The tenure of former Yahoo CEO Terry Semel, who resigned yesterday, will certainly leave a lasting impression. Prominently, Semel – a former Hollywood executive – presided over efforts to bring Big Tech and Big Media together, in part by developing an entertainment division in Santa Monica. Semel is replaced by Yahoo co-founder Jerry Yang, an all-around “tech guy” who will focus on reducing Google’s advantages in search and advertising.

So Yahoo may return to its roots, at least for now, leaving the classic Media-Tech divide unbridged. That’s fine with Michael Arrington of TechCrunch. “The valley will take over Hollywood,” he concluded. “Not the other way around.”

On the other side, Sony’s Howard Stringer also chimed in on the topic last week in the Financial Times, with some pungent words describing how media executives often think about Silicon Valley.

“You’re using our content to drive your new global companies and getting rich in the process,” he said. Tough words, but Stringer added that the hostility is starting to thaw. Now “we’re more inclined to go there and say, ‘Look, let’s work out a partnership, otherwise the combat will exhaust us all.’ ”

Why all the tension? The frequently-observed truth is that platforms – and not content creators – have been the real beneficiaries of the latest boom. Compare the expansion of YouTube and MySpace to that of NBC, and the difference is clear.

What is less clear is exactly how to find middle ground. At Yahoo, Semel and appointee Lloyd Braun made deals with “60 Minutes” and almost produced a reality TV show. Elsewhere MySpace briefly became a content provider with its (now quiet) MySpace Records. These moves, however, always lacked the simple power of a great idea. When people first saw Friendster, many asked, “Why didn’t we do that before?” No one had that response when Yahoo sent a reporter to Iraq.

By our lights, two companies – Digg and Reddit – do indeed offer salient examples of convergence. Instead of editors deciding which stories make the front page, Digg and Reddit confirmed that social software could enable users to do it themselves.

And this topic – selection – may well be the raw nerve. For decades, salaried individuals in media have enjoyed the rewarding job of choosing content for the rest of us. Presumably they’re not wild about ceding that task to websites driven by bored cubicle-dwellers working for free.

For their part, media executives would be right in saying that such websites are not ready for primetime. Media companies produce sit-coms, novels, and music – things meant to bring forth elusive, powerful emotional reactions. Disney is not going to cede its decisions to online voting at Digg. And if merely analyzing Google hits were enough to discover winning content, then “Snakes on a Plane” would be last year’s highest-grossing film – right next to a movie based on Lonelygirl15.

But what if we ran media companies with more democratic and more powerful tools – like a Digg, or (even better) a Media Predict? Instead of the posturing and power plays we see in “Entourage,” we could use the wisdom of a qualified group to find the content that really works. Modernizing the media enterprise is where real convergence will lie, as opposed to slapping a Yahoo label on a magazine.

Historically, the great leaps in progress come not from mere technical power, but in getting people to practice old businesses in new ways. Media Predict hopes to be a part of that change. And while convergence has remained elusive, Media will surprise many by being equal to this challenge, when the time is right. Semel at that point may be remembered less as a pariah and more as a man ahead of his times.