Years ago, we did a study to determine whether anyone at Google is particularly good at hiring. We looked at tens of thousands of interviews, and everyone who had done the interviews and what they scored the candidate, and how that person ultimately performed in their job. We found zero relationship. It’s a complete random mess
Group items matching
in title, tags, annotations or url
3More
5More
Space Oddity - YouTube - 4 views
-
- ...2 more comments...
-
-
Coming next: Dancing bear jumps through burning hoop! ... on Asteroid!!! :-P But seriously - Chris Hadfield did an amazing job in getting ordinary Earthlings interested in space. His educational videos can be found here: http://www.youtube.com/playlist?list=PLUaartJaon3LV-ZQ4J3bNQj4VNVG2ByIG
-
and to poison the waters of an amazing performance, here's the relevant(?) copyright law: http://www.economist.com/blogs/economist-explains/2013/05/economist-explains-12?fsrc=scn/tw_ec/how_does_copyright_work_in_space_
-
And in case you wonder, this is *not* the most expensive music video ever made. Also, launching his guitar to the orbit was still far cheaper than the cost of some guitars sold on earth. Where else can this info come from if not http://what-if.xkcd.com/45/
9More
"Natural Light Cloaking for Aquatic and Terrestrial Creatures" - 3 views
-
Cheap and scalable invisibility cloaks being developed. The setup is so trivial that I would almost call it a "trick" (as in "Magicians trick"): 6 prisms of n=1.78 glass. Nontheless, it does the job of cloaking an object at visible wavelengths and from several directions.
- ...6 more comments...
-
-
-
New video here (smaller file than previous): "https://dl.dropboxusercontent.com/u/58527156/20130613_101701.mp4" Note how close to the center of the field of view the hidden objects are. I am quite surprised that such poor lenses create such a sharp focus.
-
Well.. I would say that it is not "fully cloaking", as the image behind is mirrored as well
-
That just means that you have to double the setup, i.e., put 4 glasses in a row. Of course the obvious drawback is that you can only look at this cloak from one direction.
-
and here is the same for bigger things http://cosmiclog.nbcnews.com/_news/2013/06/07/18831348-this-invisibility-cloak-could-conceal-satellites-or-hide-your-kids?lite
-
Is this really new? I don't know, but I know that the original idea of cloaking was pretty different. When cloaking as an application of transformation optics became popular people tried to make devices that work for any incidence angle, any polarization and in full wave optics (not just ray approximation). This is really hard to achieve and I guess that the people that tried to make such devices knew exactly that the task becomes almost trivial by dropping at least two of the three conditions above.
-
I think it is very easy to call something trivial when you're not the one who invested considerable time (5 min in my case) to design a cloaking device and fill the coffee mugs with water... Also, I did not really violate that many conditions: true I reduced the number of dimensions in which the device works to 1 (as opposed to the 2 dimensions of many metamaterial cloaks). However the polarization should not be affected in my setup as well as the wave phase and wave vector (so it works in full wave optics) - apart maybe from the imperfect lens distortion, but hey I was improvising.
1More
In Head-Hunting, Big Data May Not Be Such a Big Deal - NYTimes.com - 1 views
3More
Poison Attacks Against Machine Learning - Slashdot - 1 views
-
Support Vector Machines (SVMs) are fairly simple but powerful machine learning systems. They learn from data and are usually trained before being deployed.
-
In many cases they need to continue to learn as they do the job and this raised the possibility of feeding it with data that causes it to make bad decisions. Three researchers have recently demonstrated how to do this with the minimum poisoned data to maximum effect. What they discovered is that their method was capable of having a surprisingly large impact on the performance of the SVMs tested. They also point out that it could be possible to direct the induced errors so as to produce particular types of error.
-
http://arxiv.org/abs/1206.6389v2 for Guido; an interesting example of "takeover" research
3More
They Finally Tested The 'Prisoner's Dilemma' On Actual Prisoners - And The Results Were... - 2 views
-
-
For me it is not a surprising result at all that criminals cooperate more than students. The former are partners in crime, united by being "against the system". The latter are nothing more than competitors of each other at the job market. On a side note, the majority of the students doesn't *know*, they *memorise*. There is a difference. I recommend "Celda 211". Good movie on the topic.
