In my attempt to bring a different experience in showing concrete implementations of AI search algorithms (I try not to give students the same problem on a particular course every semester), I’ve thought of giving the students a machine problem concerning a game called sokoban. Particularly, I have asked them to implement the search tree data structureand use the A* search algorithm as described in Russell’s and Norvig’s book “Artificial Intelligence: A Modern Approach” to solve certain maps. Luckily, there is an available Java-based and ready-to-use implementation of the sokoban environment as well as maps downloadable from the Arctic Blue Games website. This should make it a bit easier for the students since they no longer have to worry about implementing their own sokoban environment: they just have to study the documentation provided for with the codes.

For my class in Automata Theory and Computability (of which I am co-teaching with my mentor, Henry Adorna), I’m experimenting on incorporating the use of JFLAP for the lessons. JFLAP is a Java-based software that allows the user to build and simulate automata and Turing machine-based models. This was introduced / suggested to me by Sir Henri prior to the start of the semester. On initial play-around, it looks OK and I foresee I could use this as a “fast forward” mechanism for breezing through demonstration of machine construction and even simulation. I have used this in class once already, albeit swiftly, and I couldn’t really garner sufficient data from there yet, although I believe that the base case was satisfied properly (based on the aforementioned play-around). I really hope I’ll be able to use this extensively in the future sessions. Will keep you posted on the findings.

Diablo 3 is OFFICIAL ladies and gents!

Guess the DAWWInci Code isn’t as blasphemous as it seemed after all! :)

Registration’s over, and now all that’s left is to hold court. Here are what I will be teaching this semester:

  • Introduction to Automata Theory and Computation
  • Artificial Intelligence
  • Data Mining

Additionally, this semester also marks my first crack at a PhD, so I’ll be taking these subjects (as a student!) as coursework for the degree program:

  • Argumentation Theory
  • Coding and Information Theory

OOXML backwards compatibility led Microsoft to ODF

Excerpt from ZDNet:

Microsoft announced on Wednesday it will support ODF version 1.1 in the release of Office 2007 Service Pack 2 (SP2), scheduled for the first half of 2009. The company will also support PDF and XPS in Office 2007 SP2. OOXML is partially supported in the current version of Microsoft’s office productivity suite and, according to Microsoft’s announcement, will not be fully supported in Office until the release of “Office 14″, which as yet has no confirmed release date.

I could be wrong, but I can’t help remember the way Microsoft is trying to acquire Yahoo! through its stockholders when it failed to purchase the company as a whole. In the same manner, MS is now “supporting” ODF in order to further its bid to have its version of OOXML be recognized as an international standard.

Analogically speaking, if MS can’t take you over, they’ll get you from under. I smell a rat. A tunnel rat, that is.

Hello one and all!

Welcome to my other blog dedicated to everything computer science… at least in my realm. ;)

Here, I’ll be posting topics that will serve as a sort of supplementary reading material to some of the subjects I teach, and some commentaries on key and trivial issues in IT and CS.

All this for great justice! :)