Skip to main content

Ed Burns JavaOne 2005 Day Two rest of the day summary

Posted by edburns on June 29, 2005 at 1:20 AM PDT

You can see my trend line for blog entries is going down. No
excuses. Here is a summary post of the rest of day two at JavaOne 2005.
This was the most interesting day for me, with many Faces talks and
meetings planned.

The morning was packed. After the Amazon talk, I went straight to
the Wicket talk, where I ran into Howard. Here are the quick notes from
Wicket.

    Miko talks about meeting Jon Locke.

    Wicket, why its cool.

    Talked about how many web frameworks there are.

    What to look for in a framework. Emphasizes fun, POJO centric,
    reuse, non-intrusive, safe, scalable, complete.

    Sitting at this talk, I get the feeling there are two camps of web
    frameworks: 1: non-big-corporate, open source, but benign despot
    technologies, 2: big-corporate non-open-source, standard based community
    developed ones. Even though JSF is very open, I feel people still see
    it as closed. Not true!

    Demos: I18N, sortable tables, hello world.

    Show the RAW HTML. Uses span with an id to hook up an HTML element
    to a component. There is a heursitic that discovers the java class
    behind a page.

    Need to declare and map the WicketServlet.

    Is there an EL? Is there something like a managed-bean framework?
    Pluggable rendering? Conversion? Portlet? Integration
    wit Struts? Database integration?

    Questions

    Portals? Not yet. State management always in session (no client
    option) Eelco says they're working with JetSpeed.

    Faces has external navigation. Wicket does not.

    Martijn makes a bold statement that Wicket will blow Faces out of the
    sky for AJAX support. We'll see Martijn! Competition is good!

    Debugging? No, but they have the IDE.

After Wicket, Howard and I went tried to grab a quick lunch but ended
up chatting for a while about the difference between conferences like
JavaOne and conferences like NoFluffJustStuff as analogous to the
differences between big-corporate and non-big-corporate frameworks.
Howard shared an analogy to the analogy regarding chaos theory. In the
Mandlebrot set, all the interesting stuff happens at the interface
between the really chaotic parts of the system, and the really stable
parts of the system. That's where the fractal dragon appears. I like
Howard, he's very agreeable, practical, and fun to hang out with.

After the unexpectedly long, and expectedly not so tasty lunch I was
torn between Greg Murray and Tor Norbye's AJAX and Faces talk, and the
FedEx Struts to Faces talk. Since I know Greg and Tor and actually
contributed to some of the code they were using, I decided to go to the
FedEx talk. I'm glad I did. The Yerba Buena theatre was packed. I
think it's bigger than the Esplanade room. At today's morning keynote,
John Gage mentioned some stats about the most attended session on
Monday. I wonder which one it will be today? Kevin Hinners from FedEx
laid out three strategies for migrating from Struts to Faces.

  1. Containment: Keep the Faces part of the app separate
    from the Struts part.

  2. Rewrite: obvious.

  3. Integration: using the Struts-Faces integration library.

The most important thing about this talk for me was to see Faces
adoption inside large enterprises like FedEx. After the FedEx talk, I
had some time to kill so I prepared for tomorrow's Web Framework
Smackdown. I collected a list of features that web frameworks have that
I eventualy want to compile into a tabular format for comparison with
the frameworks on the panel.

Next up was the Portal Panel Discussion. I thought it would be
important for me to be there, but it turned out to be a bit of a vendor
pitch. The bright side was that I got some ideas on how I don't
want the smackdown to be. This one was boring! I fell asleep. For the
smackdown, I want to keep it lively and fun. I hope we succeed.

To that end, we had the technical rehearsal for the panel discussion
tonight, followed by an intimate beer and pizza session in one of the
meeting rooms where the panelists and moderator got to know each other
and practice responses to questions. At TSSJS, Floyd told me that panel
discussions work better when the participants know eachother, and there
are limited surprises. Or, if there are surprises, they're not "total"
surprises. The funniest thing about this little meeting was how I had
to use Jason's marketing schwag carabiner as a bottle opener. Not an
easy task! David Geary said, "I guess that's why you were the spec
lead!". I felt some tension between the panelists, which is remarkable,
really. I mean, we're all trying to solve the same problem, but there
is indeed a real sense of rivalry. We'll see if it comes through for
the audience tomorrow.

Next up was the Web Tier State of the Union BOF. This was the start
of this years night of a thousand web-tier bofs. Not as well attended as
I would have liked, but Adam Winer suggested we should have named it the
AJAX web tier state of the union and then people would have shown
up. Thanks to everyone who attended and asked questions!

After that I stayed through Ben and Dion's AJAX BOF, which was indeed
well atended. Those two are fun speakrs, Ben especially. I learned
later that Ben is href="http://en.wikipedia.org/wiki/John_Kenneth_Galbraith">John Kenneth
Galbraith's cousin.

Once again, I'm bushed and up too late.

Technorati Tags:

Related Topics >>