Come think with us.
Since we’re blogging from the main hypios site, you can read our thoughts, leave comments, browse problems, and connect with Solvers on the hypios network, all in one place.
Put on your thinking caps and update your bookmarks. We’ll see you there!
Pensive gorilla from ben pollard‘s Flickr.
Semantics and the future of search
In Part I of our two-part “Thinking about the future of search” series, we discussed the social recommendation side of the future of search. Here, we’ll explore the possibilities of semantic search.
At hypios, we’re really excited about the semantic web—so excited that we can’t stop talking about it. We think that data will eventually be structured, machine-readable, and linked, vastly improving search. Engines will return better (more relevant, more specific) results in an easier-to-read format.
Meaning and context
Search depends on meaning, and meaning depends on context. Getting good search results depends a lot on our ability to define our terms and specify a certain meaning. Right now, we have to put in more keywords and use a common vocabulary; in the future, we might see: Read the rest of this entry »
A few weeks ago, while writing about Dan Pink’s great talk, we introduced a puzzle…and, well, we’re still thinking about it. Of course we’re not alone; the link between competition and innovation keeps a lot of economists (and organizational psychologists) busy—especially since innovation and competition don’t seem to go hand-in-hand in the way that many would expect.
In Pink’s talk, for instance, he discusses the surprising finding that creative thinking seems to be allergic to pay-for-performance type incentives (and by extension, we argued, competition).
This struck us as intuitively right. Who wants to be distracted by the prospect of failure or ranking when solving a challenging problem? Think about the way kids are absorbed in play. Think about yourself playing Legos as a kid. Now imagine your performance at Legos if your mom offered to pay for the best construction after 45 minutes of “play.” The pleasure of total absorption in building something would have been lost.
It turns out that the dominant investment strategies for firms display a similiar aversion to competition. There is an observed negative impact on innovation in highly competitive fields, since investments seem more risky in these contexts (this effect is mitigated slightly in the case where two firms are running “neck and neck”). This means fewer firms enter a crowded field, and overall investments (viz., innovation) tend to be lower.
How many piano teachers are there?
I started taking piano lessons at age five. My parents tried everything to get me to practice, especially appealing to my pride. ”How many people can play the piano?” my father would ask. ”You’ll be glad you learned as a kid! How many adults are there who started taking lessons and wish they had continued?” He would then start jotting down figures—how many piano teachers in our town, how many students each teacher had, how many would probably quit once they went to college—and come up with an estimate that proved I would grow up to be part of a very special and elite group of pianists.
My dad has long since given up on me as a pianist, but his questions still plague a different select group: search developers. The future of search may lie in the answers. In this two-part series, I’ll examine how search engines might answer them through social and semantic improvements.
Xavier Gréhant, a PhD student working in distributed computing, solved the first problem broadcast on hypios.com. He developed what is probably the first problem-suggestion algorithm (or engine) out there, and won $3,000 for his solution. Here’s the story:
Join Xavier on hypios: become a solver!
Didn’t we just post about Twitter? Well, yes, but as a philosopher, I can’t ignore what’s happening. (Pun intended.)
I’ve always preferred Facebook’s “What’s on your mind?” to Twitter’s “What are you doing?” Here’s why:
Twitter seemed to think all you do is important. Facebook’s question seems designed to keep you from being too prosaic. Do you really want people thinking your overcooked spaghetti is all that’s on your mind? Even if you occasionally talk about your dinner (and breakfast and lunch and snack and…) at least the question assumes you have interesting thoughts to share. So on this first level, Facebook seems to be for philosophers whereas Twitter seemed to think of you as an ego-driven stupid. But users weren’t. They shared much more on Twitter than what they were doing. And if they hadn’t you wouldn’t even know that the site exists.
Today, Twitter changed: “What are you doing?” has been replaced by “What’s happening?”. Well, Twitter is growing up—it’s no longer all about the ego, like a small child, it starts to be curious about the world around itself.
Of course, as the Twitter blog says, this isn’t going to change the way anyone uses the service; it’s quite the opposite. Like in the jeopardy quizz-game, where you have the answer and need to find the question, Twitter’s team got together, looked at all the answers for a few months and then they came up with the question people already seemed to answer. That is, the Twitter team starts to take into account how the users use what they have created. The new prompt reflects the actual use of the site, rather than its intended use. It never was the micro-blogging site they intended it to be, where people only told their friends what they were doing. It at once was a broader network where you could learn what’s going on, from election protests to planes landing in rivers.
Twitter may have changed the question, but the answer are still the same (and yet, always changing).
But…Twitter isn’t that good
Sure, Twitter is a good product. The application interface is accessible and wide open. It’s a real-time social bookmarking tool that’s original in terms of virality and network recommendation. This stream fits with some people’s needs to organize content. But honestly, it’s not that good. Following too many people (something that we, @hypios, admittedly do, as we follow pretty much everyone who follows us) makes the stream change so quickly that it’s impossible to really ‘follow.’ Follow fewer people, though, and it gets boring; you’ll just see the same excessive twitterers on your network.
