RSS 2.0 New Entries Syndication Feed Atom 0.3 New Entries Syndication Feed

Show blog menu v



Use it



Sibling projects

RIFE powered

Valid XHTML 1.0 Transitional

Valid CSS!

Blogs : Latest entries

< Previous page 
JIA: Deploying Apache Axis in Mission-Critical Environments by Eugene Ciurana
[series of entries regarding my trip to JIA]

Simply put, this presentation was a font of knowledge about Axis with great delivery and good content. Eugene's presentation focused on web services in a mission-critical environment, specifically using Apache Axis. Not just why you should use Axis, but how to use it and how to secure it once you are using it.

Eugene kicked things off by running through why and when you should suggest web services to your company. Then he did a quick explanation of the 2 types of web services he considered most important, SOAP and REST, and why you should chose them, leaving out the aging but simpler XML-RPC. I don't see this omission as an issue because, as I understand it, XML-RPC is a bit of a parent to the later developed SOAP.

During the presentation, Eugene stated several times his assumptions about his audience. Assumptions about the configuration of what they are working with on the server side, and assumptions about their knowledge of Java and web services. By doing this he raises his presentation to a new level because he doesn't need to spend time explaining what he assumes his audience knows. This is a presentation technique that can be hit-or-miss but is indispensable when it hits. It can be a problem when you misread your audience and assume too much of their experience. Your presentation will flop because no one will understand or will understand very little.

Eugene also went through various starting points for building your web service infrastructure, and the caveats with each. He suggests starting with the XSD schema and generating the business objects with that, instead of starting with objects and generating the XSD or including the schema in the WSDL.

He then goes into an in-depth look at putting it all together. What it looks like, the tools to use and the pitfalls of the methods. Lots of examples and lots of code makes this section particularly interesting. He then took the time to go over, in several slides, the options for securing Axis and the best method to do so.

Eugene's verbal direction and the quality of his slides really made this presentation a must-see. After watching this presentation I realize Eugene is the kind of presenter everyone strives to be. I look forward to learning more from Eugene and using his considerable experiences in my own work.
posted by JR Boyens in JIA on Oct 20, 2005 7:51 PM : 0 comments [permalink]
JIA: The Server-side Architecture Behind OpenLaszlo Applications by Geert Bevin
[series of entries regarding my trip to JIA]

[full-disclosure: I work with Geert and therefore may be biased even if I attempt not to be]

OpenLaszlo is an RIA framework (currently Flash output only) that I've seen some incredible things done with. Pandora especially. As well as Geert's own Bla-Bla List. So naturally going in I was quite interested.

Geert started off the presentation with various demonstrations of advanced OpenLaszlo applications including an Amazon store example with nice drag and drop and a very nice UI. Pandora the front-end from the Music Genome Project which allows you to select a song or an artist and get other music that is similar to that selection. And, of course, Bla-Bla List, a simple but nicely done todo list application.

Geert was the only person in the entire conference that I saw that used video in his slides and this really made a big difference in how the explanation of the applications came across. While the videos played he walked us through what was happening at each step. It seemed that in a few places his videos went just slightly too fast for his explanation, but it never detracted from the presentation.

After this he went into the architecture of OpenLaszlo and how it changed the traditional MVC structure. This part garnered some questions from the crowd, but from my perspective (and a few others in the audience) it seemed like there were two types of people in the crowd. Those that understood MVC completely and those that had no idea what MVC was and were totally confused as to what these slides meant. I even had someone behind me ask me what the term URL (pronounced Earl) meant. I discussed this with Geert later and we both agreed that it was probably an audience issue rather than a presentation issue. However, Geert did clarify the term in his updated slides which are now available.

Earlier in the conference there was a session dealing with Ajax given by Justin Gehtland who showed an example application with a city/state lookup based on an entered zip code. Geert took this example application and reimplemented it in OpenLaszlo. So for those that attended the earlier presentation it was a nice clean example showing what you can do.

All-in-all I think it was an excellent intro presentation to the OpenLaszlo system. I would've liked to have seen more code examples, but there just isn't a whole lot you can get to in only an hour.
posted by JR Boyens in JIA on Oct 11, 2005 8:00 PM : 0 comments [permalink]
JIA: Comparing Java Web Frameworks: JSF, Struts, Spring, Tapestry and WebWork by Matt Raible
[series of entries regarding my trip to JIA]

As a minor developer on a framework, I was quite interested in the content of this session. To be able to see someone go through, code, and explain how frameworks differ would be a very useful thing for not only my development work but for the extension of RIFE.

Earlier on Matt took a poll on what frameworks are being used and what the audience knew. Which is quite a nice idea for presenters. It provides a way for you to gauge your audience and really tailor your presentation. FWIW it seemed everyone and their dog was using Struts.

Matt's presentation style is more of an informal style and it was fresh entertaining and nice. Although I might chalk it up to the early morning coffee. Later on in the presentation he had some code demonstrations that completely failed and merely wasted time, but live code demonstrations are never very easy. Preparation is always a good idea. I've seen other presenters prepare all of their code in steps beforehand, test it thoroughly and still have issues. Scripted and simulated “live” code demonstrations are probably the only safe method.

