Sunday, September 5, 2010

A Fortnight to JavaOne 2010

With two weeks to JavaOne 2010, the frequency of blog posts related to JavaOne is rising as excitement for the coming conference builds.  In this post, I reference and summarize several of these.  I also begin what I hope is a once-per day (until JavaOne begins) focus on a JavaOne 2010 abstract that seems particularly interesting to me. For today, that will be at the end of this post.

Several people with long experience with Java development and with JavaOne conferences have recognized this year's JavaOne's focus on core and fundamental Java technologies.  Back in February, Oracle Technology Network (OTN) editor Justin Kestelyn called this year's edition a "return to purity."  In that post (which highlighted the call for papers), Justin stated, "the conference is going back to its roots: 100% focus on Java technology and its related ecosystem and community - no distractions. ... Sometimes, simpler is better."  Those selecting abstracts for presentation seemed to adhere to that statement.

Approximately a week after Kestyln's "A Return to Purity" post, editor Kevin Farnham observed:
In a sense, I think this year's JavaOne will actually be more focused on core Java topics. Wasn't the old JavaOne really a kind of "SunOne" or "SunWorld" conference, in ways? That's how it felt to me, anyway. Just about anything and everything "Sun" was present at JavaOne, in one form or another.
With the presentations selected, the idea of JavaOne focusing less on a particular company (Sun or Oracle) and focusing more on the Java ecosystem seems to have been realized.  I have posted my opinion of this and Adam Bien has also observed this.  In his post JavaOne Sessions Schedule Completed - First Impression, Some Criticism, he states:
Last week I managed to schedule sessions I would like to attend. The topics are interesting and technical - really looking forward to it. Product pitches are not existent - at least I didn't found any. From the topic / session perspective - I really looking forward to this conference. From the content perspective it should be at least as good as the previous ones.

Some had previously feared that the combining of Oracle OpenWorld with JavaOne would take away from JavaOne's Java developer focus.  Had these conferences been combined into the same conference, these fears might have been realized.  However, by having them co-located but separate, Oracle has managed to keep its more proprietary and Oracle-focused presentations in Oracle OpenWorld and Oracle Develop and leave JavaOne as a purely Java-focused conference.  Although my blogger pass allows me to attend all three conferences, there is no doubt that the vast majority of my time will be spent attending JavaOne presentations.

Although I'm pleased that the topics presented at JavaOne this year tend to be core and fundamental, I am also very much looking forward to hearing the keynotes at this conference.  In the case of keynotes, I often am looking for future trends and directions and the Java Strategy and Directions keynote holds a lot of promise in this area.  Terrence Barr appears to feel the same way and states (with his emphasis), "At the top of the list is certainly the Monday keynote Java Strategy and Directions (Monday, 5:45 pm to 7:15 pm)." I also agree with Barr that "another must-attend is the Tuesday keynote JavaOne Technical General Session (Tuesday, 2:30 pm to 4:00 pm) with Greg Bollella, Roberto Chinnici, and Mark Reinhold."

The most recent JavaOne 2010 posts are not without criticism.  Douglas Bullard wonders why he has not been notified of the cancellation of GWT presentations associated with Google employees.  I think that's a good question.  Have the individual speakers not notified the conference organizers that they will not be presenting?  Or, have the conference organizers been notified and failed to update Schedule Builder and send out the cancellations?  If the only "notification" has been Josh Bloch's blog post, then I'd argue that a single blog post (even one as viral as this) is insufficient basis to cancel several individuals' separate presentations.  For every conference that I've presented at, it has been the speaker's responsibility to contact the conference organizers to state that he or she won't be presenting.  That may have happened here, but I don't want to jump to any conclusions without knowing more details.  It could be the conference organizer's fault for not notifying conference attendees with these presentations on their schedule of their cancellation, but it could also be the fault of the presenters not communicating to the organizers that they won't be presenting.

Adam Bien is not impressed with Schedule Builder and writes that "schedule builder software isn't a usability show case."  He lists several ways it could be better.

JavaOne 2010 begins two weeks from today in San Francisco, California.

JavaOne 2010 Highlighted Abstract of the Day (Day #1)
Java Unleashed: Java Virtual Machine Tuning from the Pros
One JavaOne 2010 presentation that looks particularly interesting to me and is evidence of the "return to purity" and greater "focus on core Java topics" is "Java Unleashed: Java Virtual Machine Tuning from the Pros" (S313949).  This session will be presented by an IBM employee, Trent Gray-Donald, and will be held Thursday, September 23, 12:30 pm at Parc 55 in Cyril Magnin III.  Here is the abstract for this presentation:
Worried you're not getting every last ounce of performance out of your application running on a Java Virtual Machine (JVM)? Not sure where to start? Come hear some of the leading JVM implementers describe how to tune an application until it flies. This session will provide insights into different approaches to performance tuning, from low-level profiling through application-centered approaches. We'll use several helpful free tools, including VisualVM, the IBM Visual Performance Analyzer, and the Eclipse Memory Analyzer, to get unique insights into running Java programs and unlock new levels of performance from your applications. You'll leave with an appreciation of what JVM teams use to analyze and solve many performance bottlenecks.
I like to use VisualVM, but I'm not as familiar with the IBM Visual Performance Analyzer or the Eclipse Memory Analyzer (see also Acquire Heap Dump from MAT) and look forward to learning more about them.  However, what I am most interested in are the general JVM performance tuning tactics that I can glean from this presentation.

No comments: