The 2007 Godel Prize for outstanding papers in theoretical computer science at the ACM Symposium on Theory of Computing honours two computer scientists, Alexander A. Razborov and Steven Rudich, for proving that there is no "Natural Proof" that certain computational problems in cryptography are hard to solve, and though they are thought to be unbreakable, there is no natural proof that they are secure.
Razborov, a mathematician and computational theorist, is a leading researcher at the Russian Academy of Science Steklov Mathematical Institute in Moscow, Russia. Rudich is Associate Professor of Computer Science at Carnegie Mellon University in Pittsburgh, Pa., an editor of the Journal of Cryptology - and accomplished magician. (Source)

300 years ago, Carl Linnaeus was born in Sweden. He was the creator of modern taxonomy and the binomial nomenclature. Extremely influential.

Next SIGGRAPH will feature more information about Spiderman 3 than the ordinary "Make Of" documentation. ("Spider-Man 3" featured fully-articulated, performing CG characters composed only of dynamically-generated particles and fluids for the first time.) (Source)

Writing about Poker isn't that easy these days. There's kind of disgust when bloggers write about companies advertizing Poker communities or companies making money with this type of game. For a change, we present a new scientific project: Polaris is a poker-playing computer program created by University of Alberta researchers, meaning a new milestone in artificial intelligence. The goal is to provide a Poker software that is better than any human opponent. In fact, Polaris is just one in a group of programs, each working with a different strategy, that is able to bluff. I guess the only missing feature is the poker face. A match between the software and two prominent Poker players takes place at July 23 - 24, in conjunction with the Association for the Advancement of Artificial Intelligence Conference in Vancouver, British Columbia.

Remember Pong? Sure you do. Here's a fine example of melting science and games: Stephen Taylor, student at George Mason University added realtime fluid dynamics to the classic video game and developed Plasma Pong. A fine sample of the valuation game engineering gets at universities today. (Source)

Well-known applications like Google Earth and Virtual Earth not only give you the possibility to hover above earth but to discover some cities by switching on a three-dimensional view thus giving you a hint of the city's appearance. I sometimes wish that it would be possible to visit historical sites as well. The Rome Reborn Project makes this possible with a 3D model of the eternal city. Now there's a new approach by Frank Dellaert at the Georgia Institute of Technology in Atlanta, together with his colleague Grant Schindler and Sing Bing Kang of Microsoft's research lab in Redmond, Washington. Their software constructs an animated 3D model that shows the change of a city over the years.
Coming from old photographs of the site, the 4D cities software automatically arranges these photographs according to their common features until the visibility patterns of all the buildings are consistent. It then computes a 3D animation of the city, allowing a visitor to travel through time. (Source)

PennState's statement on the importance of brands in the search engine market. Users clearly preferred their favourite search engines, even when the results show exactly the same hits. Researchers presented faked result pages of various search engines on some given search terms to test persons and asked them to evaluate the quality of the hit list. Despite the exact results, participants of the study decided that Google and Yahoo outperformed all other search engines. Branding is being tied not just to product identification but also to product performance, says Jim Jansen, assistant professor and lead researcher. So newcomers, beware! (Source)

A new book by Theseus Research CEO Karl M. Fant wants to dispel the belief that it's important to be a good mathematician for being a good computer scientist. Though I don't share this opinion I'd recommend reading the discussion about this idea. Since math is the language of science you can't do computer science without. On the other hand, you don't have to do math in your daily work, when you're, say, an administrator and try to make things done. But that's no science in the literal meaning, of course.

