Getting the Mix just right
The web is vast and, having been growing exponentially for 25 years, is likely to become unimaginably vaster if it continues on its present course. I sometimes imagine it as a representation in cyberspace of the physical universe, which may be expanding to the point where even light cannot travel fast enough to get from one part of it to another. In more whimsical moods, I think of the web as the product of the proverbial infinite number of monkeys with an infinite number of typewriters: given unlimited time they’ll inevitably produce the complete works of Shakespeare (and everybody else) without any directing intelligence. Of course, the literary gems will be buried in a mass of unintelligible gibberish.
In short, it’s a reasonable bet that whatever you’d like to read, view, listen to, share or comment on is out there somewhere — almost everything is — your only problem is to find it. Since its founding in 2001, StumbleUpon has been the web’s go-to discovery engine. Once you’ve signed up for it and specified your interests you click the button marked “Stumble” and you’re served up a page which has been “thumbed up” by Stumblers with similar interests, including the people you follow. You can then give either a thumbs up or a thumbs down to that page. And so StumbleUpon learns more about what you like and what you don’t and presumably its recommendations improve.
But StumbleUpon isn’t perfect. For one thing, the “interest” being catered to by the page you’ve just been served up may not be the particular one you’re in the mood for at the moment. (And, I suspect, many people are like Ghost World’s Seymour who doesn’t want to meet somebody who shares his interests: “I hate my interests!”) Also, experience suggests that it’s easier to thumb up a page which is undemanding and “snackable”, making such pages rise to the top. There’s a whole host of reasons why any discovery engine, however good, can always be improved upon.
Garrett Camp, cofounder of StumbleUpon, which is now owned by his company Expa, has decided to build “on the legacy of StumbleUpon” to make a better discovery engine. It’s called Mix and this is his description of how it works:
Often the best content for you has the overlap of multiple characteristics: it’s new, several friends like it, it’s related to your personal interests, and it’s trending among people with similar tastes. Mix will rank content based on all of these characteristics, to help identify the best content for you. … [S]ocial signals … won’t be the dominant factor in what you see. We will ensure you aren’t just seeing the latest or most popular content, but the stuff that we believe you will actually like the most.
Instead of a single “Stumble” button, Mix shows you a series of graphic links to pages. The first is “Top for You” and it contains just what you’d expect: Mix’s top recommendations for you personally. Beside that, across the top of the page, are similar links to each of the “interests” you’ve specified. Each of these links leads to a separate page of recommendations in that area of interest. In other words, you’re not just clicking a single button and hoping that you like the suggested page. Instead, you have a wide choice of pages which Mix has selected as likely to appeal to you.
I’ve been using Mix for a few months now and I’ve consistently been able to find pages that I wanted to read and that I enjoyed reading. I’m definitely impressed.
Garrett Camp believes that the web has entered a fourth stage, curated discovery (after directories/portals, search, and social). I hope he’s right about that. Algorithmic selection is no longer enough to turn up “the best photos, videos and pages”: an element of human intervention is required. This being the web, you won’t be surprised to find that the curators are also the consumers. When you’ve signed up for Mix, you can add pages you think deserve wider attention to Mixes, which are themed collections of graphic links to those pages. Some people create dozens of mixes on various themes.
In addition to the General mix that everybody starts with, I’ve created seven so far, most of those having to do with fiction, or books generally. If they match your interests, you might like to get started on Mix by following some (or all) of them. They are:
Short fiction on Medium [the stories I’ve recommended in my newsletter]
Short stories [not on Medium but elsewhere on the web]
Writers on writing
Fiction by me on Medium