I recently did a “tour” of the east coast of Australia for the YOW conference. This conference is pretty much like the “GOTO” conference seen in Europe - and this year toured Melbourne, Brisbane and Sydney.
I went to all three to speak, some by plane, some by car. I kept getting stuck in horrible traffic:
Made me late.
I was speaking on experiences adopting functional programming languages (Erlang, Scala and Clojure) - in the context of microservices (a subject I have written and spoken about before). The talk also included an intro by @jedws (Atlassian) and @kornys (Thoughtworks) - who both had examples/case studies from their organisations in a similar vein.
Our aim was to get people interested in trying out FP languages and techniques, use them “in anger” - in production, where they could, to see how it can make life better for them as developers (my angle was microservices: a small service that does one thing - if you have to replace it - no big deal! Also great for the easy-to-deploy paas model of apps).
YOW/GOTO is known for getting luminaries to speak and hang out with people - I thought I would drag down the tone by handing out free underpants.
There was a reason for this - I came up with the idea of “Lambda Underpants” as a way of advocating for FP. This was based on “Ruby Slippers” - hard to explain - read those links!
So - I actually went and had some “Lambda Underpants” made, professionally. This was quite awkward and I don’t ever want to do this again. This went down very well with the audience when I handed them out:
A satisfied customer.
I even promised to mail some underpants (someone was live tweeting) - which will be, hopefully, my final awkward experience this as I go to the post office and mail someone my underpants. If I never have to do that again it will be too soon.
Interesting things I learned
Jim Webber and Neo4J: I had avoided thus far learning about graph databases. No time. But this was fascinating. Graph DBs like Neo4J aren’t so much about storing docs/data, but storing nodes and their relationships - you can then query based on the type of patterns of relationships and connections you are looking for. Really, this is logic programming in a way, but with LOTS of data.
I came away fascinated and keen to find a way to have a Neo4J service on CloudBees (and find customers to champion it!). Read more.
Jeff Hawkins - computing like the mind: This was a keynote that talked about an open source project, and a product, that uses a “cortical learning algorithm” - this is a new model of neural computation similar to how our brain works - which allows “on-line learning” from arbitrary datastreams, and then reasoning over new stimulus (just like we learn - we are born - and from then on - are learning “online” - learning and new data are the same thing).
This was fascinating and showed that the state of the art has moved on in AI, as has our understanding of the brain, it was more than a little chilling, and amazing, thinking about a post-programming where machines learn online and think. He assured us no robot apocalypse was imminent, but I don’t believe him. Read here for more info.
Please, lets never speak of the underpants again.