Then there’s the functionality. I won’t go too far into that, but it’s maddening that there’s no ‘select all and delete’ option for the 200 daily automated direct messages you’ll get (which are not at all distinguished, by the way, from the real ones). You have to click on every single one of them. “Delete all” a feature that you have with every simple email account. I’ll stop here, but to reiterate, it’s not that good.
But then why is everyone talking about it?
One billion search results in Google: five times more than “Obama.” Impressive, right? But let’s have a closer look at these results. The first page: Twitter.com, About page, Wikipedia article, Crunchable profile… But when you go deeper into the results, you find that at least one out of every two results is a ‘how to’: how to get more followers, how to make a profit from Twitter; or a who’s who: who has more followers, who is more popular? This only confirms the impression you get when reading tech news about Twitter: a huge part of the media coverage is about Twitter’s popularity, marketing and social capital. Searchable interest in Twitter is predominantly about marketing!
Telling the world what you’re having for lunch might be amusing for a while. But in general, Facebook is better for that—you choose your friends, and you have more control over comments and interactions. Most active Twitter users do care about what they say. But they care more about the world (their followers) caring about what they say.
Twitter as a marketing tool is extremely advanced because it’s distributed and partially automatized (via applications). I sometimes envision of a sort of endtime’ scenario: a point where Twitter would become a place where machines tweet and retweet to other machines, where machines welcome other machines by direct messages, where there is only occasionally (if any) human attention to reprogram the way in which the machines interact and alter the content the machines endlessly send to each other.
What’s a Ponzi scheme?
Roughly: an investment scheme where the first investors get dividends, that is, they are paid back for their investment, directly from the capital of new investors. So a Ponzi scheme investment fund never generates any real revenue: it merely redistrbutes investments. To make sure that early investors have the feeling of earning money with their investment, there must always be new investors and more new investors than old ones: it’s a pyramid. Bernard Madoff recently became infamous through applying this scheme for many years until his pyramid collapsed, when too many investors wanted to have their money back.
Attention is Twitter’s capital
On a certain level of Twitter activity (and probably the one where most traffic happens), Twitter works in just this way. The capital used on Twitter is not money, but the attention of new users. Marketeers go crazy about Twitter because, especially on the internet, companies tend to associate attention—probably correctly—with potential revenue. If you pay attention to a link by clicking on it, you might also buy the product it’s promoting.
Of course, Twitter users are not investing real money, or at least, they’re not paying Twitter directly. (Even if Twitter itself still hasn’t earned a single dollar, the ecosystem around it is making money). The Securities and Exchange Commission probably won’t need to investigate Twitter’s offices after reading this post. And even if it consumes some time, and time is money, using an application to tweet doesn’t take that much time.
But Twitter is not “free,” at least from a macro point of view: the usual things sold by Twitter marketing software, “reach thousands of new customers for free” glosses over the fact that, even though Twitter lets users broadcast a message to their followers, their followers aren’t necessarily listening. And listening, in the Twitter marketing system, is an investment: an investment where dividends are not paid to the investor, but to the marketers, the people you listen to. It starts to resemble a Ponzi scheme. When dividends are paid with the money of other investors, we’re getting close to Madoff’s territory. Just…there is no “puppet-master”, like Madoff, steering investors towards the riff, like Lemmings. Twitter as a marketing device is just something that naturally evolved towards becoming a Ponzy.
The Twitter user cycle
P.S. This post was intended to be somewhat provocative and ironic. Indeed, it only applies to a certain use of Twitter, its use for marketing. There might be quite some ways to avoid the collapse: if Twitter changes its strategy, better applications are developed or, well, if Twitter generates some income. And well, maybe we wouldn’t care too much if Twitter turned out not to be such a good marketing tool after all and if the Twitter-as-marketing-tool-pyramid collapsed. What do you think?
The winner of a competition in problem-solving talks about his experience and his approach to problem-solving. Xavier Gréhant, working on a PhD in distributed computing solved the first problem that was broadcast on hypios.com: He developped what is probably the first problem-suggestion algorithm (or engine) out there.
In the last few months, most of our blog posts have been about issues surrounding Open Innovation, Crowdsourcing, web-communities, some were turned towards “pure” science, interesting facts, issues and figures. But the death of Claude Levi-Strauss (on October 30) reminded some of us of our backgrounds in disciplines like social science, philosophy and history… For most of us, Levi-Strauss was not just a familiar name, but an intellectual figure that truly influenced the way we think. And that brings me to my main point for this post: in what respect is Levi-Strauss’ thinking important (even if you’re not an anthropologist?)
