I'm at the OKFestival this week. Tons of inspiring talks. I'm writing this post incrementally because I've got lots of notes floating around... so sorry for the mess!

For the next two years, I'll be collaborating with the Art of Making project. The project investigates the processes involved in the carving of stone during the Roman period. In particular, it aims to analyze them using the insights and understanding Peter Rockwell (son of Norman Rockwell) developed during his lifelong experience as a sculptor. Eventually, we will present these results through a freely accessible online digital resource that guides users through examples of stone carving. In this post, I just wanted to report on the very first discussions I had with the sculpture and art scholars I'm working with for the purpose of creating a shared model for this domain.

In this post, I'm presenting an organized version of the notes I've taken after reading a beautiful paper by Luc Steels titled "The Symbol Grounding Problem has been solved. So What's Next?"

One of the cool things you get to do with a Kindle is being able to access your pdf library on the go; however soon enough I ran into the problem of getting all the pdf files to show up with the right metadata, e.g. name, author, creator etc..

Last month or so I started a Knowledge Representation workshop with my colleagues at CCH. The basic idea is to take a broad perspective on the various topics related to KR, and then focus on the digital humanities so to see how these approaches and technologies can be best applied to our domain.

A new interesting book on Knowledge Technologies from Nick Milton is available online. It is meant to be read also by novices so it's deliberately not too technical or complex. I had a quick look at it this morning, and I think that it is interesting even for who's already familiar with all this stuff, cause it gives a nice overall perspective on the field. Never too fanatic about the 'semantic' promises, sober and realistic when describing the features and advantages of these technologies.

It's nice when a few people's interests happen to converge. You start tackling problems together, and learning as a group. This is what happened KMi recently with the Epistemic Logic interest group. We've decided to start a seminar, trying to make sense of the 'epistemic logic' area and possibly draw some useful tips from it.

A very interesting article from Harry Halpin, whose work lies at the borderline between history of science (of computer science especially, I gather) and (Semantic) Web. I think it should be a must-read for all SW practitioners, so to understand where we (yes - I'm part of them..) stand in relation to the past...

I wrote something in the last weeks about the difficulties related to the modeling of representations and their contents. So today I just read this article by Stefano Mazzocchi, which basically hints at the same issue, but from an 'RDF perspective'. It's a very interesting article, so I just wanted to quote and comment a few passages

It's been three days that I'm struggling with concepts of content, form, representation and so on.. I wonder whether there's a well-formalized theory of representations out there.. the one in DOLCE is a useful design pattern, but I'm still reluctant to say that it is complete (I hope I'll find out to be wrong). Another clever view of the issue can be found in a tutorial by Richiiro Mizoguchi, and this is what this post is about.

If you want to work with knowledge, you need firstly to 'extract' it. No doubt about that. There're various techniques for doing this, and a whole research area called knowledge elicitation. I've recently had an inspiring meeting with Gordon Rugg, who convinced me it was worth spending time to validate my ontology, using a card sorting KA technique. I started this morning, and it is a lot of fun I must say.. Now I know what psychologists feel when they spend hours and hours interviewing people... it's something like digging the ground, from the point of view of an archaeologist. Anyway, card-sorting is quite an interesting technique (a special issue of Expert Systems is completely dedicated to it). In a nutshell, card sorting involves categorizing a set of pictures, objects or labelled cards into distinct groups using a single criterion. The results can be very intriguing..