In February I took on the daunting task of starting a new venture. It was based on an idea I had while reading a book on the low cost airline, Ryanair. I never knew you could lease an aircraft; I thought an airline with billowing amounts of cash just bought the machines and got on with it. Wrong, wrong, and wrong with a capital W. My rationale was simple: there are a lot of aircraft on the ground, let's help get them back up in the air. How can I provide a system that makes sure that both parties benefit. B2B auctions!

Aerleasing is an enterprise auction engine for the airline industry. More to the point, it's 100% Java. The requirements were simple. Use as many open source libraries as possible so there is no major outlay. Borrow no money; use only what you have on hand.

The most expensive thing was hiring an excellent graphic designer. Brand is still everything, no matter how good the programming is. With no brand identity, you're dead in the water before you even start. So February was spent phoning, talking, and listening to as many people as I could get my hands on. I don't think I have ever learned so much in a short space of time.

I settled on Sitemesh for my template framework, mainly because I had used it before and could get up and running easily. I wanted users to be able to upload assets to their auctions (such as images and documents), so the Jakarta Commons file upload was an obvious choice and easy to implement.

The first prototype got a lot of reworking after a few of my contacts commented on the system. "Can you get a PDF copy?" they asked. No problem! I had a couple of choices, FOP or iText. I found iText was excellent in providing PDF documents from a servlet. Easy to set up and I put images in the document as well. If you don't mind crafting XSLT stylesheets, then FOP is wonderful too.

In the original version of Aerleasing I provided RSS feeds (using Rome to generate them), thinking I could convert an entire industry into using this wonderful data. Not so. I spent more time explaining it and it still caused confusion. What these folks live on are spreadsheets, so why try and fix something that isn't broken. Jakarta POI was downloaded and worked on; in fact it's a work in progress but I've used it a handful of times before so I know what's going on.

I spent a lot of time thinking about how auctions could be updated in terms of their start and stop times. After a bit of Googling around I found Quartz as one thing I did require was that the time worked outside of the app server just in case the server crashed for whatever reason. I didn't want the timer stopped because the server stopped.

The site launched in early July and was creating a bit of stir with some of the industry press. Nothing like the concept of Aerleasing has been seen before, but that still didn't stop me from having to pick the phone up and cold call some companies. As a technical architect it was a bit nerving but now I actually enjoy it. They don't ask me about SOA, RSS, J2EE, or SOAP but they do ask about the system and the benefits of it.

If you have the drive, you can work in any industry you want. More to the point, you can take your skills of Java, programming, and analysis and start crafting systems that will possibly change the way people work. Some of you are doing it already. I take my hat off to you.

One thing that came out of all of this is how much users are not really bothered about technical requirements. As I said before, RSS feeds didn't figure much in anyone's thought process, quite the opposite from where I sat. This train of thought was backed up by a piece on the Forbes Web site stating that 91% of Internet users still don't know what RSS actually means. A sobering thought considering the number of RSS libraries that are out in the Javascape.

Another surprise was trying to find an auction engine. You'd think that SourceForge or one of the other open source repositories would have something. With the exception of a demo engine for Hibernate, there was nothing that I could find. Searching on SourceForge brought up nothing; plenty of utilities to put the last bid on eBay, but not an engine. At the end of the day it's no real worry. I had the ability to put my own system together. In terms of getting something out in the open though, writing my own cost me in time. I missed the Paris Air Show by two weeks.

So now I stand here as a founder of a company, all my own technology, no funding (that's another long story for another time) and I am loving every minute. If you've ever wondered, "what if....?" don't wonder any more. Do it!