Then you need to look no further than the JAOO Today blog! Here you can read all about JAOO, the events, the sessions, and much more.

Of course, you can also follow the JAOO Twittering.

We have created a 1-day program with the business people in mind with the possibility to buy a one day ticket to our conference at Tuesday October 6, 2009. This way the developers can bring their boss and the boss can bring her developers and even though they will follow seperate tracks, it can be a bonding experience for the whole department – especially with the social events during the evening.

The speakers will be specialists Michael T. Nygard, Keith Braithwaite, Eoin Woods, Jes R. Dragsted and Toby Young.

You can read the invitation here with instruction on how to register.

Business drivers & challenges in software development:

Alignment of Business and IT strategy with Enterprise architecture.
Jes Rude Dragsted, ATP, DK

Qualitative and quantitative metrics to prioritize business technology projects.
Toby Young, City Index, UK

Enterprise Architecture, Top 10 Mistakes.
Eoin Woods, Barclays Global Investors, UK

Buy it, Build it, Download It, or Browse It? Achieving Effectiveness with Enterprise Applications.
Michael Nygard, US

Real decisions, real responsibility, real management.
Keith Braithwaite, Zuhlke, UK

Last year all of the 1200 starting numbers were sold out, so if you want to join in this year you should register very soon. The run takes place Tuesday October 6.

You can register here.

We videotaped the presentation and you can watch the video here:

After talking a lot about the Bespin and performance, and bragging about how well Bespin performes on a large code base, the familiar demo ghost showed up and teased Ben and Dion: Bespin was surprisingly slow on a larger code base and according to Ben and Dion it hadn’t been the day before. Fortunately Ben later found the reason and blogged about it. And the culprit was the tab character and how it is supported in Bespin. In his blog post Ben discusses the possible ways to support tabs in code and their effect on performance including the fix that he made after the failed Bespin demo at QCon.

You can also download the slides from this presentation.

This year we are also trying something new on Twitter: A contest.

The rules are simple: This week we will ask five questions on Twitter and every time you answer a question correctly (one answer pr. question, answer on Twitter and tag it with #JAOO_contest), you get a ticket for our JAOO lottery. At the end of the contest, we will draw three winners among the correct answers, and these will be announced on Twitter – @JAOO. If you win, please write a direct message to us with your address so we can send you the prize.

The prizes are:

• An edition of the book Pattern-Oriented Software Architecture – a Pattern Language for Distributed Computing signed by Frank Buschmann and Kevlin Henney.
• An edition of the book Pattern-Oriented Software Architecture – on Patterns and Pattern Languages signed by Frank Buschmann and Kevlin Henney.
• A ticket for the three conference days at JAOO Aarhus 2009, Okt. 5-7.

See you on Twitter and at the JAOO conference!

One of the important parts is the food and this year we decided to try something new. We have testet food from a lot of great restaurants here in Aarhus (we work hard for you and are glad to do it) and to really put the chefs to the test, we decided to serve the food from the best restaurant at lunch for a whole week at our company, Trifork.

As you can see the food looked delicious and trust me: it tasted fantastic. We hope you like it too – see you at JAOO.

In the beginning this was done as a show in the exhibition hall. This was not fortunate for many reasons, one of them being that at our biggest conference, JAOO in Denmark, a whole kindergarten would sometimes pass by in order to see a show in other parts of the conference building (since this was also the music hall of the city).

Later we shuffled the speakers into a small room and started calling it the “Solution Track”. It was evident in the schedule that these product descriptions could turn into sales pitches and they did not get a lot of attention nor attendees.

Then suddenly, one year, we had so many presentations from the exhibitors and sponsors, that we were able to create a mini-schedule for them with themes and abstracts and we gave them some more attention. This resulted in a lot more attendees and happier sponsors and exhibitors.

This is a solution we have been practising for a few conferences now, and it works well. The presentations in the solution tracks get good evaluations, both because they actually send good speakers to the conferences, but also because the expectations of the attendees are set for product descriptions, and that is what they get.

So, except for the fact that these tracks will never really be respected by the Programme Committee  :-) (which is a bunch of snobbish people like myself) the solution tracks are a great success. Last year, one of the presentations in the solution track was in the top 5 of presentations.

This year, one of our solution tracks focus on Spring technology. As can be seen, part of the speakers on that track are also in the technical program and thus, also invited by the Programme Committee.

But how should solution tracks look like in the future? What kind of presentations would you like to see in a solution track? Speakers that have been developing the products, users of the products, salespeople, CTOs from the companies that create them?

If you are a .NET person, you probably already know Ayende because of his incredible diligence as a blogger and his (usually pretty bold) opinions.

If you don’t know him, I bet you know the projects he is involved in: he is a major NHibernate committer, he is currently the lead of Castle MicroKernel/Windsor, and he maintains his own catalogue of projects (Rhino Mocks among others), and he is currently developing the NHibernate Profiler. Moreover, he is an active speaker, teacher, and consultant, and he has authored “Building Domain Specific Languages in Boo” which has recently been released on Manning.

Ayende is not only cool because of his unusually high activity level – he is also a very opinionated and pragmatic software developer, often undogmatic and incredibly self-confident, and in my opinion always very convincing. He is the origin of the JFHCI motto, he is never afraid of questioning “best practices” and sometimes playing the devil’s advocate, and he never seems to stop showing off his incredible depth AND breadth of knowledge and skill – lige e.g. when he gnawed his way through the CouchDB codebase in a few days, posting reviews of the code as he went along, not having read any Erlang code ever before he begun!

Another funny thing is that Ayende seems to identify himself with a rhinoceros, hence the naming scheme used in most of the artifacts he produces – and it takes only a few seconds of exposure to his nature to understand why he sees himself that way: rhinos are big and stubborn, and they push forward with great force. Rhinos have relatively small brains, however (at least compared to their size), but I assure you that this is one characteristic that he does NOT share with rhinos

I’m guessing that his presentation will show us some of the lesser-known features – a few NHibertricks, if you will – of some popular OR/M that he has put to use in various projects to reduce friction. As I am using some popular .NET OR/M myself in various projects as well, I am very curious to learn some of the tricks he has up his sleeve! I just KNOW this is going to be good!

When my companion and I wrote our Master’s Thesis “Garbage Collection in a gbeta Virtual Machine with the Train Algorithm” we read one of Urs Hölzle’s articles – “A Fast Write Barrier for Generational Garbage Collectors”. The article described how to implement an efficient write barrier, more specifically how to use a card-marking write barrier scheme. It was a short article without a lot of complex theory and it treated the subject in a clear way making it easy to understand and inspiring to read – like candy for the brain .

The purpose of a write barrier is to ensure that the garbage collector keeps track of the object heap each time a pointer is updated.

The write barrier in a generational garbage collector ensures that the remembered set of each generation is updated (in the train algorithm the remembered set is used between cars (mini generations within a train) and between trains (macro generations)). A remembered set (proposed by David Ungar) keeps track of which objects of a younger generations is referenced (kept alive) from an object residing in an older generation.

The card-marking write barrier scheme does this by dividing the heap into cards of size 2kwords. Every card has an associated bit in a separate bit vector. A store check simply marks the bit corresponding to the location being updated. At garbage collection time, the collector scans the bit vector and, whenever it finds a marked bit, examines all pointers in the corresponding card in the heap.

At JAOO Urs Hölzle will be speaking about “Energy-efficient cloud computing” and telling us how Google made their infrastructure scalable, cost effective and energy efficient.