This is my first time at the perennially popular Jfokus conference here in Stockholm, Sweden, and so far, it has been an awesome trip. Stockholm itself is very nice, and although there is currently quite a lot of snow, everything seems to be working – even the train arrival time is accurate to the second. Coming from Britain, this is an unexpected surprise! The Jfokus venus is also superb, both in location (right next to the train station) and facilities. After battling the snowy streets I received a warm welcome at the venue, and was given all my very nice conference swag – I now (finally) own a Jfokus beanie, woo-hoo!
The first conference session day began with some rather funky engineer-style break dancing (with a professional crew, not simply a bunch of developers!) and a welcome to the conference from conference organiser Mattias Karlsson. Next up was a rather interesting satirical look at future markets within IT, exploring B2C, B2C and FOC2BFOW (friends-of-consumer-to-buyer’s-friend-of-wife…). This was an unscheduled talk, and unfortunately I didn’t catch the presenter’s name, but in addition to being a great comedy start to the day it did make several good points about the current pace of innovation and market creation within the IT industry, as well as the legacy of bureaucracy we still suffer with at larger enterprises.
Christian Heilmann presented the main keynote “You don’t need another app – you need one that survives the Thunderdome”. This talk provided some great insights into the mobile app market, and destroyed the popular myths that the most popular apps are games, and that many app developers get rich (less than 10% of popular apps earn enough to pay a salary). Christian also suggested that mobile apps are now considered passé by many, and accordingly mobile device apps should simply be treated as another form factor that consumes data from back-end services.
After the morning coffee break I attended fellow LJC-er and Devoxxian James McGivern’s very interesting “Probably, Definitely, Maybe” talk, which discussed the use of probability theory with software. Key takeaways included the use of Bayes’ theorem to adjust product rankings, a guided tour of probability distributions, and a brief discussion suggesting that if your leg falls off then you probably aren’t suffering from Ebola! There were also great sections on stochastic processes and Markov machines. I bump up against these every so often in my work, and I’m always fascinated when I do.
After a very nice lunch of chicken salad and pasta it was my turn on the stage, this time presenting my “Thinking Fast and Slow in Software Development talk”. I was fortunate enough to fill the room with 500+ people, which hopefully suggests that developers are becoming even more aware of the need for good decision making within software development (or are at least keen to learn more about it). The audience was great, and I had some superb questions, comments and feedback, which I will try to factor in to the next version of this talk.
Straight after I had finished chatting with attendees from my presentation I headed downstairs to learn about akka-http from Mathias Doenitz. This was another very interesting talk, and I am keen to explore the framework further. In essence, akka-http wraps the popular Akka actor library with its own client and server-side (async and non-blocking) HTTP stack. If you’ve been reading about reactive streams of late, then you’ll probably have heard of this concept (or the toolkit) being mentioned before.
Next was the afternoon coffee break, and I took the opportunity to visit the sponsor booths in the main hall. There are some impressive sponsors, and I had some interesting chats with the Red Hat crew and the IBM Bluemix team, both of which are pushing into the PaaS market. Also running alongside the coffee break was a series of ‘quickie’ sessions, and I snuck along to Heather Vancura’s “JCP, Adopt-a-JSR and you” to show my support for this great work. If you are interested in learning more about how the Java language and the JVM platform are evolved within the Java Community Process (JCP) then please do pop along to the Adopt-a-JSR website or get in contact with Heather. [Editor’s note: also check out this great overview by Mani Sarkar]
Wrapping up the day’s sessions for me was a session of “Advanced Java Unit Testing with Spock” by Ken Sipe. I’ve seen an earlier version of this talk before, and it doesn’t disappoint. Spock is a very interesting Groovy-based testing framework, which is well worth a look if you are exploring ways to grow or improve your testing processes (or, for example, a looking for a way to introduce Groovy or BDD to your organisation…). Key takeaways for me included that as interesting as Spock is, Groovy itself is also very cool, and the use of a dynamic language for testing can reduce a lot of the pain and boilerplate code that is typically required when testing a Java codebase.
The rest of the afternoon I enjoyed participating in the hallway track, and managed to catch up with some regular conference friends, and meet some new ones too. The evening concluded with a buffet (courtesy of King Games) and a few beers (kudos to Red Hat!), which we enjoyed while watching Stephen Chin’s live Nighthacking session with a panel of Java luminaries on stage discussing the “forgotten Java bytes”, which was essentially a walk down memory lane looking at earlier Java books and conference swag.
I had several interesting chats throughout the evening with the Mattias Karlsson (thanks for the invite!), Heather and the JCP crowd, and fellow LJCers Richard Warburton, Raoul-Gabriel Urma, James McGivern and Jim Gough and fellow Devoxiann Arun Gupta. I also had several great conversations with attendees from my talk earlier in the day, and I hopefully shared a bit more of my wisdom, in addition to learning a lot from the great feedback.
That’s it for the first day at Jfokus 2015. Both Stockholm and the conference venue itself have been great so far, and I’m looking forward to another fun-filled (and informative) day tomorrow.