ÁñÁ«ÊÓƵ

Philip Altbach displays a startling lack of understanding of what open access means ("Hidden cost of open access", 5 June). It does not mean abandoning peer review and putting all academic work on the web without context, review or quality control. Open access is simply making peer-reviewed material officially published (in the academic sense) by journals, currently available only behind toll barriers (for print and/or electronic copies), freely available online.

The simplest route to universal open access (by all researchers to all peer-reviewed papers) is for authors to self-archive their peer-reviewed and published papers in an institutional repository. To achieve this quickly, funders and universities require (mandate) academics to do this (in a sensible extension of the "publish or perish" approach to requiring the results of academic work to be given the greatest possible dissemination).

No open-access advocate suggests that self-publication of a paper without peer review constitutes "academic publication" in the traditional sense. Journal titles and standards are expected to be maintained, as has happened in high-energy physics and astronomy, which have had near 100 per cent open access for more than a decade. Papers deposited in an institutional archive include details of the journal in which they were published, and sceptical readers can check that papers they have downloaded have matching metadata with the publisher's metadata (already generally freely available online).

It would be a serious (and easily spotted) academic fraud to deposit a paper in a repository with a claim to peer-reviewed publication when it did not have that status. If - and it is a big if, given the experience in physics and astronomy, where mass cancellations of print-journal subscriptions have not happened - print subscriptions were cancelled in sufficient quantity to undermine the cost-recovery model of administering peer review, then the cost savings provided by these print subscription cancellations will more than cover the costs of administering peer review.

These cost shifts would happen because it is not in the interests of academia to allow high-quality journals to close down.

Andrew A. Adams, School of Systems Engineering, University of Reading.