of two people that showed up with a Windows machine. This led to a problem trying to build a release using rebar3. I’m sure it’s possible, but I think I would need to build it from source, on a PC. I didn’t realize this, because all of my Erlang development so far has been on the Raspberry Pi.
Garrett mentioned Lukas Larson a couple times. I have never heard of him before this, but the context in which he was mentioned (The be-all, end-all, and last word on certain Erlang related topics.), made me immediately think of the classic Bugs Bunny cartoon, Leopold.
“ETS tables allows Erlang to perform side effects.” (Forgot who said that)
“It’s not fault tolerant, if it doesn’t recover from a fault.” Garrett Smith
Mnesia was a cool database technology…. 20 years ago”, Fred Herbert?
I was surprised at the distance some attendees traveled. The guy next to me was from Kentucky, The three guys at the table nearest to mine, work for Splunk in San Francisco. I think I heard one guy mention he was from Mississippi.
Right after the conference I made sure to subscribe to the Erlang questions mailing list as that was mentioned several times as a good resource.
The after party, was held in the “VIP Room” of the Billy Goat Tavern, home of the cheezborger. And yes, in case you’re wondering, the VIP room was everything you would expect in place named the Billy Goat Tavern.
In conclusion: I had fun, learned a lot, met some great professionals, and I am looking forward to Chicago Erlang 2016.