The presentation had a lot of content, but the slow demonstrations really brought the whole thing to a standstill. The earlier slides about the architecture and design of the framework were quite instructive and even a little frightening in the case of JSF, who's diagram consisted of no less than 20 individual parts. Compare that to Webwork or Spring which had at most 5 or 10.

In conclusion, the information in the first hour or so was interesting, but after that it just dragged and fell apart. Complete with mumbling and complaining about Apple and their Java implementation and fumbling for a power cord. So in short, presenters: Please, have a backup plan. This was quite possibly the longest three hours of the entire conference.
posted by JR Boyens in JIA on Oct 9, 2005 5:34 PM : 2 comments [permalink]
JIA: Practical JMS by Chris Pearson and Phil Miller

[series of entries regarding my trip to JIA]

JMS was, before this conference, something I had actually never heard of. I had a passing acquaintance with various messaging solutions. More like I'd just heard they existed. I wasn't even planing on attending this session, but I talked with Pearson the night before the session and he convinced me that it was worth my while to attended. He refused to try and sell me on JMS, but I drug out some information about it and was immediately interested.

Before I continue further I did want to mention that Chris double-teamed this presentation with Phil Miller. another Formicary employee and I didn't catch his name. I couldn't find it in my materials nor could I find it on the JIA site. If someone would comment with the name I would be happy to give credit where credit is due. For now I think we'll refer to him as The Enigmatic Dude, or Ted.

Phil Ted started off the presentation and gave an overview of what JMS was, a few ideas of what it might be used for and just really some general information. Once the crowd was well informed about the state of JMS and a snippet of its background. Chris came to the front to take presentation over.

The presentation was oriented as sort of a Consumer Reports for JMS. Various JMS implementations were discussed and reviewed. No real “winner” was announced, but Chris went through the caveats of the various implementations and it was clear that IBM's Websphere MQ was the loser of the reviewed products which included both open-source and proprietary products.

Pearson also provided a clear client/server code sample that ran flawlessly and showed a simple example of how easy it was to use.

I've really got an itch to use JMS now because of the presentation, the problem is I don't know what for! Formicary provides financial solutions for clients and that's about the only use I heard for JMS. I would've really like to have heard something on other uses, or what other companies were doing with the technology. This detracted a bit from the after-feeling of the presentation. I felt like I wanted to use JMS, but had no outlet in my current projects or even in future uses.

I thought this presentation really hit the mark though. I learned about JMS and got the urge to use it. Which is exactly what I expected to get out of the session.

Update: Thanks to Hani Suleiman and Ted Phil Miller himself "The Enigmatic Dude" is no longer a mystery.

posted by JR Boyens in JIA on Oct 9, 2005 5:22 PM : 2 comments [permalink]
JIA: Ajax: Learn How to Develop Next Generation Rich Web Applications by Justin Gehtland
[series of entries regarding my trip to JIA]

I really must commend Justin on an excellent presentation. This was one to attend. Maybe next year TSS can record the presentations and place them online to purchase and download. I would purchase and download this presentation even though I watched it live.

Justin really did an excellent job of taking the information and presenting it in a way that everyone near me in the audience could relate to. He used concrete examples. He showed us the browser and what everything he coded did. He had all his code in steps so he wasn't writing totally live, but still allowed for the feeling of the senior programming giving you some hints.

I loved that he really seemed excited about the topic that he was presenting. The energy and enthusiasm he exuded really permeated the room and energized everyone. A three-hour session can get very long if the presenter doesn't have the drive and the ability to keep the room going. Justin's session was the only one that I pulled out my laptop and was writing code in.

The examples that he pulled out were pretty cool as well. The first was a simple City/State lookup by zip code. Really just laying the groundwork. He presented some really interesting ideas. I never thought about sending just plain raw text down the pipe. I had always just assumed that XML was the way to go and then you'd parse/handle the text in your JS code. He even showed an example of sending JS code down the pipe and then eval()'ing it on the client-side. Which initially had me a little squeamish, but he made sure that our security fears were waved. Since most, (if not all), browsers run JS in a sandbox, so it can't hurt your system anyway. But he did then say that you could access all features if you added the JS to a jar and signed the jar. So I'm not sure how safe that is either, otherwise you just get guys signing their jars with self-generated certs to push malicious code. There must be more to it than what was covered.

All-in-all good tips, good code, good information, and good ideas for development. An A+ presentation.

Update: I have been informed that the presentation was not, in fact, presented by Dion Almaer but in fact by Justin Gehtland as Dion had a family event come up. While Justin did use Dion's slides, I wanted to update the post to give credit where credit is due. Thanks to all that commented to correct my mistake.
posted by JR Boyens in JIA on Oct 9, 2005 4:51 PM : 6 comments [permalink]

< Previous page 
Google web