Jazoon's master of ceremonies this year, Reiner Grau, kicked off the
conference by evoking the mental image of Stanley Kubrick's "A Space
Odyssey" and then introduced Corsin Decurtins, the chairman of the
Program Committee. Corsin explained how difficult it was to choose
70 talks from the 250 submissions contributed this year and how
fortunate he was to have the team of 30 industry professionals as
reviewers in order to maintain high standards.

Perennial Jazoon keynote speaker Danny Coward was introduced to give
the opening talk, but not before being asked about his PhD in Number Theory,
to which he quickly replied, "I'm just one of the screaming monkeys
at the beginning of the Kubrick film". Danny was also candidly asked
to comment on his opinion on how it feels to now be under Oracle
management, to which he cheekily answered, "It is nice to be working
for a profitable company - it has been a long time".
With that, he began his talk, "Java SE and Java FX: The Road Ahead",
which detailed features that have either been just released or can be expected
in the near future. He consciously didn't cover JEE but deferred to other
nice talks in the conference for that information. Danny pointed out that
on 24 May Java turned 15 years old and told us there are 77 million
Java developers, 900 million PC desktop installations, and 2.5 billion
mobile device deployments. He

showed a
video of James Gosling from
sometime around 1992 demoing the Star 7 Java-based hand-held device with
a touch-based inertia-effect quick scrolling feature recognizable today
by smart phone users. Of the new SE features, he talked about the jigsaw
modularity effort to improve the size, scalability and speed of java
startup. Starting from 50 entities with 177 dependencies, they are now
down to 27 entities with 95 dependencies and continuing to unravel
interdependencies that have crept in over the last 15 years. He talked
about new features to help support parallelism including the garbage
first collector and a new Fork/Join API. He described the DaVinci
project whose goal is to increase the speed and simplicity of support
for other languages on the JVM, of which there are now over 200. Then
he talked about new language features: additional annotations, small
syntactic improvements such as multiple catch, inferred types in
object instantiation strings in switch statements etc. On the topic
of JavaFX, he described the quick pace of 4 releases in 18 months
which included new UI components closer to the completeness of
Swing yet scalable from mobile to TV deployments, speed improvements
via lazy binding supporting scenegraphs up to 1000s of nodes and better
startup through concurrent jar loading and faster class loading. He
described the Prism project, experimenting with a new graphics stack
which can take advantage of native GPU acceleration and closed with
a compelling demo of an application for visualizing and interacting
with Winter Olympic games medal awards by country since 1924.
When introducing the speaker of the final keynote talk, Kevlin
Henney, about his book "97 Things Every Programmmer Should Know",
Reiner Grau asked him how it came to be that so many physicists
seem to end up in

IT. Kevlin jokingly said that Geneva is where
all the European physics jobs are but there are too many physicists.
Reiner also pointed out that Kevlin wrote a patterns book and asked
how it was different from the famous "Gang of Four" book. Without
missing a beat, Kevlin replied "Well, they made an awful lot of
money." Kevlin started his talk by defending the number 97. He
first pointed out it is a strong prime but then admitted that he
wanted a number close 100 but didn't want it to be 99 or 101 because
they tried too hard to not be 100. The book is a complition of
97 nuggets of advice from a large variety of different programmers
from almost every continent. He briefly described some of these
such as "do deliberate practice" which of course involves doing
a task to master the task rather than to complete the task; "learn
to estimate" which distinguished an estimate from a target, which
in turn differentiates from a commitment; and preferring domain
specific types to primitive types, where he employed the help of
a keyword frequency visualization tool to show the difference
between examples of these two extremes and how the domain-specific
example was more expressive.