or many years I had abstained from joining Reddit, which I subconsciously viewed as a competitor to Digg, where I was already quite prominent and sometimes called “omnipresent”. But several months ago I reconciled all this nonsense and joined the site — a site that I had been sometimes criticising over at Techrights. During the weekend Techrights reached Reddit’s front page with this submission and the comments were surprisingly decent (in the past there were Novell employees attacking messengers in the comments, as means of misdirection). Things apparently do change over time, and sometimes in a positive way.

Separately, I have set up a page for me and my fiancée before our wedding (less than half a year away) and she loves it. Finally, a friend (and fellow professor) took dozens of videos and hundreds of photos with me over the weekend because we prepare a site and product called FXI (no public details on it just yet, but here is a a clue). There are some FXI logo ideas (the product is not yet up on the Web). This was made with the GIMP of course and it is far from final, only concepts.

Next weekend I hope to finally upload a lot of code I wrote in 2011-2012, accompanied by whitepapers and documentation. Busy week ahead…

ed and green hues represent the matching from dual-scale classifiers of the rear of cars. Some more bugs were removed in this latest iteration of the implementation.

he circle at the left shows whether the car is getting closer (white) or going away (red). The size of the circle is indicative of length.

esting shown on various cars excluded from the training set. I will soon release a PDF document to explain what my program is achieving and how.

CTIVISM for software freedom seems to have weakened somewhat. Truth be said, the FSF saw its power declining, key staff at the FSF left, and Android, which champions the Linux ideology in some sense, does almost nothing for GNU and very little for software freedom. Its code is Free/open source, but it also embraces DRM and other nasties.

Speaking for myself, I intend to extend my activism to freedom in general, as well as justice. For the past 5 years or so I’ve described myself as an advocate of fair competition. This also implies greater parity and more opportunities for new players, as opposed to the cabal of corporations that reign the status quo.

Yesterday I spent a long time speaking to a gentleman from Afghanistan. He came to the UK some years ago and I’ve had several opportunities to speak to him and ultimately befriend him. Things in his own country are grim, he told me, but he does not seem bothered by police state culture in the West, let alone the unjust inequality. For some people, perhaps not idealists or perfectionists, a world full of flaws is OK and not worth improving. But one needs to always be reminded that progress in society never happens without the hard work of those sacrificing their life for good causes.

George Bernard Shaw once said: “The resonable man adapts himself to the world; the unresonable on persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man.”

NE of my clients, SH Property, recently had the site redesigned. As I do architecture and programming for a living in a field which is very competitive, I often have to depend on tools that lower down the costs. The means by which I rebranded the site at the domain level only required Apache redirects, which sure saves a lot of money. It is worth noting that Libre/Open Source software was used almost exclusively to build the site and the previous logo/header/banner (when it was called SJ Property Investment), for instance, was created using the GIMP. It was temporary. Here it is:

While Libre/Open Source software usually costs nothing to acquire, it’s far from impossible to make a living with it; services and skills — requiring manpower — are the real scarcity.

ASED on further experiments, at the expense of performance in terms of framerate we can easily improve accuracy to the point of perfect tracking for particular cars. This is done by increasing the sample (window) size on which the tree is trained.

Using 20 positives and only 10 negatives for training.

20 positives and 20 negatives with a sample size of 40×40