WDSC Webcast Today

COMMON is hosting a webcast today called “WDSC: What you Didn’t Know it Could Do”. Originally George Farr was supposed to do the webcast but something urgent came up and he will be on an airplane during the webcast, so I’m the backup :)

I can’t say as I’ve ever delivered a WDSC webcast with that exact title before, and the description implies that there will be something here for everyone; both experienced RSE users and those who have never seen it before. So it should be interesting :) I’ve decided to pull together some slides that show some of the useful, but perhaps not so well known, smaller point features from various parts of the RSE. Of course, I’ll also do a highlight of new stuff in 7.0

The webcast is free for COMMON members. So if you are interested and available from 1 – 2 EST today it would be great if you could join us.

Updated June 1st:  The webcast replay and slides can now be found here and the Q&A here.

WODA 2007

No, this is not a post about star wars, it’s about dynamic analysis :) This past Tuesday I attended the Fifth International Workshop on Dynamic Analysis in Minneapolis. Despite it’s close proximity to Rochester, this is not an IBM affiliated conference. It’s an academic conference where researchers from various Universities and research organizations present their research.

There are two main categories of analysis: static and dynamic. Static analysis is done without having to run the program. In WDSC, the outline view, program verifiers, application diagram, WebFacing conversion, and Java refactoring tools all use static analysis.

Dynamic analysis is done while the program is running. The profiling and logging tools in WDSC (which are inherited from the Eclipse TPTP project) are another example of dynamic analysis tools. These tools gather memory, execution time and method coverage information as your Java program runs. The information is then analyzed to assist you in isolating memory and performance problems in your programs.

I find that static analysis tools are more commonly used today in commercial software development. I believe one of the reasons for this is that they’re easier to integrate within the overall development environment since there is no need for the user to run their program and have some agent collect information as it runs.

Another difficulty with dynamic analysis is it slows down the execution of the program. This seemed to be a focus area at the workshop; each presenter reported (or was asked during question period) what performance overhead their research added to the running program.

However, research in dynamic analysis seems very promising. The keynote speaker, Rajiv Gupta, presented his research on detecting the location of software faults and strategies on how these faults could be avoided dynamically (i.e. on the fly). In regards to performance, he also touched on some research that uses multi-core processors, where one processor is running the program and another processor is doing the dynamic analysis.

The Value of Accessibility

For decades, accessibility for people with impairments has always been a corporate priority at IBM. Yet, for the most part, the software industry tends to treat accessibility as an afterthought, even though there are various government standards such as Section 508. As our user interfaces have progressed from terminals, to desktop clients, to web pages and continue through to immersive three dimensional environments, such as Second Life, providing for accessibility becomes more difficult.

The Evolution of Product Information

A few posts ago, I expounded at length about how people are partly unfamiliar with the usage of their devices. Until we discover a process that allows us to learn how to use things instantly, such as with a consensual mind meld, or through direct neural imprinting, we will continue to have this affliction. Even if we had one of these abilities, some of us would no doubt struggle to use it effectively.

Customers that use evolving products like WDSC, and perhaps your own creations, go through a constant learning process, which includes looking for answers to specific questions. The implication in Don Yantzi’s RSE Advanced blog post is that the power of any tool is best realized after users have gained sufficient knowledge and experience in using and combining the multiple abilities of the tool.

It is a difficult task to provide detailed information for any relatively complex product. At times it can seem like a hit and miss proposition, like casting a net only to discover that there are holes in it. To complicate matters, these questions are being asked by people with varying levels of experience, from novices who ask general introductory questions, to experienced users who ask specific, granular questions. So your net has to be both thick with large holes, and thin with small holes. Also, questions vary in scope. Some questions may involve a process that touches many features of your product, while others are confined to a small component. Is there a solution to this problem?

RSE Advanced

Claus Weiss and I did a presentation at COMMON titled RSE Advanced. This was a look at advanced features of the RSE (included in the standard edition of WDSC), NOT the features of RSE only included in WDSC AE. In the presentation we covered customizing the workbench, filters and filter pools, launch configurations, and working in a team environment. [Because I talked too much] we ran out of time and didn’t cover iSeries projects or working disconnected.

