News: What were your impressions of JavaOne?
JavaOne 2006 was a successful, impressive show. The floor had lots of energy, and the interest in the technologies was easy to see and feel. The session content was really good, with a lot of focus on technology transfer and less marketing... provided you ignored Sun's obvious focus on NetBeans and Glassfish. Honestly, a focus on Glassfish is acceptable. Glassfish is the reference implementation for Java EE 5, probably the largest change in the enterprise APIs since their introduction. Sun almost has a duty to focus on the future of enterprise programming, and for better or for worse, Glassfish is the first step towards that future. In addition, most of the "Glassfish presentations" had a primary focus on Java EE rather than Glassfish itself, with the application server mostly serving for what it really is: an implementation that contains all of the technologies. The Netbeans focus was less defensible. While Netbeans has shown quite a bit of vitality in the last year with a very active development/release schedule and a much improved user experience, it's still an IDE that manages to stand apart from other IDEs mostly because it's a Sun product. It's good that it's open source and freely available. It's also very good that it has largely abandoned the worst qualities of its past, and the community involvement around Netbeans has managed to go from something Sun says is there ("We promise, people use it!") to something actually palpable and visible. However, it is still just another IDE, and thus the constant barrage of Netbeans marketing was tiresome, although occasionally informative. With all this, JavaOne still managed to accept the existence of an entire ecosystem of Java, including competitors to Sun's product line, with an enthusiastic endorsement of Java's future development (even if some disagreed about specific features of that future.) Sun managed to market its open source products and new open sensibilities very well, and - in Your Humble Editor's opinion - fairly graciously, with a commitment to visibility, community participation, and even competition. The new acceptance of JRE redistribution for Linux is great news, no matter what dissenters might say (and there are dissenters, of course, who look every gift horse in the mouth.) The questions about Sun's vitality haven't gone away, but one thing's for sure: Sun is responding to what people have said they wanted. Plus, Sun managed to offer its products while making a compelling case that, while they obviously endorsed their own product lines, the competition was good in a broad sense. In a discussion with one of the Netbeans team, the statement was made to the effect that the best thing about Netbeans being viable was that the competition between the "big editors" - Eclipse and Netbeans - was beneficial to both, by forcing them both to improve. That was the message that nearly every Sun representative had - one way or the other - which managed to dilute the force of what marketing there was, and made it acceptable. These opinions are, however, just that: opinions. What were your impressions of JavaOne? What sense did you get about Java?
- Posted by: Joseph Ottinger
- Posted on: May 20 2006 07:39 EDT
- Re: What were your impressions of JavaOne? by Sergey Smirnov on May 20 2006 16:17 EDT
- Impressions by Lars Nielsen on May 21 2006 02:18 EDT
- any news on JBI and SCA ? by Gerald Loeffler on May 21 2006 10:41 EDT
- Re: What were your impressions of JavaOne? by Cameron Purdy on May 22 2006 10:06 EDT
I used to expect that Java EE 5 was announced explicitly during the JavaOne 2006. I asked Sun's guys could we say now what GA of JavaEE 5 is out. The answer was "Yes. It happened a week ago". This event should be explicitly announced. Otherwise, people, who are waiting for the new features, especially, the interoperability between JSF and JSP, still in the same waiting mode. In contrast, Google announced GWT and it really made a rumor at tow last days on the JavaOne. I agree, the JavaOne becomes more and more technical than marketing event. It is good and bad. This is good because people expect it is a technical exhibition, but not a trade show. It is bad, because less number of companies who present on the floor. It was a lot of spare place this year. What I met new is a lot of solutions for mobile devices. Motorola, Nokia , SavaJe Technologies, Sprint and others widely presented this area of the Java Technology. -- Sergey : https://ajax4jsf.dev.java.net/
I stopped going to JavaOne after the really bad show of 2003 when the conference seemed to have completely lost the plot and people were calling it SunOne. Looking around people do seem to think it has recovered and gone back to a technology focus. Also the inclusion of non Sun technical input is to be widely appreciated as generally the biggest changes in the J2EE space has come from non Sun sources in the last few years. In the end I am yet to be convinced that the $2500 tag is worth it (and I live in San Francisco). Especially when you compare it to other events like he Sysposium (that was not so great this year) or No Fluff Just Stuff.
I stopped going to JavaOne after the really bad show of 2003 when the conference seemed to have completely lost the plot and people were calling it SunOne.I've heard people say that TSSS was down somewhat this year... It's the first one I've missed, so I hope you're right :-) JavaOne this year was definitely better than it has been. I've only been to 2003, 2005, and now 2006, but this was definitely the best one I've been to. The technical sessions were good this year, and I had a hard time deciding which I wanted to schedule. The floor was typically unimpressive, but people seemed excited in general.