-
Did the theoretical payoff matrix take into account what happens when you next encounter the person you cheated out of their coffee/cigarettes?
2More
Voyager I has officially entered interstellar space - 2 views
5More
Physicists Discover Geometry Underlying Particle Physics | Simons Foundation - 7 views
-
- ...2 more comments...
-
-
Definitely elegant.. although if true, it does put two of my closest friends out of a job...
-
Sounds like a great tool for calculation and may provide some deeper understanding. But: I think their comments about space and time are misleading. Often you can ignore space and time when you just want the probability of an event (and it makes your calculations easier) but especially in the low-energy regime an event is clearly localized.
-
2More
The Truth About Google X: An Exclusive Look Behind The Secretive Lab's Closed Doors - 4 views
-
Space elevators, teleportation, hoverboards, and driverless cars: The top-secret Google X innovation lab opens up about what it does--and how it thinks.
-
Interesting insight indeed, I see quite some overlap with the ACT mantra, athough they have 250 people and an outdoor playground.. To Teller, this failure-loving lab has simply stepped into the breach. Small companies don't feel they have the resources to take moonshots. Big companies think it'll rattle shareholders. Government leaders believe there's not enough money, or that Congress will characterize a misstep or failure as a scandal. These days, when it comes to Hail Mary innovation, "Everyone thinks it's somebody's else's job," Teller says.
17More
Massively collaborative mathematics : Article : Nature - 28 views
-
- ...14 more comments...
-
-
"Similar open-source techniques could be applied in fields such as [...] computer science, where the raw materials are informational and can be freely shared online." ... or we could reach the point, unthinkable only few years ago, of being able to exchange text messages in almost real time! OMG, think of the possibilities! Seriously, does the author even browse the internet?
-
I do not agree with you F., you are citing out of context! Sharing messages does not make a collaboration, nor does a forum, .... You need a set of rules and a common objective. This is clearly observable in "some team", where these rules are lacking, making team work inexistent. The additional difficulties here are that it involves people that are almost strangers to each other, and the immateriality of the project. The support they are using (web, wiki) is only secondary. What they achieved is remarkable, disregarding the subject!
-
I think we will just have to agree to disagree then :) Open source developers have been organizing themselves with emails since the early '90s, and most projects (e.g., the Linux kernel) still do not use anything else today. The Linux kernel mailing list gets around 400 messages per day, and they are managing just fine to scale as the number of contributors increases. I agree that what they achieved is remarkable, but it is more for "what" they achieved than "how". What they did does not remotely qualify as "massively" collaborative: again, many open source projects are managed collaboratively by thousands of people, and many of them are in the multi-million lines of code range. My personal opinion of why in the scientific world these open models are having so many difficulties is that the scientific community today is (globally, of course there are many exceptions) a closed, mostly conservative circle of people who are scared of changes. There is also the fact that the barrier of entry in a scientific community is very high, but I think that this should merely scale down the number of people involved and not change the community "qualitatively". I do not think that many research activities are so much more difficult than, e.g., writing an O(1) scheduler for an Operating System or writing a new balancing tree algorithm for efficiently storing files on a filesystem. Then there is the whole issue of scientific publishing, which, in its current form, is nothing more than a racket. No wonder traditional journals are scared to death by these open-science movements.
