The internet’s data mine isn’t bottomless; you actually hit an inpenetrable barrier fairly quick, and it’s something far older than the web. That kachink of your information shovel recoiling is due to the simple fact of language. You can translate keywords, of course, but you’re still always translating into some other non-universal language, and translating imperfectly. Looking at collective behavior across different world languages is then limited, locking up quality research into different “ljinguistic silos,” as put in a new study out in Nature’s Scientific Reports journal, courtesy of researchers at University College London. And there are thousands of them.What you can look at, however, is numbers — years, perhaps, as the new UCL paper does. The report looks at something called the future orientation index, which is basically whether or not a group of internet users is more or less interested in the future than the past. So, take a given year — they took 2008, 2009, and 2010 — and look at whether it’s more or less likely that users were searching for the following year or the prior year. In their experiment, the UCL researchers looked at this information across different countries and different GDPs. There’s little ambiguity about the results: poorer countries are far more interested in the past; richer countries, the future.It’s a simple, short study with a result that can be interpreted pretty much any way you’d like. But it’s a start at using search information — via Google Trends, in this case — to look at global issues. The explanations suggested are fairly duh but worth repeating: “Firstly, these findings may reflect international differences in attention to the future and the past, where a focus on the future supports economic success,” they write. So, the suggestion is that cultures that are more future-focused in the first place are more likely to succeed. Makes sense.The authors also state that “these findings may reflect international differences in the type of information sought online, perhaps due to economic influences on available Internet infrastructure.” Which makes just as much sense, really. As far as interpretations go, this is a blank slate. But a wicked interesting one. Like, imagine that this is a function of connection speed and internet capabilities and the amount of information a user is even capable of searching. In the hierarchy of search, what then would put the past ahead? Is there more correlation for past searches to current practical needs?Try this: list your past 20 searches. Divide them by “future” and “past.” Now divide those further simply by more and less practical searches. Practical might be paying a bill or doing tax information, searching for medical advice, finding weather information, etc. Less practical might be searches relating to entertainment and gaming and music etc. What do you find? Are the searches relating to the past more need-oriented? If not, what else can you come up with that explains this?An interesting side-note not mentioned in the study is that dates and numbers are quite universal, but so is something else: images. A true image search will be limited to higher GDP countries for a while yet — searching based on image inputs, that is — but one imagines a time when search — and the associated mining — isn’t limited by language. By way of a very loose example, check out the above image search results for “Great Britain 2013” and “Senegal 2013.” (I chose Senegal because, of the poorer countries, it has some fleeting semblance of stability. Also, these aren’t the first results, just the first meaningful result in the top five of each.) Interpret that. In any case, one imagines pictures as at least another potential crack in a frontier with vast implcations for interpreting the world, and predicting it.Reach this writer at michaelb@motherboard.tv.
Advertisement
Advertisement