Welcome to 2010! This is going to be a very interesting and terrifying year for me, because in theory, I will have successfully defended my Ph.D. by the end of this year.

Today I jumped straight into things, and finished my final submission for ASWEC 2010. As part of this submission, I exported all of the metrics I obtained during this research into a format accessible to the web, as discussed earlier.

Before the end of last year, I also submitted a short paper to NZCSRSC 2010. This paper simply covers my research, my results and my future direction, and it will also be an opportunity to create my first poster. Once accepted, I will be putting a copy of this paper online as well.

Before the end of January I would like to have big chunks of my thesis complete. This will let me judge exactly what kind of resources I need to complete this by my goal end date of May (with two months spare).

This week I have been placed on the Program Committee for the AI Mashup Challenge of ESWC 2010, thanks to my supervisor. I am really excited, this is my first PC position and I am looking forward to participating!

Today I happily got notification that my paper submission on Model Completion to ASWEC 2010 has been accepted, so I will be presenting this paper next April. I got some really good feedback, so will hopefully get the camera-ready copy finished before the end of December.

As part of the publication process, I will also be putting some of the raw data from the project online onto openiaml.org in the near future.

After a couple of hours of convincing Eclipse to play nicely, version 0.4.1 of the IAML modelling platform has been released.

This is a minor release which internally restructures the way that operational modelling is generated into code. This allows for loops, which previously would cause code generation to fail.

It should now also be possible to install the plugin without having to install all prerequisite features as well. The process to discover the correct balance was not straightforward, and I'm not sure if it's possible to add this into some sort of automated test.

I am disappointed that it took two months to release this minor update, but this is mostly due to research requirements as well as the conference/teaching over the last few weeks.

This week has been both productive and painfully slow.

After ASE two weeks ago and teaching a second-level paper last week, I decided to put writing on a little bit of a break and focus on the modelling language. (It needs more expressivity before I can apply sensible model checking to it.) Progress was fast until yesterday and today; I just spent 3 hours tracking down that in upgrading PHP 5.2.4 to 5.3, an ini setting default was changed.

This week I made a very major change to the way operations and conditions are executed internally, so lots of my test cases are no longer passing. I am strongly hoping that I can get all of my test cases to pass by the end of the week, which will allow me to release version 0.4.1. (I am trying to stick to a more agile and flexible release schedule in the future, with releases measured in weeks, not months.)

If you are developing in Eclipse and it is quite slow, Eclipse includes a heap meter. To enable it, go into Window > Preferences > Show Heap Status. I just found about it today, but it looks like this:



The trash can will force garbage collection, and you can also get the meter to render the maximum heap (so you know if Eclipse is approaching the maximum setting).