-
here we go ... nice controversy! but maybe too many things mixed up together - open science journals vs traditional journals, conservatism of science community wrt programmers (to me one of the reasons for this might be the average age of both groups, which is probably more than 10 years apart ...) and then using emailing wrt other collaboration tools .... .... will have to look at the paper now more carefully ... (I am surprised to see no comment from José or Marek here :-)
-
My point about your initial comment is that it is simplistic to infer that emails imply collaborative work. You actually use the word "organize", what does it mean indeed. In the case of Linux, what makes the project work is the rules they set and the management style (hierachy, meritocracy, review). Mailing is just a coordination mean. In collaborations and team work, it is about rules, not only about the technology you use to potentially collaborate. Otherwise, all projects would be successful, and we would noy learn management at school! They did not write they managed the colloboration exclusively because of wikipedia and emails (or other 2.0 technology)! You are missing the part that makes it successful and remarkable as a project. On his blog the guy put a list of 12 rules for this project. None are related to emails, wikipedia, forums ... because that would be lame and your comment would make sense. Following your argumentation, the tools would be sufficient for collaboration. In the ACT, we have plenty of tools, but no team work. QED
-
the question on the ACT team work is one that is coming back continuously and it always so far has boiled down to the question of how much there need and should be a team project to which everybody inthe team contributes in his / her way or how much we should leave smaller, flexible teams within the team form and progress, more following a bottom-up initiative than imposing one from top-down. At this very moment, there are at least 4 to 5 teams with their own tools and mechanisms which are active and operating within the team. - but hey, if there is a real will for one larger project of the team to which all or most members want to contribute, lets go for it .... but in my view, it should be on a convince rather than oblige basis ...
-
It is, though, indicative that some of the team member do not see all the collaboration and team work happening around them. We always leave the small and agile sub-teams to form and organize themselves spontaneously, but clearly this method leaves out some people (be it for their own personal attitude or be it for pure chance) For those cases which we could think to provide the possibility to participate in an alternative, more structured, team work where we actually manage the hierachy, meritocracy and perform the project review (to use Joris words).
-
I am, and was, involved in "collaboration" but I can say from experience that we are mostly a sum of individuals. In the end, it is always one or two individuals doing the job, and other waiting. Sometimes even, some people don't do what they are supposed to do, so nothing happens ... this could not be defined as team work. Don't get me wrong, this is the dynamic of the team and I am OK with it ... in the end it is less work for me :) team = 3 members or more. I am personally not looking for a 15 member team work, and it is not what I meant. Anyway, this is not exactly the subject of the paper.
-
My opinion about this is that a research team, like the ACT, is a group of _people_ and not only brains. What I mean is that people have feelings, hate, anger, envy, sympathy, love, etc about the others. Unfortunately(?), this could lead to situations, where, in theory, a group of brains could work together, but not the same group of people. As far as I am concerned, this happened many times during my ACT period. And this is happening now with me in Delft, where I have the chance to be in an even more international group than the ACT. I do efficient collaborations with those people who are "close" to me not only in scientific interest, but also in some private sense. And I have people around me who have interesting topics and they might need my help and knowledge, but somehow, it just does not work. Simply lack of sympathy. You know what I mean, don't you? About the article: there is nothing new, indeed. However, why it worked: only brains and not the people worked together on a very specific problem. Plus maybe they were motivated by the idea of e-collaboration. No revolution.
-
Joris, maybe I made myself not clear enough, but my point was only tangentially related to the tools. Indeed, it is the original article mention of "development of new online tools" which prompted my reply about emails. Let me try to say it more clearly: my point is that what they accomplished is nothing new methodologically (i.e., online collaboration of a loosely knit group of people), it is something that has been done countless times before. Do you think that now that it is mathematicians who are doing it makes it somehow special or different? Personally, I don't. You should come over to some mailing lists of mathematical open-source software (e.g., SAGE, Pari, ...), there's plenty of online collaborative research going on there :) I also disagree that, as you say, "in the case of Linux, what makes the project work is the rules they set and the management style (hierachy, meritocracy, review)". First of all I think the main engine of any collaboration like this is the objective, i.e., wanting to get something done. Rules emerge from self-organization later on, and they may be completely different from project to project, ranging from almost anarchy to BDFL (benevolent dictator for life) style. Given this kind of variety that can be observed in open-source projects today, I am very skeptical that any kind of management rule can be said to be universal (and I am pretty sure that the overwhelming majority of project organizers never went to any "management school"). Then there is the social aspect that Tamas mentions above. From my personal experience, communities that put technical merit above everything else tend to remain very small and generally become irrelevant. The ability to work and collaborate with others is the main asset the a participant of a community can bring. I've seen many times on the Linux kernel mailing list contributions deemed "technically superior" being disregarded and not considered for inclusion in the kernel because it was clear that
-
hey, just catched up the discussion. For me what is very new is mainly the framework where this collaborative (open) work is applied. I haven't seen this kind of working openly in any other field of academic research (except for the Boinc type project which are very different, because relying on non specialists for the work to be done). This raise several problems, and mainly the one of the credit, which has not really been solved as I read in the wiki (is an article is written, who writes it, what are the names on the paper). They chose to refer to the project, and not to the individual researchers, as a temporary solution... It is not so surprising for me that this type of work has been first done in the domain of mathematics. Perhaps I have an ideal view of this community but it seems that the result obtained is more important than who obtained it... In many areas of research this is not the case, and one reason is how the research is financed. To obtain money you need to have (scientific) credit, and to have credit you need to have papers with your name on it... so this model of research does not fit in my opinion with the way research is governed. Anyway we had a discussion on the Ariadnet on how to use it, and one idea was to do this kind of collaborative research; idea that was quickly abandoned...
