This week and next Oxford University will be debating, behind closed doors, the reforms recommended in the recently published North report. The easy bit for both Oxford and Cambridge to decide on (and therefore the one that will be debated least) interests the government and the public most: how to get a wider range of students into these universities.

Here necessity drives. The government will demand evidence of success as the price of retaining the universities' additional subsidy now paid through the college fee. What is not so easy is persuading more good students from comprehensive schools to apply. Maria Eagle's tale (page 17) is all too familiar. Unless they come from educated and affluent homes (as many do), comprehensive pupils see these universities as foreign countries. They find them not so much daunting as unattractive.

The universities are busy in schools trying to persuade more students from poorer families to apply. It is unlikely however that they will go so far as to consider the Texan and now Californian (page 11) solution to skewed intakes. This is to make places available automatically to top students from every school.

Such a solution could not work for Oxford and Cambridge alone. Their combined entry is too small to allow them to offer each of over 4,000 secondary schools even two places each. The University of California has nine campuses with more than 150,000 students and 4 per cent from each school will only account for half its intake.

ÖйúAƬ

Nonetheless the government might like to think about a variant that would achieve greater inclusiveness and raise aspirations in less successful schools. Something needs to be done. The latest figures for admissions (page 52) show yet again that students from poorer socio-economic groups remain under-represented across ÖйúAƬ as a whole despite recent expansion and free tuition.

A scheme adapted to British circumstances might, for instance, couple an offer to top academic students in each school to some version of the scholarship scheme now being introduced by the Canadians (page 11). It might be possible to offer a set percentage of students a scholarship generous enough to encourage them to choose universities attracting above a given percentage of research funding.

ÖйúAƬ

This would be the same sort of encouragement to academic talent as the National Endowment for Science, Technology and the Arts is supposed to be going to offer to young people showing marked talent in music, performing and visual arts. With the research assessment exercise under scrutiny it would be possible to build in some qualifying threshold for universities to participate in what would be in effect a scholarship scheme for future researchers and academics.

Oxford's debate meanwhile is concentrating on the North report's recommendations on teaching, learning and examining and on the university's internal structure. The flashpoint on the first is likely to be the recommendation discussed by Jennifer Davey (page 14) that assessment arrangements be altered to mitigate total reliance on final examinations. The recommendation carries more than a hint of political correctness, addressed as it is to women's relative lack of success in winning firsts. This has worried some at Oxford for a while. But then it worries educators everywhere that women seem to do better on continuous assessment and men on sudden-death exams.

Making changes in assessment mainly to help women would be patronising. It also risks throwing away arrangements which have, inadvertently, made an Oxford degree a highly successful form of job training for work in the media. It is no accident that so many Oxford graduates go into journalism after a regime of weekly essays to fixed deadlines plus sudden-death finals. These are stringent tests of ability to turn a topic round fast and write quickly and coherently under pressure. If people are serious about diversity, Oxford should be encouraged to retain this distinctive, if brutal, characteristic of its degree.

The most difficult of the North recommendations for Oxford to tackle are those relating to governance. While highly democratic systems of governance are desirable in a research-driven academic community where outcomes cannot be planned for, Oxford, and to a lesser extent Cambridge, are too self-absorbed and inward looking.

There are no lay people involved in their deliberations unlike other universities where the chairs of council and council members come from outside. North has made some recommendations that would streamline Oxford's governance but they do not include any substantial opening up to outsiders. Reform under such conditions is hard.

ÖйúAƬ

Meanwhile the forces of assessment are massing on the universities' borders, presenting the greatest threat yet of outside interference with matters the universities regard as their province. Cambridge has been resisting demands that it welcome the auditors, fudging the issue because of the burden of assessment it is also facing.

Burden of work is not the only reason these universities are wary. The Quality Assessment Agency has a predisposition to fall in with the government's enthusiasm for all those things Dearing recommended: credit accumulation and transfer; threshold standards; compulsory components in the undergraduate curriculum and centrally recruited and trained external examiners who report to the agency.

The agency is however showing signs of flexibility towards those universities that do well in subject assessments. There is perhaps an opportunity here for those who wish to excuse themselves from complying in detail with the new regime to set up arrangements that will satisfy the requirements for accountability without risking tamping them into a standard mould.

ÖйúAƬ

The most promising development in this area is that being driven by Alan Gilbert at the University of Melbourne under the title Universitas 21. The scheme brings together a group of large civic universities round the world with roughly similar profiles. They are joining together to accredit each other, develop curriculum packages and share external examiners. The group includes Birmingham, Nottingham, Edinburgh and Glasgow in the United Kingdom, the National University of Singapore, Canada's McGill, Toronto and British Columbia, the University of Auckland and a number in the US.

Neither Oxford nor Cambridge are part of this group and probably should not be. They could, however, galvanise their own international peer group for this purpose involving say, Harvard, MIT, Stanford, the Australian National University, the Sorbonne and others. There is no reason why such a set should not overlap with others like Universitas 21 or the US-dominated international group of metropolitan universities.

As ÖйúAƬ at its top level becomes increasingly international, calibrating standards cannot remain a narrowly national matter. The government recognised this in its response to the Dearing report but did not develop its thinking on how, practically, international experts might be involved.

The Melbourne model (with which Lord Dearing as chancellor of Nottingham, and a member of Melbourne's council, is also involved) may be a useful one. It would be welcome if Oxford in its debate could lift its eyes from its internal preoccupations and consider worldwide networking of this kind.

ÖйúAƬ