Online higher education won't achieve its full promise until Olivia Munn is teaching courses with titles such as "Optimizing Google for Mobile Web Performance." But at least the field is moving in the right direction.

While traditional colleges and universities have played a primary role in the development of massive open online courses (MOOCs), the online pedagogical innovation that has garnered the most media attention over the last few years, at least one company is showing that educators can have names like Google, Intuit, and Salesforce as well as Harvard and MIT.

That company is Udacity, a Silicon Valley start-up that was founded in 2011 by Sebastian Thrun, a moonlighting Google employee and former Stanford professor who pioneered the technology behind driverless cars.

In November, Fast Company reported that Udacity was "abandoning academic disciplines in favor of more vocational-focused learning." This was something of an overstatement-Udacity had never embraced purely academic learning enough to suggest that it was now somehow abandoning it. Reflecting Thrun's background, Udacity's first two courses were "Building a Search Engine" and "Programming a Robotic Car," both of which had more than a little vocational focus. And as early as October 2012, Udacity had announced its intention to partner with companies like Google and Autodesk on courses such as "HTML5 Game Development" and "Interactive Rendering."

This, it turns out, is a promising approach. As Fast Company explained, "The companies pay to produce the classes and pledge to accept the certificates awarded by Udacity for purposes of employment." And while MOOCs have traditionally been open to anyone who wants to take them, at no cost, Udacity is increasingly encouraging students to pay for its courses. While it still offers open access to all its courseware for free, it also now offers students a chance to "enroll" in a handful of classes. These students will get personalized coaching and detailed feedback on assignments and projects, plus a "verified certificate of accomplishment" upon their successful completion.

With an early registration discount of 30 percent, the price for "Introduction to Salesforce App Development," for example, is $105 per month. You complete courses at your own pace. A class might take you two weeks to finish, or it might take you two months or more. (If it's the former, you've still got to pay for a full month.)

Udacity is offering these new services to combat MOOCs' greatest perceived weakness: their low completion rates. In many cases, fewer than 10 percent of the people who register for a MOOC successfully complete it.

These changes were largely borne from a partnership the company pursued with San Jose State University (SJSU) in early 2013. Udacity offered online versions of three SJSU math classes. SJSU students could take these classes, for credit, for a substantially lower fee than they would have had to pay for SJSU's real-world versions of the classes (which the university also offered the same semester).

At the end of the semester, the Udacity students fared worse than the students who took the conventional courses. For the three classes, the pass rates for the former ranged from 23.8 percent to 50.5 percent, while the pass rates for the latter ranged from 45.5 percent to 76.3 percent.

This outcome delighted Old Education loyalists. "Sebastian Thrun has proved beyond a shadow of a doubt that real higher education can't be automated," exclaimed the Colorado State historian Jonathan Rees at his blog More or Less Bunk. Rebecca Schuman, an education writer at Slate, dubbed Udacity's effort an "embarrassing failure."

But while the completion rates for these three MOOCs were lower than the completion rates of the on-campus classes, they were, in the greater landscape of MOOCdom-where, remember, completion rates tend to hover under 10 percent-fairly impressive. By past standards, Uda­city was doing something right.

What it was doing, essentially, was cribbing from traditional colleges and universities. For all their faults, these institutions are quite adept at getting students to pass their classes. But it's not their superior pedagogy that facilitates these positive outcomes. It's their superior incentives.

First, they charge a substantial amount of money. While high tuitions serve as a barrier to entry to many, those who can pay have an increasingly strong incentive, in the form of sunk costs, to get a return on their investment. Second, universities offer something of high value for those who complete their course of studies: a college degree that promises access to a lifetime of higher wages. Finally, they're extremely selective about whom they teach: the schools with the strongest reputations for academic excellence are also the schools that take the greatest pains to ensure that only the most motivated, talented, and intelligent students have access to their curriculums.

In the San Jose experiment, Uda­city emulated some of these tactics, charging a fee and offering credit. In return, it got better results. Now it's incorporating these methods into the fare it offers everyone.

Compared to current college tuition, the $140 that Udacity is charging for a month's worth of support while taking "Data Wrangling with MongoDB" is nothing. On the Web, anyone who forks over $140 for content of any kind pretty much qualifies as an unusually engaged user, so even modest user fees will likely boost a course's completion rates. (As Kevin Carey, director of the Education Policy Program at the New America Foundation, has pointed out, one reason MOOCs have such low completion rates is that these figures typically include anyone who interacts with a given course, including the thousands of people who register but never actually make it to the first lesson.)

More importantly, Udacity is also now going beyond mere college credit and offering "certificates of completion" that actual companies will purportedly use as the basis for hiring decisions. If Udacity can eventually show that companies like Salesforce.com, Intuit, and others are regularly hiring its graduates, on the basis of their Udacity coursework alone, the company's completion rates will start soaring. In turn, MOOCs will finally start democratizing higher education in a real and dramatic way.

To truly live up to their potential, MOOCs can't just exist as a way for colleges and universities to replace the sage on the stage with a sage on the screen, and thus reduce their costs a bit. Instead, they should be used to liberate learning from campuses altogether. They should be used to eliminate multi-year tuitions, four-year degrees, and all the other centuries-old artifacts that higher education's traditional providers currently rely on to keep their services exclusive and their rates high.

The really good news is that MOOCs are just getting started. In the same way that HTML eventually went way beyond the blink tag and user-generated content evolved from bookmark lists to Wikipedia, digital courseware will continue to improve at a remarkably rapid rate.

We're already beginning to see the end of the auteur era of higher education, where a single expert develops a course, presents it to students, and is also largely responsible for offering guidance, motivation, and feedback to the students who take it, even if he or she is not especially gifted in all of these various functions. (At most, he'll have some harried grad students to pick up some of the slack.)

With MOOCs, the lone expert now needs collaborators, if only to operate the video camera. Eventually, more and more courses will be created by teams, just as movies and video games are now. Subject experts will supply the content. Teaching experts will devise courses that are customizable and highly responsive to your individual learning style. Hollywood actors will present the material.

And, finally, pedagogical Siris, more present, patient, supportive, and astute than any flesh-and-blood instructor angling for tenure can ever hope to be, will be there to hold your hand all the way through "Optimizing Google for Mobile Web Performance." "Our records indicate that students with your demographic characteristics score 10 percent better on assignments and quizzes when the material is presented by Adam Sandler rather than your first choice for instructor, Olivia Munn," your course's dedicated intelligent assistant will pleasantly advise you. "Would you like me to change your instructor to Adam Sandler?"

For those who believe that higher education should be personalized, inexpensive, as accessible to working mothers as it is to third-generation Yalies, and geared toward helping students acquire skills that employers actually desire, utopia is on the horizon.