A digital wand that you may 'paint' sounds on is a recent project of Ph.D. candidate David Merrill (MIT) together with fellow Ph.D. candidate Hayes Raffle that will be shown at this year's SIGGRAPH. The device lets you record a sound and combine it with any surface you may scrub it against. Unfortunately, I haven't found a sound demo yet, but this tool promises fun with interesting sound effects. (Source)

Projects like this remind me of my own Storm and Stress time when I had to develop a sound system for auralizing a Virtual Reality system (long, long ago). I've got a then revolutionary Polhemus system, the Paradigm Audio Library, a Yamaha Synthesizer (think it was a SY99), an Ensoniq DP4, an Atari ST with Cubase and a Silicon Graphics target system the software had to run with. Though this was really hard work (I had to learn Unix (IRIX on a SGI), C and that sound library with a half year. This deadline was fixed.), I never had more fun and never learned more about Unix/C/socket programming than in these days. It's great pleasure to see that there's still a place for innovative and geeky projects like the one above.

An interesting approach for the visualization of network structures is presented at this year's ACM SIGCOMM by computer scientists of the University of California, San Diego. Their graphs resemble kind of digital dandelions and is for experimenting with network structures and effects on them. Besides that, their graphics are very beautiful. The system's source code will be made publicly available. (Source, Original paper)

Certainly you know the famous TV show CSI: Crime Scene Investigation. Here's a bunch of special reports to probe further.

Having read the abstract of Laura Beckwith's PhD dissertation, I'm getting a bad mood. Beckwith and her adviser, Margaret Burnett (Oregon State University) were interested in the question for a possible key to make computer science (and the CS industry) more attractive for women. One of their tasks was to analyze how people use computers to solve problems. One of their results is that men are more likely to use advanced software features than women, indepent from their confidence in computer skills. Their experimentees had to find out bugs in various formulas within a spreadsheet and were allowed to use a debugging feature that would help them in detecting the errors. Even unexperienced male users who were less confident in computer tasks used that advanced software feature more often than the female users. Even experienced femals users were unlikely to use that debugger. The only way to raise the number of women in this setting was to add some more options to the debugger. In its first version, the debugging tool let users mark values "right" or "wrong." For the next version Beckwith added options for "seems right maybe" and "seems wrong maybe," and suddenly some tests had more female debugging users than male ones.

This is plain depressing. Does software really need "pink buttons"? What's wrong in saying "1" or "0"? Will "Excel for Women" become a reality and a joke for "real" users?

Beckwith's dissertation is available online.

I would be interested in a comparable study in different cultural circles, where women aren't that shy with math: Israel, India, Iran to name a few. Anyone? (Source)

Anti-Phishing Phil, developed by computer scientists of Carnegie-Mellon University, wants to improve people's alertness on email scams and phishing emails. Nice idea (though it's using only names and sites of american bank houses)!

Best way, nonetheless, is to be suspicious about every email that claims to come from your (or another) bank. Knowing how to read the headers of email messages isn't a bad idea, either.

Two Nobel Prizes for German scientists in two days have made our politicians praise the merits of our scientific research. Indeed, the Nobel laureates Peter Grünberg (together with France's Albert Fert) in physics and Gerhart Ertl in chemistry did a great job and deserve all respect entitled to a Nobel Prize winner.

But: whoever has or had some working experience in scientific jobs in Germany knows that circumstances aren't nearly as rosy as described by Mr. Ertl, who told to the press that he never had any problems in performing his research and that he cannot listen to the ubiquitous moaning about missing money and chances in German research any longer.

It's not a coincidence that both scientists don't work at universities: Ertl works for Max Planck Society, Grünberg for Helmholtz Association, both organisations are dedicated to basic research (in contrast to Fraunhofer Gesellschaft that is more application-oriented). Additionally, one must consider that a Nobel Prize winner is honoured for something he has done several decades ago. There is no evidence for paradisic conditions at German universities in these days, the reverse is true.

The enduring brain drain in Germany provides for a difficult situation for the remaining scientists. Because of Germany's special support system for basic research done by the German Research Foundation (DFG) (most basic research here has to be supported by them because only few companies are willing to invest in basic science), the need for German scientists is ungrantable since many of them already have left the country. So Germany is falling behind in many important disciplines, escpecially biotechnology and information technology.

The Bologna Process and its development turns German universities inside out. Cutting back graduations to a low-level bachelor and providing a master's degree only for a small 'best of' selection of students won't provide for masses of future Nobel Prize winners. Instead, German students for the most part are qualified for 'academic' jobs in the industry but not for big science. Eventually to study in Germany recently means to have enough money: now our universities rake in usually 500,- Euros fee per semester - and we don't have grant programs yet that are comparable to these in the US. So pay more for less education: a whole generation of students is taken for a ride and most of them don't even notice because most protests against tuition fees just withered away.

In contrast to claims for "excellence" and support for scientific "elite" there's no room for elite research on German universities any longer. The amount of 1.9 billion Euros German gouvernment plans to invest in several universities have to reach till 2011. This will be less than Stanford's budget during a five year term.

I remember a quote from my study years that said: The good ones are leaving. Those who were studying biology with emphasis on genetics (and wanted to continue their research after getting their diploma) usually went to the US or Japan. The transformation of computer science to a mere service-oriented discipline has been lamented by me in several postings. Considering science as just a tool for further utilization, denying scientific curiosity, making scientific research exclusive for established networks only while blaring out hymns of praise about German science is nothing but window dressing and can't be swept away by the merits of two Nobel Prize winners.

Attractive insights into the Internet's structure is done by researchers of the University of Southern California Information Sciences Institute: they sent a ping to every know IP address and got a breathtaking map of all allocated Internet addresses (Source, Details).

Australia's stepping towards colossal research: Australian CSIRO (Commonwealth Scientific and Industrial Research Organization) launches the "Terabyte Science" project to make its science "internationally competitive". Go Aussies, go! And don't invest in mainframes. (Source)

Avatars' reactions on the behaviour of an anti-social, ill-mannered software bot in Second Life raises the questions on the ethics of such tests. Scientists of the University College London dept. of CS developed an automated avatar searching for other isolated avatars, in order to see if SL users expect others to give them the same personal space as the do in Real Life. The results are somewhat mixed and to tell a "male" from a "female" avatar in SL isn't maybe the best idea since the avatar may be of one sex but its user might be completely different. The question for the admissibility of such studies is nonetheless interesting. I guess that a bad experience in SL could be as unpleasing as a comparable situation in RL, depending on how much the user takes SL for real. (Source)

"Study shows Google favored over other search engines by webmasters" - you don't say! By crwaling more than 7,500 Web sites and examining their robots.txt files (if present), a Penn State study found out that 40 % of these web sites had a robots.txt (10 years before, it was a mere 10%) and that most of these mirrored search engines' market share. The study doesn't tell why. But IMO this is obvious: which principal ever heared of search engines other than Google? (Source)

Having a look towards the southern hemisphere is always rewarding. Even if you notice that your own history is repeating. Scientists at the Bioengineering Institute at the University of Auckland are modeling virtual bodies of real people and, as it was during my Sturm und Drang period ten, twelve years ago, they want to save the world. So their project tries to reduce obesity, help surgeons, support animal rights, advance drug testing and help customers choosing the right clothes. Okay, supporting animal rights and clothing people is somewhat new, but the rest is an old hat. This won't pay off, folks! (Source)

Interesting article about recent strategies in fighting SPAM. I like that "Behavioral Blacklisting.”

A recent Northwestern University study examined students' choice of social networking sites (SNS) dependent on their race, ethnicity, and parent's education. Results of this study contradict popular belief that all students are using Facebook, a very popular SNS. Eszter Hargittai, author of the study, found out, that Facebook is the site of choice for white students, Hispanic students prefer MySpace, Asian and Asian-American students like to use the less popular sites Xanga and Friendster. No special preferences could be observed for black students.

Also interesting is the statistical relevance between parental schooling and SNS preference: if a student's parents have a college degree, their children are likely to use Facebook. If their parents are less educated, MySpace is preferred by these students.

Further examinations were done on students' living situation: those living with their parents were less likely to use SNSs than their more independent living fellow students. So the belief that social network sites improve everybody's social connections has to be questioned, it seems that already connected student profit mostly from SNSs. An interesting implication of this study is that online and offline actions are more tightly bound than expected and that social and ethnical background reflect in one's online behaviour.
A total of 1060 freshmen from the University of Illinois, Chicago (UIC) were interviewed for this study. (Source)

A comparable study for German circumstances would be interesting. Though at this time there is just one major portal (the notorious StudiVZ) it could be worth investigating where (inside its discussion groups and forums) students of different backgrounds are organized.

Thinking about future technology, even that of the near future, will always include a life that is augmented by robots. Robots, up to now devices somewhere between automated production and science fiction will be ubiquitous helpers men of the future won't be able to live without. So a recent prediction by futurologist Ray Hammond hits the news and he's predicting lots of useful things: the rapid change of our lives since the 1980s will continue and even accelerate; people will get implemented sort of geo-tagging in order to send help if they feel ill, personal DNA mapping, gene-therapy drugs, and stem cell research will extend people's life up to 130 years and more. The environment will go somewhat to the extreme, but mankind will handle this by using more energy-efficient measurements, the Internet will be even more connected, and robots, to come back to this article's topic, will become important helpers in everyday life by caring for the old and the young.

But it's not all about robotic nannies. Let's have a look besides our everyday lives. Noel Sharkey, CS Dept. of the University of Sheffield has written a thoughtful article in the November issue of IEEE Computer magazine, that deals with autonomous weaponry that will change the way men will conduct war and the ethical implications.

For Sharkey it's obvious that autonomous robotics research projects have been harnessed to manufacture killing machines. Almost every leading nation is increasingly using semiautonomous robotic devices for military purposes, the US Future Combat Systems projects will have spendings estimated to exceed $230 billion. More than 4000 robots are already operating in Iraq and Afghanistan (ideal recent testbeds for evaluating weapon systems and new kinds of warfare).

Ethical questions arise for engineers and computer scientists: while robotic systems so far have been mainly used for clearing explosive devices, thus saving lives, the US Army introduced the first three Forster-Miller Talon/SWORDS (Special Weapons Observations Reconnaissance Direct-Action System) armed combat robots into Iraq.

Usage of semiautonomous weaponry usually is covered under current ethical warfare discussion. In the end, a human being decides when to kill. This is comparable to usage of traditional weaponry where a pilot might press the red button and ethical responsibilities can be loaded on him and the chain of command. But these developments will change fast according to a prediction of James Canton, CEO of the Institute for Global Futures, that within a decade detachments with 150 soldiers and 2000 robots will be possible. It is just a matter of time when fully autonomous systems will be applied: cheap manufacturing, fewer support personnel and being able to send less human soldiers put them on the agenda.

However, there are no ethical guidelines in place for the inevitable collateral civilian deaths autonomous systems pay responsible for. Though a lot of developers, engineers, scientists, and technicians are involved in the development process, there is no real chain of responsibility defined if autonomous systems one day are going to decide to kill a human being. The idea of robot consciousness and thought are just subject of the academic debate and this terrifiying scenario where machines are killing people is a new territory that isn't even rudimentally discussed by society. Then, possibilities of Artificial Intelligence (AI) are often overrated; all the worse when it's about decisions about death or life.

Military's promise to save lives by reducing the body count is a mere proposition since nobody has ever fought a robotic war. A possible side effect would be a lowering of the hurdle to fight a war because less soldiers would be involved (but this doesn't tell anything about civilian deaths) and costs could be reduced. But even the possibility of an artificial conscience would emerge new problems because an autonomous system faces a lot of difficulties when trying to find it's way. Under no circumstances such a complex AI will be ready before deployment of the first systems. We have to consider if we want handle with questions about the responsibility for potential war crimes caused by machines that are just executing killer codes. (Source; IEEE ComSoc members or pay per view)

The author is member of ACM and IEEE and thus bound to their codes of ethics.

