Thoughts and ponderings: April 2005 Archives

In an article titled 'Is Computer Science Science?', published in Communications of the ACM (48) 4 of April 2005 (pages 27-31), Peter J. Denning explores the scientific nature of computer science. As he correctly postulates, this is a discussion that has been going on for a long time, and regularly comes back to me personally when I am working my Ph.D. research, when I ask myself questions like 'Is the work that I am doing really of enough academic quality?', including all the variations that are possible.

In the article, Denning explores the difference between pure research, which focuses on knowledge for the sake of knowledge, and applied research, which focuses on knowledge of demonstrable utility. The other distinction that he makes is the one between qualitative (inexact) science en quantitative (exact) science, which deals with prediction and verification by means of observation, measurement and experiment.

My own research focuses on certain aspects of web services technology. By taking a more conceptual perspective, I explore the security implications of this new paradigm for developing loosely coupled and cooperating information systems.

My drive for exploring the security implications is not "because I can", but because I believe that security is an essential success factor in reaching a situation in which information systems all over the world are able to interact and cooperate in ways that we can currently only dream of.

This point of view, combined with the definitions given above, classifies my research as applied research; I want to further the state of technology by understanding its use better. Or, in other words, by creating knowledge of demonstrable utility.

My research methodology is mostly engineering driven. I develop models and I use these models to bridge the gap between "the real world" and "the technology world" by exploring the differences between the model and both worlds. This is done through case studies to validate the match with the real world and via prototype development to validate the match with the technology world.

Looking at the difference between qualitative and quantitative science, my methdology is a quantitative one. My approach is model-driven, which is validated by experimentation.

So, combining the aspect of 'why do you research' (applied vs. pure) and 'how do you research' (quantitative vs. qualitative), I feel confident to position my work in the category of applied quantitative research.

While writing an article for a conference on cooperative information systems, it took me ten pages to come to the conclusion that I did not know what I wanted to write.

This in itself is worrysome, since it can either mean that I do not have enough grasp on the material that I want to publish, or that the material is simply not there. After carefully re-reading the pages that I wrote, I decided to rearrange the structure of the paper a little and I switched the second paragraph with one that was close to the end. Suddenly, it made sense again!

By pouring my thoughts onto paper into a structured format, I realised that my conclusion was really an introduction, and vice versa. A valuable lesson!

Another day, another semester. Today, I have started preparation work for Database Systems course for the 2004 fall semester. While trying to come up with nice SQL assignment, I could not help but realise that the year 2005 is creaping up on us. Where did time go?

Anyhow; as always, I gather that the semester will be an uneventful once again. I will lay out the rules of the game during the first lecture, only to have the first "offender" try to break them a few weeks later. Like other years, that student will be the (first) sacrifical lamb, and life will be good for the rest of the year. Let's see how much mutton I will accrue this time around.

Let's see who will be the first.....

I have been active in online communities for a long time. Having been introduced to the wonderful world of the Internet in the late eighties when I took out Clifford Stoll's Cuckoo's egg from the public library, I was enthousiastic (to say the least) when I became a student at Tilburg University in 1992.

I was lucky enough to be selected for a pilot program for "Student Email". The membership of a small elite group of Internet enabled students (as I viewed myself) spurred me to look further. At university, I found a "Library information management system" (LIBS) which consisted of predefined telnet and/or gopher sessions organized in a menu structure. Fortunately, some of the telnet sessions were broken and generously provided us with a prompt that allowed us to connect anywhere we wanted.

It was then that I started hanging around on bulletin board systems. The first one that I spent time on was on a bulletin board at the US national center for supercomputing applications. After getting booted of that system because we were not American, we ended up at Auggie BBS (currently residing at bbs.sunspot.org). There, I was introduced to other online activities such as IRC, muds, moo's, empire games, etc.

The reason for this remeniscing is that for one reason of the other, I am getting the impression that there is a critical mass to these online communitities that should never be exceeded. If, for some reason, the community grows too large, it will explode. Any online community that I have ever been a part of blew up at one point in time. Usually, the fall-out affects many people deeply. The fact that the community is online does not mean that the members of that community do not care. People are deeply involved and take such breakups very hard.

The cause of this "life cycle of online communities" eludes me. Is it simply a matter of too many people trying to push their way of thinking onto the comminity, or is it simple statistics that determines that in any community, the few "rotten apples" will show up.

Ego, or Destiny?