Looking around people do seem to think it has recovered and gone back to a technology focus. Also the inclusion of non Sun technical input is to be widely appreciated as generally the biggest changes in the J2EE space has come from non Sun sources in the last few years.
In the end I am yet to be convinced that the $2500 tag is worth it (and I live in San Francisco). Especially when you compare it to other events like he Sysposium (that was not so great this year) or No Fluff Just Stuff.
the ajax, most of the j2se and generice j2ee stuff (ejb3, jmx, come to mind) were decent. as was most of the JME stuff. The only things that stood out that sucked were the SCA/SDO demonstration/talks (too theoretical) as was the RIM talk on J2ME. anyone catch the RIM one? it was hilarious, i felt like i was back in college watching someone give a speech in one of my speech comm classes.
The lines were long and it took quite a lot of time to get from one session to another. The food was average. The JavaOne party was a joke - the beer lines were even longer than the normal ones. Good idea to have this XBOX 360 gaming area. Some of the sessions were very good, like the security sessions, Gigaspaces had their please-remember-us session that was good and finally, some sessions did not have javaone quality. Consider online session evaluations and lets-meet-X-nationality-parties. Credits to the Geronimo party. Me.
a question to those fortunate enough to attend J1: last year JBI was very much in the spotlight but i seem to sense a much more sober attitude to JBI in this year's conference coverage. Yes, Open ESB was announced, but that was about it, it seems. Could you shed some light on where JBI seems to be heading? E.g., will there be efforts to explicitly add distribution to JBI? Also, SCA and SDO have been mentioned time and again for many years now, but it's very hard for me to judge when they will be fact rather then vision. Tuscany is very important (including its Data Access Service for SDO) but seems to have a long way to go. Did anybody actually show a working SCA/SDO implementation? Are there efforts to standardise SCA in the JCP? many thanks, gerald http://www.gerald-loeffler.net
Did anybody actually show a working SCA/SDO implementation?IBM ships a product called WebSphere Process Server (WPS) that supports an early implementation of the SCA spec. This is a version of SCA that predates what is currently being specified jointly by IBM, BEA, Oracle, SAP and others, but nonetheless demonstrates the basic idea of service modularity and wiring assembly. All BPEL, Business State Machine (BSM), Rules and POJO services created in WPS are created and deployed as SCA components. I presented a snapshot of the WebSphere Integration Developer (WID) tool showing visual assembly during my session on the SOA Programming Model (TS-3608) -- this is just one way in which SCA may be rendered by different tool vendors, but is an example of how SCA development can be. Support for SDO has been shipping in products from IBM (WebSphere Application Server), BEA, and XCalia, that I know of, for a while.
a question to those fortunate enough to attend J1:Well, I don't even know what SCA and SDO are but there was a nice NetBeans plugin for visually developing BPEL - which was a kind of hand-in-glove capability in conjunction to OpenESB. It was enough to whet my appetite to pay more attention to this stuff and perhaps start to mess around with it to see what is possible. Some ease of entry tools will make that more palatable.
last year JBI was very much in the spotlight but i seem to sense a much more sober attitude to JBI in this year's conference coverage. Yes, Open ESB was announced, but that was about it, it seems. Could you shed some light on where JBI seems to be heading?
Did anybody actually show a working SCA/SDO implementation?Yes, SCA and SDO were shown working in the Tuscany BOF session on Thursday evening. You can try it out for yourself by downloading the Tuscany runtime and samples from http://incubator.apache.org/tuscany/ Tuscany does have a way to go, but it is making rapid progress now. Please feel free to join in the efforts to create a first class runtime for applications based on an SOA.
Are there efforts to standardise SCA in the JCP?Service Component Architecture deliberately aims to deal with service components written using many different technologies, one of which is Java. As a result, the main SCA spec is currently being evolved by an industry collaboration with the aim of submitting it to one of the open standards bodies, since it would be strange to deal with standardization of material relating to BPEL or C++ (say) in the JCP. Yours, Mike.
I wrote up my impressions here: http://jroller.com/page/cpurdy?entry=recovering_from_javaone A couple quick points: * Sun did a much improved job organizing this show. Yes, as with any big event there were lots of little snags, but kudos to Sun for a job well done. * For a few years now, I've been encouraging Sun to open up the floor to startups and open source projects, to try to "democratize" the pavilion a bit. This year, there was a "startup row", which is a step in the right direction. * There were a lot more people at the show this year, and the energy level was totally pumped. Peace, Cameron Purdy Tangosol Coherence: Clustered Shared Memory for Java