-
I don't really see much the problem with giving credit. It is not the first time a group of researchers collectively take credit for a result under a group umbrella, e.g., see Nicolas Bourbaki: http://en.wikipedia.org/wiki/Bourbaki Again, if the research process is completely transparent and publicly accessible there's no way to fake contributions or to give undue credit, and one could cite without problems a group paper in his/her CV, research grant application, etc.
-
Well my point was more that it could be a problem with how the actual system works. Let say you want a grant or a position, then the jury will count the number of papers with you as a first author, and the other papers (at least in France)... and look at the impact factor of these journals. Then you would have to set up a rule for classifying the authors (endless and pointless discussions), and give an impact factor to the group...?
-
-
urgently!! btw: we will have the ACT christmas dinner on the 9th in the evening ... are you coming?
2More
Need to analyse vast amounts of data? Listen to it! (NASA) - 2 views
2More
Cut EU red tape: Report from the Business Taskforce - 2 views
-
some interesting recommendations part of a current discussion between the commission and some member states in a general, and apparently popular with all parties, drive to reduce bureaucracy to, among other things, stimulate innovation. Firms face a challenge. They produce superb products, offer world-class services and benefit from being able to sell to a European market of 500 million customers. But they are often encumbered by problematic, poorly-understood and burdensome European rules. The impact is clear: fewer inventions are patented, fewer sales are made, fewer goods are produced and fewer jobs are created.
-
what hypocrites ... "As businesspeople, we are convinced that these and many other problems must be addressed if British and European firms are to compete in the global marketplace. We need regulation to operate in a pan-European market. We are not against regulation per se. But we need regulation that is pro-growth and pro-innovation." sure ....
1More
Dream job for Tobias and the like? - 2 views
7More
Meet The Man Who Paid A Record $335,000 For Virtual Property - Oliver Chiang - SelectSt... - 7 views
-
- ...4 more comments...
-
"He says he made the purchase partly because he wants to be able to spend more time in the virtual world. Before, he was averaging 10 to 20 hours per week. He wants to be able to spend about 40 to 60 hours a week now, basically making running the virtual asteroid a full-time job. (He'll also be cutting back on the time he spends developing software in real life.)"
-
From what I remember when I visited the developer/producer company HQ, he wouldn't have to pay any taxes. If he has a virtual business he might have to pay them a license fee. If you want to start a virtual bank, you would need to buy a banking license. The money thing is quite regulated in this enviroment, so probably that's why property prices can be quite high.
-
-
Remember the study but have completely zapped that this was with this company ... GSP rules :-)
-
so how does that state get his money from this type of economy? where is the VAT in there?
-
Last time I checked the "state" was still loosing money. But their main income is the sale of resources. Mostly new land, but I believe at some point they wanted to sell their initial planet too.
3More
Why Can't PCs Work More Like iPhones? - Bits Blog - NYTimes.com - 1 views
-
-
I would love to see people working on touchscreens all day, orthopedics would have a field day :) Anyway, the answer to the original question is: because the JesusPhone is an appliance, a PC is not.
