Archive for December 2005
How Ubuntu saved my WinXP (and my day)
excuse the extreme geeky digression.
I’ve been enjoying Ubuntu Linux on my laptop for the last couple of months. unfortunatly, I’m giving a talk that requires ToonTalk, which does not yet run on Linux. So I bit it, and installed winXP. And SP2. And enabled updates. and waited for them to download. and installed them. and checked for updated again. and had to install the new update service. and checked again. and waited for them to download. and installed. God I miss RPMs! You know, Linux, it’s so easy..
Anyway, just when I was happy, even had FireFox and Toontalk up and running, I foolishly went into disk managment and set the Linux partition to “active”.
I mean, you’de think that active means active, and boot means boot. no. I restart and get a “can’t load OS”. damn.
Don’t really feel like starting from square one, now do I?
No worries.
Insert the Ubuntu live cd, boot ‘er up, and open a terminal. Then:
> cd /dev
> sudo chmod a+rwx hda*
> parted
> (parted) set 2 boot on
done.
They know their media
Want to raise some money for your student group? Promote Linux? feel like a TV star? No problem. Just put your knowledge into a documentary / tutorial, and stick it up on video.google. Ssam Media!
http://video.google.com/videoplay?docid=-6104490811311898236
spot the differences
The Economist ran a review of Jerome Karabel’s The Chosen. Fascinating. Yet another nail in the cofin of the myth of American Egalitarianism.
Turns out that in the 1920s top American Universities where concerned by the proliferation of Jewish students. Not that they were racist, they were simply taking their economic responsibilities seriously. You see, in Colombia, so many Jew boys have entered that the nice rich protestants are now staying away. Solution? add character based selection. After all, we all know that Jews are “effeminates, the precious and affected, the unstable”, while private school boys were “virile, masculine, red-blooded he-men”.
Something like this -
Just to show that it was all business, in the 1960s the guvs were worried that the campuses might fall pray to racial riots. The sollution this time? Make it easier for black students to get accepted so that you can claim immunity by virtue of diversity. Unfortunatly, you can’t differentiate blacks by their manliness (or, maybe you’re afraid to), so a specific diversity clause needed to be added.
put your friends in their place
Buddy map is a social networking tool that mashes your buddies over a google map.
dawn yang dafne teo and now bruna bianco, or ant colony information dynamics
So, I’ve said scale-free networks, I’ve said fish-school statistical mechanics, and here it is. Msrs. Ramos, Fernandes and Rosa have a model of social learning based on multi-agent simulation of ant colonies.
They begin by explaining:
Flocks of migrating birds and schools of fish are familiar examples of spatial selforganized patterns formed by living organisms through social foraging. Such aggregation patterns are observed not only in colonies of organisms as simple as single-cell bacteria, as interesting as social insects like ants and termites as well as in colonies of multi-cellular vertebrates as complex as birds and fish but also in human societies. Wasps, bees, ants and termites all make effective use of their environment and resources by displaying collective “swarm� intelligence. For example, termite colonies build nests with a complexity far beyond the comprehension of the individual termite, while ant colonies dynamically allocate labor to various vital tasks such as foraging or defense without any central decision-making ability. Slime mould is another perfect example. These are very simple cellular organisms with limited motile and sensory capabilities, but in times of food shortage they aggregate to form a mobile slug capable of transporting the assembled individuals to a new feeding area. Should food shortage persist, they then form into a fruiting body that disperses their spores using the wind, thus ensuring the survival of the colony.
And conclude:
Evolution of mass behaviours on time are difficult to predict, since the global behaviour is the result of many part relations operating in their own local neighbourhood. The emergence of network trails in ant colonies, for instance, are the product of several simple and local interactions that can evolve to complex patterns, which in some sense translate a meta-behaviour of that swarm. Moreover, the translation of one kind of low-level (present in a large number) to one meta-level is minimal. Although that behaviour is specified (and somehow constrained), there is minimal specification of the mechanism required to generate that behaviour; global behaviour evolves from the many relations of multiple simple behaviours, without global coordination (i.e. from local interactions to global complexity. One paradigmatic and abstract example is the notion, within a specified population, of common-sense, being the meta-result a type of collective-conscience. There is some evidence that our brain as well as many other complex systems, operates in the same way, and as a consequence collective perception capabilities could be derived from emergent properties, which cannot be neglected in any pattern search algorithm. These systems show in general, interesting and desirable features as flexibility (e.g. the brain is able to cope with incorrect, ambiguous or distorted information, or even to deal with unforeseen or new situations without showing abrupt performance breakdown) or versability, robustness (keep functioning even when some parts are locally damaged), and they operate in a massively parallel fashion. Present results point to that type of interesting features.
