Steven,

Some notes of clarification:

> sound a note of caution: Author-side payment is the right model for the
> steady state of the learned periodical corpus once the transition has
> been made to the new medium. But to attempt to charge authors in the
> early years -- when paper still prevails, and the new medium (and
> journal) is still struggling to establish its bona fides with wary
> authors worrying about quality control, permanence and academic credit
> -- is to risk an early and needless demise: The transition must
> initially be supported by outright subsidy. Author-side payment
> comes later.

I think there are two senses two "author-side payment". In the restricted
sense, the author side (author or author's institution) *makes a payment*
to the enterprise that publishes the article, in the sense of making it
public. This may even be one of the existing, commercial publishers.
In the broader sense, the author side *carries the cost* of publishing
the article. This is what happens when a university sets up an
"XX university electronic press" and covers its costs in the same way
as it has always met the costs of its university library - that is, by
a share of the endowment or goveernment or other grants.

In the latter case, on can of course say that the cost is "subsidized".
However, philosophically, one can also wonder whether a service which puts
one copy of an article up for general use is more like a library or
more like a publisher? Old distinctions vanish with the new technology.

The First Publication Archives assumed by the ETAI represent "author side
payment" in the broader sense of author side cost covering. We don't
exclude the possibility that some participating universities want to do it
even in the stricter sense of explicit payments, but for several reasons
it seems to be a less likely development: universities' way of thinking
about economic matters, the long duration of the publication commitment.

We also assume a certain generosity on the part of the first FPA's, so
that during the intial time they accept to publish papers from colleagues
at other universities who haven't yet gotten around to organizing their
own. This doesn't seem to offer any problems in practice.

>                 Open Peer Commentary:
>         A Supplement, Not a Substitute, for Peer Review

This was also our intention; we didn't foresee relying *entirely* on
open peer commentary.

The key points in ETAI's review system are:

1)  Arranging to have several "filters" instead of one. More
    specifically,

    -  First publication archives are in charge of making sure the
       language is correct and the typographical quality is sufficient.
       The ETAI doesn't look after that on the level of the individual
       article. (If an FPA shouldn't do that task right, we talk to them
       about it as a policy issue).

    -  First level of quality screening by the (local) editor of the
       FPA

    -  Increased emphasis on author-side self-control because once
       published, an article can't be revoked. If you publish garbage,
       it will be with you for life.

    -  Syntactic checks when the article is received and posted in the
       ETAI News Journal (starting point of the international review
       process). The article must have an abstract which *specifies a
       concrete result*. Articles which profess to "discuss" something,
       or where the abstract just says "we propose a new approach to
       ... which solves some of the problems associated with earlier
       approaches" are not received. (This means that some kinds of
       quite worthwhile articles don't fit in, but so be it)

    -  Open review starts immediately on posting in the News Journal,
       but the author must wait three months before deciding to submit
       the article to the ETAI for acceptance. This means that he or
       she has a chance to receive some feedback and think things over,
       and possibly to publish a revised version of the article.
       (But the old one doesn't go away).

    If the author submits the article for acceptance, then there is
    also confidential peer review of the conventional kind. However,
    it is assumed that the acceptance reviewers get an easier job
    (and fewer jobs!) because of the earlier filters along the way.

2)  Facilitating the job for the confidential peer reviewers, partly
    by reduction of workload - number of articles to review - along
    the lines just described, but also:

    -  Some aspects of reviewing are better done in the open side of the
       reviewing, such as kind reminders that a particular reference to
       related previous work is missing.

    -  An organizational measure. Instead of having one large editorial
       board of many tens of members, the ETAI is a consortium of
       specialized areas, each with its own area editorial board.
       We hope this will lead to a stronger identification with the
       review work on the side of the editors-review committee members.
       Working groups of 50 members don't function well *anywhere*, let
       alone when the they are geographically dispersed, unpaid, and
       there are no consequences when you don't do the job.

    -  Since it is publicly seen which papers are submitted when and
       accepted/rejected when, the pressure increases considerably on
       the area editorial board to complete the acceptance decisions
       on time. This, combined with the small size of every such board,
       should be a guarantee that their work gets sufficient priority.

    -  A small number of strict and well-defined criteria for when to
       accept resp. reject an article.

I certainly agree with you that an explicit decision about acceptance
by a respected group of senior scientists in the area of the contribution
is a necessary component of the system, and that that decision must be
preceded by confidential review. The issue for us is how to reduce the
present load, or overload on the reviewing system.

We will of course look carefully at the experience from earlier attempts
along similar lines that you mention. Thanks for your encouraging words!

Erik Sandewall