-
Luckily there is open source, so neither Steve Jobs nor NY Times decide for us what sort of OS we have to like! I'll join the "Jihad for the command line" troop!!
5More
Higgs hunters face long haul - 2 views
-
to reduce the chances of the LHC being derailed again by a similar accident, physicists at the Geneva lab have decided to run the collider at just half its design energy for the next 18-24 months.
-
Once the 7 TeV run is over, CERN will shut the LHC down in 2012 for a year or more to prepare it to go straight to maximum-energy 14 TeV collisions in 2013. This will be a complex job that will involve replacing some 10,000 superconducting magnet connections with more robust ones.
-
choosing to stay at lower energies is a big price to pay in terms of the Higgs search. "We will need more than twice the data at 7 TeV compared to that needed at 10 TeV to reach the same discovery potential," she says. "At this energy we can at best expect to exclude a Higgs with a mass between 155 and 175 GeV."
-
no Higgs boson before 2013... and a replacement of 10,000 superconducting magnet connections ! Reminds me of the the gravitational detectors... no detection before an upgrade in 2013...! There are the big announcements to make the cash flow... and reality !
-
Higgs is almost 81, so he should better invest in his health if he wants the Nobel prize... But who cares, it's another 5 years window where high-energy theorists can produce nonsense with no experimental evidence. They should be happy!
3More
Robotic roach creates order from chaos : Nature News - 8 views
7More
mentored by the Advanced Concepts Team for Google Summer of Code 2010 - 4 views
-
you propably already know,I post it for the twitter account and for your comments
- ...4 more comments...
-
once again one of these initiatives that came up from a situation and that would never have been possible with a top-down approach .... fantastic! and as Dario said: we are apparently where NASA still has to go with this :-)
-
Actually, NASA Ames did that already within the NASA Open Source Agreement in 2008 for a V&V software!
-
-
indeed ... you are right .... interesting project btw - they started in 1999, were in 2005 the first NASA project on Sourceforge and won several awards .... then this entry why they did not participate last year: "05/01/09: Skipping this years Google Summer-of-Code - many of you have asked why we are not participating in this years Summer of Code. The answer is that both John and Peter were too busy with other assignments to set this up in time. We will be back in 2010. At least we were able to compensate with a limited number of NASA internships to continue some of last years projects." .... but I could not find them in this years selected list - any clue?
-
but in any case, according to the apple guru, Java is a dying technology, so their project might as well ...
-
They participate under the name "The Java Pathfinder Team" (http://babelfish.arc.nasa.gov/trac/jpf/wiki/events/soc2010). It is actually a very useful project for both education and industry (Airbus created a consortium on model checking soft, and there is a lot of research on it) As far as I know, TAS had some plans of using Java onboard spacecrafts, 2 years ago. Not sure the industry is really sensible about Jobs' opinions ;) particularly if there is no better alternative!
4More
Girls Get Math: It's Culture That's Skewed - 2 views
-
"There's a gender stereotype that boys are better at math than girls are, and stereotypes die very hard," Hyde told LiveScience. "Teachers and parents still believe that boys are better at math than girls are." The researchers provide several possible cultural factors keeping females from excelling in math, including classroom dynamics in which teachers pay more attention to boys, while even mathematically gifted girls are not nurtured. In addition, stereotypes may drive guidance counselors and others to discourage girls from taking engineering courses.
-
The guidance counselor at my high school discouraged me to study physics but was very excited when I was contemplating to become a teacher. Maybe I should send her this article...
-
Oh yeah, real new!! And in the 90s it was obvious that girls are smarter but discriminated, today its obvious that the poor boys are neglected; some years ago female teachers were proven to discriminate even stronger against girls than male teachers and today politicians demand more male teachers... because they would pay more attention to the neglected boys! Great, that's what I like about sociological research, every couple of years one can sell the same old story again and again and again... sorry, I'm in a real bullshitter mood today!
-
gotta agree with luzi here. girls at secondary school in the UK have been outperforming boys for years now after numerous government programmes. As to guidance counsellors - if they were any good at guidance wouldn't they have better jobs?