Stay up to date with notifications from The Independent

Notifications can be managed in browser preferences.

Can the internet answer every question?

Does God exist? Should you buy a pet rat? Is sleeping with the boss a good idea? Life is full of difficult questions. But, says Nick Harding, the internet could have the answers

Wednesday 29 April 2009 00:00 BST
Comments

Every day we have to make hundreds of decisions, some are as simple as choosing what sandwich to eat or what pet to buy, others are more complicated and come loaded with moral consequences. Should I sleep with my boss? Should I believe in God? Should I call my mother? They're all tricky judgement calls to make. So it's no wonder people have sought to delegate decision-making and shirk their responsibilities since the dawn of time. We've asked the Gods, we've asked the tarot cards, we've asked the life coach and we've asked Russell Grant. The bottom line is that decision making is hard – so why not get someone else to do it for us? In the Seventies, as inLuke Rhinehart's novel The Dice Man, the favoured method to absolve responsibility was the dice, all you had to do was roll a double six and you'd be heading to Rio on the back of a moped. Times have changed and today, if we want to let to fate determine our actions, we can let technology decide.

Thanks to an array of evolving websites and internet-based communities, we are able to let the internet take the sting out of decision making. In a few months, a new decision-making site, Hunch.com, will launch (although you can sign up for a preview now). In a nutshell, Hunch aims to provide answers to problems, concerns or dilemmas, on hundreds of topics. Hunch's responses are based on the collective knowledge of the Hunch community. Users input information and the site software uses that content to make decisions. Content is narrowed down by a specially created algorithm to cater specifically for the likes and preferences of the user, based on a profile of that the user provides. The site is designed so that every time it's used, it learns something new. It aims to save its users 'strenuous cognitive labour'. If you believe the blurb, it's the democratisation of decision-making. It's founders theorise that collective decision-making is superior to individual decision making. Ask Hunch, "what should I have for breakfast?" and it asks you a series of questions and ascertains from the answer what cereal you are likely to prefer.

While other question-answering websites like WikiAnswers and Yahoo!Answers, rely on threads of answers and are the equivalent of screaming towers of Babel ringing with voices of a thousand bored, opinionated geeks, Hunch hopes to shine through the information murk with one true voice.

Hunch co-founder Caterina Fake explains: "Hunch asks you questions to find out what you're like and what you like. It creates a "taste profile" of you and people like you, which combine with topic-specific questions to deliver a hunch just for you. We'll get to know the user in the simplest possible way, then help them make decisions."

Currently, the site is being developed and has around 200 invited users – the idea is that the more people that use it, the better it will get. The algorithm Hunch uses as its mathematical framework was built by a team of MIT computer scientists with backgrounds in machine learning, but it is users who will ultimately determine the quality of content and decisions it provides.

Kelly Ford, Vice President of Marketing said: "Users tend to contribute knowledge to topics they know, and extract knowledge from topics they don't know as well. That give and take is really the whole point of Hunch: to let many people benefit from the knowledge and cognitive work of others, to help people make smart, concrete decisions."

Fake, one of the founders of image-sharing site Flickr, admits she is unsure of Hunch's ultimate potential. "I anticipate users will surprise us by doing things with the software we don't expect."

So what can we expect when Hunch opens its doors to the general web populace? It provides recommendations on topics like health, finance and "what kind of rock band will I like?". It even guides users through decisions like "Should I believe in God?" to an answer based on their individual beliefs and theological views. However, it's doubtful Gordon Brown will be consulting it when he plots UK economic policy. Despite the technology, Hunch feels like a self-diagnosis branch chart and, at the mercy of user input, it contains plenty of inanity. If you want to know whether you could fight a bear and win without putting your life in danger, Hunch has the answer. You can also find out if you are smarter than Sarah Palin (almost certainly) and discover what martial arts weapon is best for you.

While mankind waits for Hunch to shoulder responsibility for its decisions, there are other sites to turn to. Like WikiAnswers.com which, earlier this year, announced that its users had asked their ten millionth question. Unlike its sister site, Wikipaedia, WikiAnswers uses an online community to answer specific questions while monitors keep check on the quality of the information provided. In 2007 it was the fastest-growing site in the US. Is it any good at making decisions for you? That depends on whether you are comfortable letting the kind of person who asks "what is the body for?", "who is the weakest man alive?" and who admits "I recently committed a serious crime, what should I do?" to pronounce on your life. WikiAnswers is popular but it's no oracle, you can't help but feel that behind it there is an army of teenagers, cramming GCSE homework and hoping someone can tell them the names of the three types of tree that grow in the African rainforest.

While WikiAnswers employs people to adjudicate what happens on its pages, Yahoo! Answers is more of lottery for decision-makers, a place where people can toss their problems into cyberspace and let the world pass judgement. It is responsibility-shirking taken to its logical conclusion in the information age, a place where, at any one time, you can log on and ask an assortment of strangers to pontificate on your life or alternatively play puppet master with other people's dilemmas. Recent Yahoo gems include "what drink gives you the most powerful belches?" and "where do I find male models? I'd like to date one". Inane? Maybe. Fun? Possibly.

If Yahoo! Answers was wearing a T-shirt printed with the words "I'm with stupid", the arrow on the apparel would be pointing to Answerbag.com, a repository of stupidity, stocked by stupid people answering stupid questions. Hot topics here include "can you remember what you had for dinner two days ago?" and "if I defecated on my desk, how long would it take for the bacterial to cover the whole surface?". Place responsibility for your actions in the hands of answerbag.com users and be prepared for the consequences. In essence, the site is a social network with users asking each other questions to elicit conversation threads, like the deliciously ironic "what mindless activity keeps you from dwelling on your troubles?".

While all these sites claim a question solution function, SideTaker.com takes the web-based decision making process a stage further and gives users' actions moral validity. The login community sits in judgement on arguments and domestic disagreements. Rowing couples can post their sides of a specific argument on the site and users will adjudicate on who is right and who is wrong. You can also post individual gripes. Confess that your wife's inability to stack the dishwasher correctly riles you, and SideTaker users will tell you whether you are being unreasonable and what you should do to remedy the situation (for the record, according to the site you are petty and if it annoys you that much, show her how to fill it properly).

The one thing these sites have in common is a reliance on user input, and when it comes to their effectiveness as decision-making tools, that is their downfall. If you invite all-comers to proffer advice, much of what you get back is opinion.

As technology commentator Andrew Keen explains: "In the old days, if you wanted answers you'd sit around a table and ask people or go to the pub and talk to people. But we are increasingly fragmented so we are trying to reinvent community, for the most part unsuccessfully, on the internet. If you are looking at a site for guidance on whether or not you should believe in God, and that site has employed 1000 professors of theology to go through the branches of questioning, then it is credible. But using collective input is always going to be fallible because those giving input can always have their own agendas." Keen points to a future beyond web 2.0 where advanced artificial intelligence determines fate for is. "The next stage in the evolution of the web is predicted to be a concept called the semantic web, a networked artificial intelligence that knows us as much as we know it. Hunch is more web 2.0 than the semantic web. The most reliable content, in my opinion ,is still curated."

So until the semantic web evolves, it seems the dice is just as effective as the web if you really want to throw caution to the wind.

Join our commenting forum

Join thought-provoking conversations, follow other Independent readers and see their replies

Comments

Thank you for registering

Please refresh the page or navigate to another page on the site to be automatically logged inPlease refresh your browser to be logged in