Still looking for inspiring examples of successful women who dared to become engineers? Eleanor Baum is but one of many; here's her story.

In order to help people to make their virtual worlds or games more vivid, Stanford computer scientist Vladlen Koltun provides a neat software that creates masses of trees. Trees are well-measured objects and difficult to construct manually, so this tool is a real help for producing complex 3D data. Provided forms are ranging from naturalistic to fancyful. (Dryad homepage, Source)

Sometimes you're stumbling upon things in the press and the web you can't but think you've been put into a science fiction novel, albeit a frightening one. So I'm reading about this gene2.0 company, 23andme, where you may use a so-called spit-kit to catch some of your saliva and send it to that bio-startup. For US$ 1000,- you'll get some sort of genome analysis based on half a million single nucleotide polymorphisms in order to develop a "detailed genetic profile". But it's not the biology nor the doubtful diagnostic method that annoys me.

Stephen Murdoch, PhD doctorate at the University of Cambridge Computer Laboratory on the principal weakness of 'covert channels', that means anonymizing systems like TOR. Watching the patterns of behaviour of different users reveals their intentions, if not their identity. A nice reminder that usage of anonymity tools alone is no guarantee for surfing incognito at all. (Source)

What now? Dispersion of behaviour, adding a random element in your online practice, usage of multiple anonymity tools. A high price to escape the eye of Mordor.

They know how to water your mouth (if you are into electronics, engineering, or any other scientific discipline): IEEE digitized several historic proceedings, starting with 1913's Proceedings of the IRE, and made them accessible through IEEEXplore. Read everything about the hot new stuff of our fathers and grandfathers, learn about the history of our now common gadgets and have insight into the very first articles on TV, the transistor and many other things. Only for IEEE members and only if you subscribed to the accordings journals! Great stuff! (Source)

Blogs and bloggers were subject of several studies so far, however the readers and commenters were still unknown - until now. A recent study of University of California, Irvine, discusses the blog-reading habits of 15 participants (if 15 blog readers are enough to draw a complete picture and if there are cultural differences among the nations is another question that would be wort considering).

Maybe the fact of having done that study itself is more interesting than its results: so the researchers (led by Eric Baumer, doctoral candidate at UCI’s Donald Bren School of Information and Computer Sciences; Mark Sueyoshi, international studies and East Asian cultures undergraduate student; and Bill Tomlinson, informatics professor) found out that there are "lurkers" as well as savvy commenters, that there are different opinions about what makes a blog, that reading a blog has become a habit, like checking e-mails or reading the news, and, mostly surprising, that readers feel a "responsibility to make insightful contributions". In other words: all quiet on the western front. (Source)

Never being mainstream, Genetic Programming (GP) has though interesting applications in various fields, like financial mathematics, process control, signal processing and computer graphics, among other things. I noticed a short hype in the early 90s, but GP soon became a valid and reliable tool. I remember a big plate of GP-built graphics programmed by Karl Sims hanging on a wall at one of my former employers. His website shows many of his works that are beautiful as well as scientifically interesting. For all of you who missed the basics (and for me as a reminder) the authors of A Field Guide to Genetic Programming (Riccardo Poli, Bill Langdon, Nic McPhee) provide a fine book about GP on their website: 200+ pages about Genetic Programming, pure information science and no-nonsense. For free. Simply great!

If you're looking for female role models in science: Daphne Koller is the right one! (Source)

(The NYT article has quotes like "[S]he tries to persuade undergraduates to stay in academia and not rush off to become software engineers at start-up companies." - that brings tears in my eyes. BTW, her projects are commercially successesful - in contradiction to many windy startup enterprises.)

