On June 27, 4,000 people in Nashville's Convention Center celebrated the 25th anniversary of chartering. Minnesota's innovation in a 1991 piece of legislation has spread remarkably: Today, across the country, about 6,800 charter schools enroll some 3 million students. In Minnesota, 166 schools enroll more than 51,000.

"The most visible and substantial education reform of the modern era … the best example of reinventing government applied to public education," is how Chester Finn and Bruno Manno describe chartering in their new book from Harvard Education Press.

Public education, long a public-sector utility, is now a two-sector system. The district sector dominates — enrolling 55 million students and spending annually about $600 billion. The charter sector is small — but large enough and dynamic enough now to attract opposition from big-city districts that fear its growth.

This new controversy is forcing some basic questions: What is the charter sector for, anyway? How does this new second sector fit in America's strategy for improving its public education?

As opposition rises, confusion reigns. I see four ideas moving:

1) Ignore chartering. Most students are in districts. Realistically, the challenge is to get existing schools to do better. We should set standards, measure performance, impose consequences. It's the accountability model for "standards-based systemic reform."

For 30 years, this notion that the problem is one of performance has dominated national policy. Yet "performance" barely improves. To keep betting all the chips here is a risk, but not a necessary risk, and so not ethically an acceptable risk. Design also matters, for systems and schools.

2) Stop chartering. Word about the Broad Foundation wanting to create thousands more charter ("quality") seats has agitated Los Angeles. Chicago, Philadelphia, Detroit and other districts are blaming their financial crises on the growth of enrollment in "charters." So the impulse is to cap, contain and downsize the charter sector. Saving the district comes first.

This is unlikely and undesirable. These districts' problems have other origins. Removing the pressure from chartering would remove the incentives for change. States will not want to go back to the public-utility model they found so frustrating before 1991.

3) Combine charter and district. Create an elected superboard overseeing but not running autonomous schools, both charter and district. David Osborne, author of "Reinventing Government," sees this "portfolio" idea emerging in Washington, D.C., New Orleans and Denver.

It might be unrealistic to believe an elected board will tolerate school autonomy. Wanting to run the schools, it might block innovation. Politics likes sameness. Michigan rejected the superboard idea for Detroit.

4) Replace the district. The idea here is that the district is an unsuccessful institution; the sooner it disappears, the better. Get more charter management organizations (CMOs) running "high-performing"' schools.

This, too, is unlikely. And unnecessary. The district sector performs acceptably in most places. Most people like traditional school. Improving existing schools is part of a two-sector strategy. CMOs are not conspicuously interested in innovation; they push a model of school not suitable for all.

There is an alternative to all this, a centrist strategy.

Good ideas abound for producing better schools. The difficulty has been with the "how" of change. The idea of superintendents changing district schools comprehensively has proved unsuccessful. So, let's be practical: Try a different "how."

Successful systems change gradually, as innovation spreads. These are open systems. Someone tries something different. Always there are "early adopters." Nobody has to adopt the different. More do, as the innovation improves. Some lag. We see this diffusion of innovation all around us.

We don't argue about whether communications should use land-line phones or cellphones, conventional autos or hybrid/electric alternatives. We do both; the new spreads while the traditional improves.

We want education to be a successful system. So we should let schools and teachers try things. Use the charter sector to generate new forms of school and new approaches to learning, and encourage districts to adopt these innovations.

This can work in Minnesota. Our charter sector differs significantly from most states':

• It is diverse: 53 percent students of color, 55 percent on free lunch, 20 percent limited English proficiency, its full share of special education.

• All schools are nonprofit. Almost all are free-standing; perhaps five in CMOs.

• Most schools are traditional, but a significant number innovate — with project-based learning, age 3/grade 3 models, language immersion schools other than Spanish, computer gaming, career academies. The association of charter schools is inventorying the innovation.

• The schools get most of the operating money available to district schools.

• Minnesota has multiple authorizers (entities that approve and oversee schools). Its "single-purpose" authorizer is an innovation.

• Another innovation here is the idea of organizing teaching on the "partnership" model common in most professional areas: law, medicine, architecture, engineering. Nowhere else have union people formed an authorizer to spread this model.

Minnesota's charter sector has schools that build around the individual student's interests, making motivation central. Motivation matters. And is usually missing: Who would assert traditional school is arranged to maximize motivation?

More and more superintendents are ready for change. It is harder in the district sector to introduce and sustain the "different." But leadership increasingly understands gradualism. Lakeville's Impact Academy is an example. The website of my organization, Education|Evolving, tracks district innovation. The state is starting to help, with its Innovation Zone legislation and, in 2016, with a $500,000 grant program for teacher-powered schools.

The "partnership" innovation can retain and attract top-quality teachers. It is attractive to unions whose members want professional responsibility: Many in Minneapolis and St. Paul hate "regimented learning."

Last November, 220 teachers from 23 states met in Minneapolis to consider how to create and operate the "teacher-powered" arrangement. Half were from charter schools; half from district schools. This is innovation-based systemic change at work. (See www.teacherpowered.org.)

No other adequately effective strategy for public education is visible. Deploring the problem and reaffirming the goal without a "how" for getting there only builds frustration. It is time to be practical; time to try "innovation-based systemic reform."

For those "in charge," this will not be easy. No more "comprehensive" change. No more mandating solutions from the top. No more hunting for "the one best way." Instead, copy successful new approaches.

Schools and teachers encouraged to try things probably will depart from the consensus. That's OK; the idea is to try things. Give them time: No innovation is perfect at the start.

In schools trying things, the concept of performance will be multidimensional, with success judged on balance. Is "quality" ever one-dimensional? Think about your car: Quality includes purchase price and operating cost, style and color, capacity and reliability, speed and safety. Schools can work to get children proficient in English and math and also be developing critical and creative thinking and social/emotional skills. Don Shalvey, who heads K-12 work at the Gates Foundation, says this is the direction they are looking.

What to do?

The rule is: When what you're doing isn't working and you don't know for certain what will, you try several different things.

For the charter sector, that means state government creating a climate of encouragement for innovation, organizing itself to support "different." Also, creating more single-purpose authorizers that, like Innovative Quality Schools, request proposals for "different." It means money, public and private, for start-ups.

For the district sector, it means improving the innovation zone legislation. Perhaps re-creating the innovation office in the state Education Department and covering districts along with charters. It means pushing Minneapolis to give real autonomy to its "partnership" schools. It means encouraging St. Paul to consider whether student misbehavior might be a symptom of a failure in its schools to offer a program students find engaging.

For the governor and education commissioner, it means making aggressive use of the flexibility given states by the recent national legislation.

The potential for new-and-different is enormous. Minnesota could, like New England, be trying competency-based learning — it could even (gasp) consider doing away with age-grading. Minnesota could personalize standards; make them rigorous but different, say, for students headed into engineering and students headed into music.

Let's build a charter sector encouraged to innovate, that will push the district sector also to try things. Let's get education changing the way successful systems change.

Ted Kolderie is co-founder of Education|Evolving.