BloJJ

Exercise 3

It was 1988 or '89, and I was driving my first car, a white boxy Talbot Horizon that had been originally acquired on 23F, the day a civil guard broke into the Spanish Congress and popped a few bullets into the ceiling. I was going to my favorite comic book store, and was about to get my monthly fix; I was in an T-intersection, where I had to yield. Peeked left, peeked right, nobody's coming, and went right ahead, since I couldn't see a thing. But that was enough to crash into the midsection of a car that was coming from my left.
The thing could have stopped right there. We descend, produce insurance papers, maybe yield at each other "Where are your eyes?" "Why are you driving so damn fat in an interesection?", and we part in different directions, never to see each other.
Maybe that's what the other driver wanted to do, but he got his pedals mixed up, and instead of braking, he sped up. Which was not the right thing to do, since he wasn't facing ahead, but a little bit to the right. He crashed into the back of a car that was parked at the other side of the intersection. That car was jerked from his position, and lurched forward towards a person that was starting his motorbike in the space between it and another car. That person had just enough time to jump and avoid being crushed between the suddenly animated car and the other, stopped car.
All in all, 5 vehicles involved. Nobody was hurt, and we signed what we had to sign. The problem was that, since I had just bought the car, I hadn't transferred insurance from the old owner to me, so it was not clear at all if I would be covered or not. Eventually I was, but for a couple of months I was mentally adding up the tab. Instead of sleeping.
Since then, I haven't been involved in anything so bad. And I've made sure I was insured.

Etiquetas: A novel in a year

Exercise 2

I want to write a novel because... just because. I gives meaning and purpose to that part of the day, the week, the year, where you have nothing else to do. It's a challenge, too. Writing short stories is easy (not easier than writing a paper, though), but writing a novel... it's in another ballpark. And you end up knowing yourself by facing challenges and getting over them. Or under them. So novels are self-introspection devices.
Tags: a novel in a year

Alternate endings to the Harry Potter franchise

Since the actual ending has been so disappointing, there are thousands of alternative endings out there, including my third, first and second attempts. But I have found this one, from the HPPP, particularly funny, including a "Brokeback mountain" and "24" ending...
Etiquetas: hp, fanfic, harry potter, book franchises

Exercise one

From A novel in a year:

The day after my eight birthday, my father told me which one of his eight wives was my real mother. And, boy, was I disappointed.

A novel in a year

PJorge gave me a copy of this book, and it includes exercises. Since the book is in English, I guess it's only fitting to do them in that tongue, although I don't intend to write anything close in extension to a novel in English (unless I can somehow stitch together 20 papers and call them an experimental postmodern scientific novel), so I'll post them here in the category called Homework.

Back to Alife

Just knew (in a rather roundabout way) that the paper we submitted to ALIFE XI on a new ant-based clustering algorithm, called KohonAnts, has been accepted in the Alife conference. Which is just great, since this was our first paper on this new (but promising) algorithm, and it's really encouraging. And it means our return to the Alife conference, which we hadn't worked since our widely ignored paper on a model of species abundance in 1996. Pity I won't be able to go there personally, one (or several) of my coauthors will...
Etiquetas: alife, artificial life, conferences

Using a couple of processors (or more)

I obviously harbor a lot of respect for Donald Knuth, but what he says in this interview (via /.) is plain wrong:

The machine I use today has dual processors. I get to use them both only when I’m running two independent jobs at the same time; that’s nice, but it happens only a few minutes every week

Beeeeep! Wrong. If you have a dual or quad processor, the OS is delivering processes to them all the time. So you're using all of them all the time. Right now I'm listening to music, and using a browser, while I'm running X and various widgets. That's only on the surface, deep down, there are a few dozens process running. So you're running two independent jobs (and scads more) all the time, and you get to use both processors all the time. Not to full capacity, of course (right now, my own processor only takes a 20% load), but you get to use them all the same.
Etiquetas: knuth, latex, great minds, gurus also make mistakes

Old School Python

Marcelo has mentioned I'm kind of old school (for a good meaning of the world "old"), and I really am. I was there when Guido van rossum hadn't produced 1.1 of Python and
still answered newbies in the USENET newsgroup.
However, I seem to remember that later on, after yet another silly question, Guido van Rossum himself answered with something like "You make lots of questions you should solve yourself", but have been unable to find it. Maybe it was direct email. BTW, that might explain the fact tha I stick to Perl. Which, at that time, was into version 4.

Field guide to genetic programming

Today the del.icio.us hot list featured A field guide to Genetic Programming, a book that was presented during EvoStar and which is written by several big characters from evolutionary computation: Bill Langdom, Riccardo Poli and Nic McPhee. Yo can buy a printed copy from Lulu or you can download it for free; it's got a CC license. One of the coolest thing is the flip-o-rama movie in the bottom corner of the pages, showing the evolution of a GP solution.

Lost papers

From time to time, I find myself going back to papers I wrote 8 or 10 or 15 years ago and finding them, if not outstanding, at least publishable; if only because I'm reading some other papers that deal with the same problem and were published (not by me) several years after that.
What happened to them? Who knows. Maybe they were sent to a journal and they flunked, and then sources were lost in a hard disk crash, along with the experiments that feed them, which eventually condemned them to oblivion. If I remember correctly, that was the case of a paper I wrote around 10 years ago about how to use (or not) non-coding regions of chromosomes in evolution. No sources, no experiment, so the only way out now is to reproduce research all over again. Which is a bore.
It all boils down to a piece of advice to my younger self: nothing is gained with a paper going down the drain. Publish it. Anywhere. If it flunks over and over again, upload it to ArXiV and be done with it. Keep it backed up somewhere, because ten years on, you will be be sorry for not having published it.