Cringely published an amazing paper on crime in the USA. Turns out that in 1982, a study was paid-for by the American Department of Justice. Three people were involved: Michael Block, Fred Nold, and Sandy Lerner. Cringely believes their study showed that the current sentencing guidelines would lead to a poor, more crime-ridden USA (and it did). The study was “hidden away”. Turns out that killed himself in 1983. Block became a law professor and won’t comment to Cringely about the study. Sandy Lerner went on to found Cisco.

A few things are amazing. The suicide of a researcher who possibly felt like a loser. It reminds me of Wallace Carothers who invented Nylon. It is unclear to me how you can feel like a loser after inventing Nylon, but apparently someone did. The second one is that the USA knows and knew that they were headed for a crime-ridden society and they went ahead anyhow. Why? I can’t figure it out. Lastly, there is the little detail that the statistician part of the study, Sandy Lerner, founded Cisco. This is an interesting contrast with the other fellow who killed himself.

Thanks to my colleague Jean Robillard, I found out that philosophers do Knowledge Management too! Following a request I made, Jean suggested I read an Outline of a Theory of Strongly Semantic Information by L. Floridi.

Of course, I’m a naïve reader, but still. I think I grasped some very important things.

He starts out by asking how much information is there in a statement? Well, in a finite discrete world (the realm where Floridi appears to live), you can reasonably define “information content” in terms of how many possibilities the statement rules out. For example, if my world is made of two balls, each of which can be either red or blue, so my world has 4 possible states, and I say that “ball 1 is blue”, there are only 2 possibilities left (ball 2 is either red or blue) so I could say that I’ve ruled out 2 possibilities and so my information content is 2. If I say “both balls are blue”, my information content is 4. You can see right away that a self-contradictory statement (”ball 1 is blue, both balls are red”) rules out all possibilities as well, so it has maximal information content. A tautology (”ball 1 is either blue or red”) has 0 information content. Floridi is annoyed by the fact that a self-contradictory statement has maximal information content.

In section 5, he points out that statements are not only either true or false, but they have a degree of discrepancy. So, for example, I can say that I have some balls. This is a true statement, but with high discrepancy. However, I can say that I have 3 balls when in fact I have 2 balls and while false, this is a statement with lower discrepancy, and maybe a more useful statement. Apparently, he borrows this idea from Popper, but no doubt this is not a new idea.

He comes up with conditions on a possible measure of discrepancy between -1 and 1. -1 means that the statement is totally false and matches no possible situation (”I have 2 and 3 balls”), 0 means that you have a very precise and true statement (”I have 2 balls”), and 1 means that I have a true, but maximally vague statement (”I have some number of balls”). What he is getting at is that both extremes (-1 and 1) are equally unuseful, but that things near zero are equally useful (either false or true). Let’s call this value upsilon.

Then, he defines the degree of informativeness as 1-upsilon^2.

This solves the problem we had before. The statement “ball 1 is blue, both balls are red” will now have an upsilon value somewhere between -1 and 0, so it will have some degree of informativeness, but nothing close to the maximal. The statement “ball 2 is either red or blue” will upsilon = 1 and so will have a degree of informativeness of 0. Finally, “ball 1 is blue” will have upsilon positive but less than 1, and possibly close to 0, so that it will have a good degree of informativeness.

That’s what I got out of it for now.

It appears that my favorite Montreal newspaper, «Le Devoir» has some sort of blog. I’m not quite sure how it works yet, but they have RSS feeds and everything a good blog should have.

They say they want to do it to improve communication with their readers.

As is often the case, however, their RSS feed has only a few words in each description… making it useless to me. If you want to force people to come to your blog, take into account that people’s time is precious and they can only visit so many sites daily. At least, it looks like the content is free.

There is actually something troubling about «Le Devoir» on-line edition. They’ve opted for a free content/no registration on half the news as long as you receive the journal (the rest of the news being totally free to all). Alas, you have to log on and so on, so I limit myself to the free content. One wonders whether they wouldn’t be better off opening up totally. The blogging goes in the right direction. Don’t hide your product.

Update (due to Zeke): According to Paul Wells from MacLeans, «Le Devoir» is one of the only journals in North American making a profit out of the on-line edition.

We just submited an article to the Journal of Algorithms and we were told that starting in 2003, the editors have stopped accepting papers. One alternative appears to be ACM Transactions on Algorithms.

It seems like the entire board of the Journal of Algorithms had resigned some time ago. I had no idea that Elsevier and other big publishers were in such troubles. I had heard about the Journal of Machine Learning…

It feels like soon, all the big journals will have moved to an open or semi-open setup. Very scary for big publishers. Very scary. Yes, they’ve been making ever larger profits, but it may all come down to a stop really soon. Tipping point coming!

Interesting ACM newsletter on e-Commerce: SIGecom Exchanges.

Also, quite a number interesting track for SAC 2005 including data streams and e-commerce.

I also found out about IEEE CEC 2005.

Ernie’s 3D Pancakes has a post on anonymous academic bloggers. To me, this is an interesting question. I use my own name everywhere on this blog. You can easily figure out where I work, what I teach and to whom, where I publish and so on. You can even find who my son is and so on. I think that Jeff correctly points out that feeling you need to be anonymous is probably misguided. The likelyhood that a colleague is going to come to my blog, read it, be insulted, and try to hurt me on the job, is very, very slim. One reason for that is that I would never bad mouth a colleague on my blog: it just wouldn’t be fun and interesting for my target audience. The likelyhood that a reviewer of a paper I submitted would come on my blog and be insulted and reject my paper is also very slim. However, reviewers have many more reasons to wrongly reject a paper and if you start worrying about this sort of thing, you are not out of the woods!

So, I use my own name. There.

Owen told me about SIGCSE tonight. I never knew that ACM had a interest group in Computer Science Education… maybe I should join?