-k.

My son is almost 16 and nearing the end of his last “mandatory” year of school. He doesn’t want to go back next year. He can’t see the point of school. It’s a difficult thing to deal with. I’m passionate about him continuing on though the rest of high school and into university. I was talking with him last night and I explained it this way.

When you were five, if I’d just let you eat what you wanted then you would have lived on lollies and ice cream. Now you’re a little older you understand that you can’t live on lollies and ice cream. You know that although they taste good, you just need to have other things in your diet (like coke and meat pies ;)).

We went on to talk about things he might want to do. He doesn’t want a desk job, fair enough, but he has it in his mind that all jobs out of university are boring desk jobs. “I don’t want to sit in front of a computer all day.” He a gregarious kid, outgoing, happy and a bit of a charmer. I’m trying to show him that there are careers that will allow him to be involved with people, get outside, that going to university doesn’t mean doing a desk job.

It’s an ongoing process.

An updated version of Necessitas is now available. Necessitas is an Eclipse plug-in that adds a class-path container to manage a projects jars using the ivy dependency manager. You get the advantages of ivy without leaving eclipse.

Key features for this release:

• Start-up time greatly reduced (especially for poor bastards like me on a modem
• Update to Ivy 1.2a
• Takes advantage of the new “cache only” operations in 1.2a
• Will attempt to assign source achieves if they are available
• You can check for updated dependencies without downloading them
• Ivy messages are available while updating
• Lots of internal cleanup

It can be downloaded from the usual location (http://eclipse.oneill.id.au/updates/).

Spring MVC, Web Work, Cocoon, Tapestry et al provide us with great frameworks to build our web applications on, but are these frameworks the way forward or are they the last remnants of the old republic?

In a recent post to the Cocoon mailing list Stefano Mazzocchi (the father of Cocoon) asks if changes in client side technology are making these advanced web frameworks less relevant.

… and the more I learn how to driven the client, the less I feel the need for advanced server frameworks. Is it just me?

Stefano is right, things are changing, the web is becoming a true application platform (just take a look at how many patents google has taken out in this area over the last 12 months) and the current crop of frameworks are going to either change to meet our needs or fade into obscurity. It’s not just the frameworks but also the infrastructure on which they are built. As Greg Wilkins points out in his entry on continuations in Jetty 6

The advent of AJAX as a web application model is significantly changing the traffic profile seen on the server side. Because AJAX servers cannot deliver asynchronous events to the client, the AJAX client must poll for events on the server. To avoid a busy polling loop, AJAX servers will often hold onto a poll request until either there is an event or a timeout occurs.

The Servlet specification with its focus on simple request/response interactions will need to change to support asynchronous event based modes of interaction. The current crop of servlet containers use a thread per active connection which has worked in our simple request/response world may not be practical in an AJAX environment. NIO and it’s event based networking will be the key but the current servlet specification was written well before NIO and has no real support for event based processing of requests. HTTP itself isn’t all that AJAX friendly but with the web browser being the likely application client for the foreseeable future and the slow moving w3c we are going to have to live with it.

There is space to make a mark. To take a mental shift. To be apart of the evolution. It’s an exciting period to have some free time to investigate emerging technologies.

The project I’ve put my heart and soul into for the last two years has been cancelled.

I’d pretty much had enough of the project a few months back (see Time to move on?) but my manager, someone I respect, convinced me that things would change, well they did. The business is a fit of senseless stupidity and corporate political maneuvering killed the project. Not officially of course, that wont happen until July (next financial year)
. The official line is that it’s being “re-scoped”. All the contractors and many of the senior staff (one person in the entire team is remaining with the company) are being released to pursue other opportunities. I left on Friday.

The hardest thing about leaving the project is leaving the team. If you work in a corporate environment you often end up working with a large number of “also rans”. Our team wasn’t like that. It was filled with smart, energetic, committed people. We took everything that was thrown at us and turned it into a success (at least from a technical perspective). When problems arose we worked through to find the correct rather than the easiest solution. It paid off countless times and although discussions were at times heated everyone respected everyone else.

The final debriefs were hard. I sat in one meeting that involved the development manager from a downstream system. I have little time for fools and this team (with one or two exceptions) is full of them and the development manager is one of the nastiest pieces of work I’ve ever had the misfortune to work with, she a true obstructionist. Unfortunately she leads development for the public facing on-line sites so is able to use fear as a lever on the business. Anyway at one point our project manager was talking about the re-use of some of our work, and praised the development team for the quality of our output. She smirked, humphed and asked if there was any hardware she could steal from the project. When she was told no, she got up and left. I nearly lost it.

I’m proud of the work the team did, it’s a shame the system never made it to production.

• We built content storage system that was tested with 24.5 million pieces of content ranging from single lines of text to multi-megabyte image files
• We developed a method for streaming content to and from the repository that provided a predictable level of performance and memory usage
• We provided a sophisticated validation system that supported complex business rules with real-time (in editor) performance
• We created a event driven content distribution system
• We provided notification hooks for work-flow and other systems interested in repository changes and events

I’m now looking for my next challenge. I’ve spoken with old colleagues who have exciting things on the boil. The future is bright.

As a final ironic twist one of the external vendors (who was effectively part of the political power play that bought us down) has spoken to one or two people from the project about integrating our ideas into their product.