No problem: NASA's Java Mission-planning and Analysis for Remote Sensing (JMars) program has been made available to Open Source by Arizona State University's Mars Space Flight Facility. (Source)

If you skip all the defence-from-terrorist-attacks blurb, Visual analytics from the National Visualization and Analytics Center (NVAC) is a nifty tool for discovering structure in the masses of everyday's unstructured data. Like these elegant "do-what-I'm-thinking" softwares that you can see so often in TV shows. But this one seems to work. (Source)

RepRap is the next must-have gadget! The "replicating rapid-prototyper" makes use of additive fabrication (molten plastic builds 3D objects). Another astonishing fact is that this machine is able to produce itself! And - get prepared - it's open-source and you are allowed to build your very own RepRap! (Source)

According to a recent study by the RISP (Rhein-Ruhr-Institut für Sozialforschung und Politikberatung e.V.; an institute for applied sciences at the university and the city of Duisburg) IT workers are risking their health. IT workers working in software development projects and consulting are suffering from psychosomatic disorders - fatigue, nervosity, sleep disorder, stomach trouble - four times more often than workers in other sectors.

There is a rise in chronic exhaustion of 40%, usage of antidepressants is 60% higher than in other sectors, usage of psychotropic drugs beats the average value of all employees with even 91%.

Possible reasons are no surprise. They can be found in:

• Project work: IT workers are often working in several projects at the same time. Contradictoriness have to be solved by the individual workers.


• Particularization of work: instead of getting the whole picture, many IT workers just treat with small parts of a project, thus experiencing an assembly line work effect.


• Stress by management: restructuring in companies and the urge of self-organization for the individual worker unload management tasks on people who once decided to do technical jobs (instead of an management job).


• Changes in performance appraisals: by replacing performance ratings on efforts instead of results, workers are likely to expand their work time.


• Fear of losing the job.


• Constraint of constant self-improvement, even when companies won't pay for courses. Often IT workers train themselves in their spare time (especially freelancers).

83% of all interviewed managers (n=700; managers of companies with more than 200 employees) also admitted to have problems: concerned about losing their job they suffer from bad sleep. A third of them feels overcharged, nutrition often consists of fast food to save time. (Source)

This is not a Frankenstein scenario, we don't even have to evoke pictures of brain-manipulated people walking like zombies in order to perform the will of their master: no, recent approaches of poking electrodes into the brain deal with new methods in reconstructive surgical procedures. The target is no less than making blind people see again or paralyzed people walk. Where will this development go to? Will we become cyborgs? The institute deliveres interesting details.

Loyola University researchers created an artificial neural network that is able to predict (at a 90 percent probability) if death row inmates will get executed or not. Most significant factors were gender and school education, so if you're a dumb male you better change something! But the creepy part is that there's a piece of software able to predict one's future. Though the death row of a prison is a somewhat reduced world, it seems to me that further predictions about more important life issues will be possible in mid-future and that you might get a prognosis that you will not necessarily like and that might be nonetheless true. (Source)

A recent study of University of Minnesota researchers found educational benefits of social networking sites, contradicting the results of a Pew study three years ago. According to Minnesota, usage of sites like MySpace and Facebook adjust the computer and technology knowledge of low-income students to that of more advantaged fellows.Especially features like "technology skills, creativity, being open to new or diverse views, and communication skills" would be given by social networking sites. "Students are developing a positive attitude towards using technology systems, editing and customizing content, and thinking about online design and layout," University of Minnesota
learning technologies researcher Christine Greenhow says. (Source)

I'm not convinced about this. "Thinking about online design and layout" indeed would be needed at many MySpace sites! And what's the worth of these many social contacts you have? Are they real friends or just passerbys of your daily communication walk? I can't see why the pure quantity of contacts in my address book makes me technically affine or why this rises my creativity. I would also make a difference between people that just do the ususal copy-and-paste jobs, take a video from YouTube, some MP3s and presto! from others who really try to tell who they are and what they found interesting! (Design issues wouldn't be a problem here.) Social network sites, as well as the ordinary website, are just tools that you can use for any purposes possible. Thinking that they increase computer knowledge and other skills because they are there, isn't really convincing.

