How many times has this happened to you? You’re reading an article online, and you come across a word or a reference you don’t know — or a topic or term you want to learn more about — but it’s not hyperlinked. So you copy the text, pasting it into Google. Your Google search leads you to a Wikipedia page, which leads you to another Wikipedia page, which leads you to a New York Times article from 1942, which leads you to a New York Times article from yesterday, and before you know it, you’ve lost track entirely of the article that started you on your info-finding interlude in the first place.

If you’re like me, the answer is: Many, many times. The flip side of the web’s status as the greatest repository of information the world has ever known is that its information can easily form a kind of black hole when it comes to user attention. Hyperlinks allow us — hey, encourage us — hey, almost force us — to flit about from site to site across the vast expanse of the web, indulging our curiosity at the cost of nothing more than a click and a bit of time. But when sites don’t present links where there really should be links (argh!), users are often put in the position of having to hack those links through the old cut-and-paste, leaving the source that initially sparked their curiosity, often never to return.

Apture, a kind of web-based context engine, refers to this friction-filled practice as “leaking.” And leaking, as its name suggests, isn’t ideal — for web users or publishers alike.

But! Apture has just launched a feature that could change that. “Hotspots,” which builds on Apture’s Highlights function, leverages the curiosity of the crowd to determine, quite literally, the web’s “missing links.” The feature measures and tracks the number of times users have highlighted certain text within Apture-enabled websites; and then — here’s the very, very cool part — automatically creates a cluster of contextual information to support that text, via Wikipedia, YouTube, maps, and more. It’s Apture’s in-line-but-also-expandable version of a hyperlink.

“Our mission here at Apture is to discover your sparks of curiosity and satisfy them with full learning experiences,” Tristan Harris, Apture’s CEO, wrote in a blog post announcing Hotspots. “With Apture Hotspots we’re taking that to the next level by making curiosity itself go ‘viral.’ Each time one person is curious about something, Apture Hotspots will spread that curiosity to thousands of others.”

Crowdsourcing curiosity: It’s an intriguing proposition. While Apture isn’t the first outfit to experiment with communal highlighting — Kindle’s “Popular Highlights” provides a similarly social reading experience — the fact that this is happening on the open web, enabled by Apture’s single line of JavaScript, is significant. Hotspots aren’t about highlighting text so much as moving beyond it: identifying the need for relevant contextual information that lives beyond the confines of text itself.

The bonus for publishers and users: The entire process takes place within a single webpage. And that’s important, particularly, from the publisher perspective, especially if engagement metrics like time-on-site stats play an increasingly important role in the business side of web content. When you consider the sites that have the highest overall engagement on the web — sites like Wikipedia, Facebook, and Twitter — one thing they share is the fact that “everything is interconnected” within them, Harris told me. Those sites can almost literally capture our attention because the elements that our attention might follow are, often if not always, housed under the same domain. Apture’s brand of in-line, on-page context, Harris says, leads to users spending two to three times longer on sites than they otherwise would — and with 1.7 times the pageviews. Hotspots will likely expand on that.

But Apture has a broader goal, as well. In some ways, Harris points out, the premise that Hotspots is building out — the ability of the community to tighten intellectual connections among discrete bits of information — is an old one. Vannevar Bush, in the early part of the past century, discussed the idea of a “human brain”; Kevin Kelly, more recently, has applied that notion to the workings of the web (“a neural net that can learn”). Both visions are based on the notion that links, both literal and figurative, are core units of collective intelligence. They’re not secondary to our “human brain”; they’re necessary features that allow it to function. As Harris puts it:

Imagine the web is one big global brain: every page is a neuron (it holds information), and every link is a synapse that connects pages together. Every day we use the web, we make it smarter. When we click a link, we’re thickening and strengthening the connections between two pages. And every time we create a link, we’re teaching the web an idea. And just like the brain, it’s not the pages that make the web powerful, it’s the connections.