Anyway, what we royal subjects like to really moan about these days is Rip-Off Britain. We just love a bargain more than anything, and thanks to globalisation and the sweatshops it creates, we’re swamped with more bargains than ever before. Jeans for a fiver? No problem. DVD player for fifteen quid? You got it. 5p more for a litre of petrol?

<silence>

You what? You’re ‘avin’ a larf aintcha?

Just as we love a bargain, we absolutely detest being ripped off. And as our pound gets stronger against the dollar and the euro, the more, so the theory goes, retailers will sneakily mark their goods up, and the more we get ripped off.

A case in point is computer software. I read today about the new version of Roxio’s CD/DVD software, Toast. I checked it out on their website and it’s yours for $79.99. Unless you’re a Brit that is, in which case it’s £69.99 which translates to, at today’s exchange rate, $136.27. A SEVENTY percent increase.

And this happens across the board. Adobe charge $649 for Photoshop CS2 to North Americans and offer it to us at the bargain price of £569.88, or $1109.59. 70% again. Apple are a little more considerate and only mark up OS X by 34% (although songs cost 55% more on iTunes UK than on iTunes US). These blatant rip-offs are dwarfed, however, by Microsoft’s Office 2003 Professional which is a steep $379.99 on Amazon.com whereas Amazon.co.uk is charging a positively vertical £437.99 ($852.79). A colossal mark up of 124%.

And there’s not a lot we can do, other than, I suppose, moan.

Some people just object and persevere with GIMP and OpenOffice. Some will try out an ‘evaluation copy’ they found via Bitorrent. Most Brits, though, will begrudgingly accept it, buy it on credit, put the kettle on and get back to the Daily Mail. Now, about those trains…

The essence of the theory is that the more oil we find, the more we suck out of the ground every year, and because oil isn’t renewable (not in our lifetime anyway) at some point the amount we suck out starts to decrease rather than increase. But, we are told, this decrease isn’t a period of steady decline where we all start running our SUVs on vegetable oil, it’s exponential which means production will fall rapidly until the amount of energy needed to dig it up is greater than the energy you get from the oil itself. At this point, no one can get hold of any oil which means the world will come to a grinding halt and we all strip to our loincloths, become feral savages and start building wattle and daub huts in the hills. Grrr.

But I digress.

Java developers have had it good for years and right now it’s boom time. It’s top of the heap in the latest popularity index, there are more Java jobs on www.jobserve.com, www.monster.co.uk and www.jobsite.co.uk than any other language and they pay more. Even Universities are teaching it to freshmen*.

Now, here are your typical skills a Java developer needs experience of these days on your average ‘enterprise’ project:

• Java (’core’)
• JMS
• JDBC
• Struts
• JSP/Servlets
• Hibernate
• Spring
• XML/XSLT
• Ant
• Eclipse/IntelliJ

That’s some barrier to entry and it’s getting higher all the time, with the increase in popularity of SOA and Web Services, AJAX, and no doubt several other ‘new’ frameworks and enterprise patterns which I’ve yet to hear about. Here’s a picture:

I think Java-based production is peaking and productivity has long-since peaked — the ratio of satisfaction returned on energy invested is spiralling down.

We have reached Peak Java.

Programmers like being productive. They like new and cutting edge technologies. If I had a fiver for every developer I came across who was learning Python or Ruby on Rails in their spare time, I’d have… thirty quid or something.

There’s change coming. People are talking about it more and more. Java will be around for a long time but the next generation of programmers, today’s hackers who will be tomorrows CTOs, haven’t got time nor the inclination to learn ‘the stack’ — they’re far too busy churning out killer web applications in PHP, Python and Rails.

As for me, I’m running to the Welsh mountains to build a hut and learn to subsist on a diet of grass and rainwater before the oil runs out.

• * http://www.joelonsoftware.com/articles/ThePerilsofJavaSchools.html

I’ve recently had this problem and I managed to solve it by not forking the batchtest task, and instead forking the junit target and setting the forkmode attribute to once, like so:

<junit printsummary="true" fork="yes" forkmode="once">
<batchtest todir="${tmp}">
</batchtest>
</junit>

Google appear to follow the release early, release often mantra of agile software development. They release products early with a ‘beta’ tag and reduced functionality, then release updates containing new features at frequent (although not regular) intervals. Gmail is a perfect example of this. Even though Yahoo have also started to roll out the beta of their Gmail-rivalling web mail system, you get the impression that their development is somewhat less agile than Google. More Microsoft than Thoughtworks, more donkey than race horse.

Rather than working in agile utopia, it’s probable that Google operate a much more formal, regimented system which combines aspects of several methodologies. They probably have significant requirements specifications, design documents and their UAT phase will be long and scrutinous (although this would be the case irrespective of development methodology), but the simplicity of their products, their productivity and the way they listen to customer feedback is a clear sign of a belief in the underlying principles of the agile manifesto.

Any Googleites out there want to share their experiences?