Says Dave Patterson, head of the University of California, Berkeley's Parallel Computing Laboratory at the Usenix conference in Boston two weeks ago. Parallelism would be the future and he emphasized that there's urgent need for software developers to get to know how to write parallel code. But I agree with Andrew S. Tanenbaum who states that taking this next step after serial computing will rise more problematic software than ever before, because things won't be easier when they run parallel. Maybe we should stick to Donald E. Knuth ("The Art of Computer Programming") who stricktly avoids parallelism and just delivers the best in serial computing. So it isn't dead, of course, as you have to be able to write very good serial code to be able to write good parallel code. Besides the question if there will be developers at all with skills beyond PHP and Ruby. (Source)

The Days when Wired magazine editor Chris Anderson released his "Long Tail" and told us that (once more) the Internet would change everything seem to be over. Anderson's much praised theory describes a society that is less 'hit'-driven in products and markets but that allows a lot of niche markets to grow by means of the Internet thus producing an ever growing choice of products - in contradiction to traditional offline markets. This theory gained much attention and was eagerly adapted by marketing people and IT geeks: the first ones because growing niche markets would mean more pitches, the other ones because Anderson's propositions predict more emphasis on the Internet and the people who use and design it. Especially bloggers favoured this idea, since it gave hope that even the shallowest idea would have the potential to create a market, or at least would have a voice. Lots of the recent Web 2.0 hype deals with Anderson's ideas and the fate of many obscure companies is dependant on this doctrine of salvation.

It turns out that things are not that easy.As Anita Elberse, a marketing professor at Harvard's business school, lets the numbers speak for themselves: by using rigorous statistics she points out in her article that online and offline shopping behaviour isn't that different at all. It doesn't seem that consumers are eager to break the narrow borders of the common offers in shops, but that there is an element of social conformity in cultural consumption that makes us want to have the same things as all others do, also in online shops. The boost of individuality predicted by Anderson is thwarted by our wish to let others suggest to us what we would like to do.

What does that mean to us, the common Internet users and bloggers? It means that there's a big mainstream with all the bells and whistles we already know from traditional media and that most people find their satisfaction in following this. The tiny rest remains unread - like this blog. ;-) But is this really surprising to you - presumed you're not a Web 2.0 evangelist? (Source)

I'm really curious about this. A SPAM detection that is inspired by the mechanisms of the immune system would indeed be something new. More about this at the website of the international Alife conference and here at this place, as soon as I get more information.

I don't know anything about the formulas being solved there. But the evolutionary computation of a a century-old algebra problem done by researchers from Hampshire College in Massachusetts and the State University of New York proves its strength. The user is allowed to play God a little bit by selecting the elements of computation to use and by determining the measurement of desired designs. Starting with random combinations of elements, each computation step generates offsprings that are "recombined" and that may "mutate", then evaluated (thus simulating some kind of 'fitness'). The solution emerges during this artificial evolution process. The aforementioned researchers, Hampshire’s Lee Spector (genetic programming), and SUNY-New Paltz’s David M. Clark, along with Spector’s students Ian Lindsay, Bradford Barr and Jon Klein, won the 2008 Genetic and Evolutionary Computation Conference’s top prize. (Source)
Learn about the basics of Genetic Programming.

This is real in-depth research: researchers revive the study of the development of roots at molecular, cellular, and organ levels. Since the research group consists of mathematicians, engineers, and computer scientists it will be interesting to see which results will get unearthed. (Source)

An interesting study on measuring bloggers' stickiness to their topics, introducing a new coherence score as a means of measurement. How noisy is your blog? (Link may require payment)