The figure of “structuralism”
Levi-Strauss is the most prominent name associated with what is commonly called “structuralist” thinking. Structuralism is a theoretical approach that was born with Ferdinand de Saussure’s pioneering work in linguistics. Its general principle is to grasp an entity (such as language, in the case of Saussure), as a system of elements (rather than as a set of atomistic elements). To think in terms of the system means that each element can only be defined in terms of the relations (of equivalence or opposition) it has with other elements of the system. Take language, for instance. Saussure’s approach showed that meaning in language can be best explained not by assuming the “identity” or pre-existing “content” of the linguistic signs it is composed of (which, added together, make up a language), but by assuming the priority of the system, and the relations between the combined elements. In this case, meaning has to be thought of oppositionally (as the result of a contrast between opposing pairs) and differences between terms assume prominence. Saussure goes even further: we do not choose, when we speak, to combine certain elements of language. In language it is always a matter of pre-existing possibilities, combinations that form language itself (the structure) command social agents when they are speaking. The structure commands the agents, which is a very strong thesis.
In this view, structuralism is a fairly deterministic theory. From it’s origin in linguistic theory, Levi-Strauss’ theoretical & field work applied the structuralist framework to the problem of understanding how societies, and, in particular, non-western societies (like the South-American Bororos) worked. Structuralism proved an especially powerful tool, allowing Levi-Strauss to explain the meaning and evolution of social entities and the social system better than competing models used in anthropology. Other explanations were either historical (basically explaining the existence of a social entity by giving an account of how it appeared) or functionalist (a tradition represented by Boris Malinowski and Franz Boas that analyses social forms through the function they fulfill within society, e.g. marriage is for stable reproduction); both are tautological. That is, they don’t really offer explanations but just repeat our still unexplained preconceptions. Levi-Strauss opposes to these two theories an analysis that’s mainly relational: “examining the logical structures that underlay relationships rather than their contents” is the mission of modern anthropology as he sees it.
De-Naturalization
Applying his method allowed Levi-Strauss to make powerful claims about social reality; in particular, he is widely known for his analysis of kinship. The claimed universality of the functionalist idea of family – father, mother, children – was directly undermined by Levi-Strauss’ analysis (as ethnocentrist and complacent), which showed that in some societies, for example, the figure of the uncle was more prominent than that of the father. Later work in theories of gender attacked the purported “naturalness” and universality of men’s domination in families, picking up on Levi-Strauss’ work in this field.
In which sense is the “structuralist” analysis of kinship novel? Here’s a famous example, developed in Structural Anthropology: let’s call “A” the relation between uncle and nephew; “B”, the relation between brother and sister; “C” the relation between father and son, and “D” the relation between husband and wife. Levi-Strauss discovers the following suprising fact: in every society, A is to B what C is to D. That is, Levi-Strauss shows that while the individual elements vary in their relations with each other, there is a certain invariability at the system-level. This conclusion is the result of the “structural investigation”: “the search for the underlying patterns of thought in all forms of human activity.”
A “handbook” against racism
Levi-Strauss’ work on “savage” societies, decomposing myths, symbols or social events that might appear totally irrational to us, made these phenomena more understandable. Decomposing the elements, he showed how a relatively small repertoire of meaningful entities are combined (each time differently) in myths, rituals, etc in different cultures and historical epochs. According to reknowned sociologist Pierre Bourdieu, this analysis of fundamental units (say characters or plot elements) rather than the content, showed the underlying rationality of often stigmatized rituals, religious practices, etc… It was now harder to judge things that seemed at first glance strange or unfamiliar as absurd or barbaric. Structuralist analysis is a powerful lens, allowing the social scientist the possibility to see meaning as a construction out of universal elements–flattening the perceived differences between cultural productions. It’s this aspect of Levi-Strauss’ work that I take to be most significant and most durable: his books are, as Bourdieu said, a “very powerful handbook against racism.”
What’s a “classic”?
When a famous intellectual dies, you can hear, among other intellectuals, politicians, etc… a sort of uniformly reverential tone: s/he was a “classic”, a prominent public intellectual, a national treasure, extremely influential, etc etc. Some of them, with lots of nostalgia, might also say: “he was the last giant” (I think in France we’ve lost twenty or thirty ” last giants” this year alone). But, with Levi-Strauss’ death at 101 years of age, no one can deny he loss of an important intellectual figure. What’s really clear in this case was Levi-Strauss wide influence outside of his field, anthropology–his influence, that is, on the larger culture. No one, among philosophers, sociologists, historians, and even mathematicians (structuralism brought a lot to cybernetics, for example), could ignore his name; and most of them know a bit about its work. I think here we have something that can be said to be an objective criterion for what it means to be a “classical” intellectual figure. Given increasing specialization, it will likely be harder and harder to find individuals who’ve affected the progress of disciplines other than their own.
We pay him homage.
We’re happy to announce that hypios has its first problem in sustainable development. One of our Seekers is looking for a biodegradable battery with the same power as a traditional battery. The need for this innovation emerged in a very specific environment (reflected in the problem’s specifications), so a potential solution will only have to work in this environment. But prospective use of this kind of device is not limited to the domain where the problem emerged.
We really hope that this problem will find a solution on hypios, and that there are more of its kind to come! It might seem like a tiny problem, but it reflects one of the biggest environmental issues of the contemporary world: the production of energy (and especially the immense source of energy represented by nuclear power plants) produces non-degradable and non-recycable waste. Perfecting the biodegradable battery is a small but crucial step towards a sustainable world.
Become a fan
