California made big news recently with something even more massive than MOOCs: a legislatively proposed “statewide platform through which students who have trouble getting into certain low-level, high-demand classes could take approved online courses offered by providers outside the state’s higher-education system.” That, at least, was how it was described in a recent Chronicle of Higher Ed article. Predictably, the Chronicle soon published an opinion piece declaring (in its headline) that this was “A Massively Bad Idea.”

Nevertheless, it’s an idea that’s catching on. In the March 27th issue of Inside Higher Ed, the top headline was “Economies of Online Scale,” a story about statewide systems in New York and Florida that will presumably improve on the California idea by being more organized, by offering not just aggregations but single-source conduits of online ed. Not content with online courses from a number of institutions on a number of platforms, each will provide one unified system, either through one institution (in Florida’s case, the University of Florida) or through one platform (Blackboard in the case of SUNY). Why? The single most telling remark comes from SUNY Chancellor Nancy Zimpher: “I think the problems the country is trying to solve simply cannot be solved one institution at a time.”

SUNY is in fact planning to “allow up to a third of the credits for certain SUNY degree programs to come from outside institutions, including MOOCs,” according to Zimpher, and apparently the system is in talks with Coursera, a company whose careful affiliation with the nation’s top tier institutions of higher ed was the focus of an exposé/critique in IHE just days before, an article titled “Coursera’s Contractual Elitism.” But that elitism is not a problem from Zimpher’s perspective; on the contrary:

Being able to bring in credits from courses taught by professors at more elite institutions – Stanford University or Duke University – could help improve student perception of a SUNY education to being much more than a “degree of convenience,” the chancellor said.

It can be vertigo-inducing to look at problems and solutions from the heady heights of massive courses and vast systems, of one-ring-to-rule-them-all platforms and status-seeking outsourcing. If the issues are ones of teaching and learning, of access and degree-completion, it might be good to look at things from the bottom up no less than the top down. What works in all modes of college teaching was documented by Chickering and Gamson decades ago in the seven principles of good teaching, applied by Steve Ehrmann to online ed in the Flashlight Project that documents successful teaching online. Basically, and this is hardly counterintuitive, what works is lots of interaction and feedback.

These are not the distinguishing features of online instruction getting play in the news these days. Instead, we have fundamentally presentational (“sage-on-the-stage”) instruction in huge courses with no real feedback from or interaction with the instructor; the results are predictable: appalling completion rates in courses taken mostly by people who are not college students and taught by professors whose own institutions wouldn’t give credit to them (as noted in a year-end retrospective on MOOCs). Most faculty involved, as a recent survey of MOOC profs in the Chronicle revealed, are “professors with no prior experience with online instruction.”

If this feels like a bandwagon to jump on — and Bill Bowen’s thoughtful commentary “Walk Deliberately, Don’t Run, Toward Online Education” is full of cautions —  it might be good to review the lessons of two decades of successful steady growth in online education. If the extant online offerings, MOOCs excepted, seem “fragmented” — the problem underscored in the “Economies of Online Scale” article — it may be because they were developed in ways the stay close to the course goals and established policies of the institutions offering them, and there may be something to that. Bigger isn’t always better, especially when what’s good happens at precisely those levels of interaction that scaling up pares down.