But like most big ideas there is a suspicious lack of detail. What kind of networks do they want? And what kind of education system will we end up with if they take hold ?

“The thrust of all we do is about schools working together,” says an adviser to Charles Clarke, the Education Secretary. But the problem for teachers is that there is considerable confusion about the blueprints on offer, and little clear evidence about which ones work. Labour has said it wants to see diversity in education, but unless some clarity emerges they may end up with chaos.

At one point it looked as if federations - enshrined in the 2002 Education Act - were the favoured blueprint. There are 33 pilots on the ground now, with pound;16 million of central funding, and while there are several models the most common characteristic is a single group of governors overseeing several schools. There may be a “superhead” as well.

Federations are used to solve specific problems, for example joining together groups of primary schools in rural areas, or as a rescue mechanism where a head leaves a failing school. But federations, especially the “hard” variety involving legally-binding structures, are not popular with everyone.

When Mr Clarke came into office he let it be known he favoured “softer” networks. And teachers were suspicious of superheads and talk of mergers and takeovers.

By contrast, informal school partnerships are blissfully free of central government interference. They are gaining momentum thanks to the work of Tim Brighouse in Birmingham.

“Collegiates” - as he called the new clusters - may start out as talking shops but can end up as radical collaborations, for example sharing post-14 provision through common timetables.

“I am trying to get a voluntary bottom-up movement going across the capital,” says Mr Brighouse, now Commissioner of London Schools.

School partnership is spreading as an idea. “I am sure there are hundreds of examples of informal networking around the country. There is no one model, and no one is claiming to have all the answers right now,” says Sophia Parker, of the think-tank Demos, which is working with the Department for Education and Skills on networks.

La Swap, a consortium sixth form in Camden, London, is one example which, like most collegiates, emphasises the individual ethos of each “home school” while offering something extra - a contrast to the “superschool” model of many federations.

So can the Government really just stand back and let collegiates run their course?

The problem is that voluntary collegiates can lead to politically explosive end games. If you swap children around, for example, within a school cluster at 14, what happens to parental choice? And what about school league tables?

Mike Tomlinson’s proposed reforms would lead to more networking as heads scramble to provide a wider range of courses post-14. He thinks this has implications for league tables.

“If we are going to have anything like the proposed model, it is less important where the learning is delivered compared to the fact that it is delivered, so how do you assign the results ?”

The problem for ministers is that networks probably provide the only mechanism by which the Government can deliver its reforms given limited resources, despite Chancellor Gordon Brown’s recent largesse.

Don’t expect to see the issue in the manifesto for next year’s expected election. There are too many difficult questions to be answered on the hustings and no hard data from the pilots. But if Labour gets its third term, there is little alternative to making networks work.