I wouldn’t consider most of those topics on their own to be overly advanced, with the exception of maybe filter pools. But that’s just because they are implemented in a weird way to make them more flexible. Filter pools are really just a group of filters. The tricky part is realizing that subsystems don’t “own” filter pools they just reference them. This allows subsystems from different connections to reference the same filter pool.

The main thing I wanted to focus on in the presentation was how to combine all these things to customize the RSE to adopt to the way the user works and therefore make them as efficient as possible. So we dived into options for customizing the workbench layout, views, preferences, keyboard shortcuts and LPEX, LPEX parsers, LPEX actions, user defined actions, customized compile commands, and multiple connections. We also discussed how to set all this up so one person can do the customizations and then share them with other team members.

I think I’ve mentioned in this blog before, that each of these features on their own is good, but it’s the power of combining them and using them together that makes them great. Understanding the possibilities and how to combine them is the advanced part. In case you are interested here is the RSE Advanced presentation.

The Continued Longevity of the Client-Server Model

A recent news article, where IBM’s chairman, president and CEO Sam Palmisano declares that the ‘PC client-server model has run its course‘, raises questions about the future of desktop applications. My first reaction would be to say that there are applications, such as WDSC, that are best suited for the desktop. But is this really true?

Two Minutes with George Farr on WDSC

I just came across this article on SystemiNetwork of a quick interview with Product Line Manager, George Farr. It’s short, but answers five questions:

1. So George, we heard about some major internal changes at IBM between Rational and WebSphere — what’s all that about?
2. Now that’s interesting . . . OK, then, what is all the fuss about the screen designer in WDSc?
3. So, are you still saying that the screen designer will be left in Advanced Edition?
4. For now then, what should customers use for designing 5250 screens and reports?
5. We’ve got a few seconds left . . . anything else you’d like to add?

See the answers here.

WDSC 7.0 Webcast Tomorrow

Tomorrow I am doing a System i Network webcast on WDSC 7.0. The webcast is sponsored by Aldon. During the first half hour I’ll present on what’s new in WDSC 7.0, focusing on the RSE and highlighting what’s new in the non-RSE stuff. During the second half Aldon will be presenting source change management and how their stuff integrates with WDSC.

You can register for the webcast here. As an added benefit it looks like System i Network will be having a draw for 3 TAO 1.4 digital picture keychains during the webcast. I wonder if the speaker gets included in the draw :)

Implications of Your Internet Footprint

Recently, I was made aware of a present-day practice by some employers to research potential employees by examining what they find on the Internet. Curiosity led me to investigate the nature of my own Internet footprint. I found the usual suspects, such as this blog and a few patents that I had authored. But, I also discovered a few things that I had forgotten about, including an article that I wrote for a print publication, some photographs, and bug reports that I had submitted. I soon found myself wondering, in the unlikelihood of a change in my job status, whether my Internet footprint accurately reflects what I want to convey to prospective employers, and their customers, or in the case of self employment, what I would prefer to convey to my own customers. How well does your Internet Footprint fit you?

Report from COMMON

Expo just closed up at COMMON so I now have a chance to write this post and do some work. Here are some things I have found:

• Lots of interest in the RSE. I just walked past George Farr’s RSE session and it’s packed. According to the COMMON Conference Daily that room can hold 200 people. Pretty much all the chairs are used and some people are sitting on the carpet along the wall.
• I’ve had a number of people ask if you can use OVRDBF with compiling from RSE. Yes you can. The trick is to turn off batch compiles from the Remote Systems > iSeries > Command Execution preference page. Then the OVRDBF command and compile commands will run in the RSE server job associated with the connection (which is a batch job so it still doesn’t take interactive cycles).
• Much to my surprise, nobody asked about Screen Designer, Application Diagram and WDSC AE in the sound off at opening session.
• I attended Scott Klement’s Web Services from RPG Using HTTP API. He does a great job of describing what is a Web service. Highly recommended as an intro to Web services (the session is repeated Thursday at 12:30).
• A lot of people are saying “We are at a crossroads and need to come up with a Web development strategy”. Meaning they feel they need to get serious about Web development with System i but they don’t know which way to go (EGL, PHP, Java EE, CGIDEV2).
• Nazmin found us a good Indian restaurant to eat at last night :) The Gandhi Palace inside the Ramada Plaza Hotel.

That’s it for now.