Re: Stable Self-Archiving Software

From: Stevan Harnad <>
Date: Fri, 9 Jan 2004 13:09:10 +0000

On Fri, 9 Jan 2004, Jan Velterop wrote:

> The potential for instability you describe lends support to the necessity of
> inclusion in the definition of Open Access of this:
> "['open access' means that:] The article is universally and freely
> accessible via the Internet, in an easily readable format and deposited
> immediately upon publication, without embargo, in an agreed format - current
> preference is XML with a declared DTD - in at least one widely and
> internationally recognized open access repository (such as PubMed Central)"

Well, I suppose "widely and internationally recognized" includes virtually
all universities and research institutions (and hence their open-access
eprint archives), but even so this seems to be an arbitrary narrowing
of the necessary conditions for counting as "open access" (as opposed to its
desiderata). Surely an unaffiliated author who self-archives his Nature
article on a public website has made it open access.

(The "immediacy" criterion for open access status is more concrete,
functional, and ensurable than the "permanence" criterion --
because permanence, like existence, is always somewhat open-ended and
uncertain. If we insisted that permanence was a matter of necessity
to qualify as open-access we'd also have to insist on omniscience or
prescience. Rather like saying that a scientific hypothesis is only
tenable if there can never be any counterevidence to it.)

No. Beggars can't be choosers. The problem today is that there is far,
far too little open-access content, *not* that there is plenty of it,
but some of it may not be future-proof! Perennity (like OAI-compliance)
is a desideratum for open access, not part of the definition.

(Having said that, a publisher-policy that uses temporary toll-free
access as a "hook" -- only to take it all offline again after a fixed or
arbitrary period -- is certainly not providing open access! But that is,
if anything, only a negative necessary condition: It is *not* open access if
it is *known* or *intended* to be temporary only. Nor, for that matter, is
it open access if the access has been gerrymandered to prevent downloading.)

But why are we doing all this scriptural exegesis instead of providing open

> (from the BioMed Central definition:
> We deposit also in HTML
> and PDF, but both are of course based on the underlying XML.

Format too, is a desideratum, not a definiens. (Beggars can't be choosers.
When the cupboard is bare, you don't insist that all candidate contents must
first be certified organic!)

> Meanwhile, all research articles published by BioMed Central are also
> deposited (actively, where we take the initiative, as well as passively,
> where the depository collects) in a range of other depositories and not
> just in PubMed Central... We have a deliberate policy of redundancy in
> that respect.

Redundancy too is welcome and desirable -- but not definitional!

But now we come to the true rub: The gratuitous "free/open" distinction,
yet again:

> All Open Access material published by BioMed Central, be it from our own
> site or from any of the deposits, is not just free, but truly open: it
> allows for the material to be used, re-used, and incorporated in e.g.
> databases and course-packs freely, allows unlimited print copies (some other
> OA definitions limit print to 'a small number of copies for personal use'),
> and does not even exclude commercial use (if anybody can sell OA material
> that is freely available from many sites, well, good luck to them), all on
> only one stipulation: proper attribution of the author(s).

See the reply to this regrettable attempt to co-opt the concept of "open
access" in order to make only (certain) OA journal publishing contracts
the criterion for what counts as open access:

    "Free Access vs. Open Access"

To repeat:

(1) Most of those re-uses already come with the online territory: Insert
a link to the open-access article's URL into the databases and course
packs, send its URL to whoever needs to print it off; if anyone can sell
OA material that is freely available to anyone, anywhere, any time, good
luck to them (but the toll-access journal that may hold the rights may go
after you, so be forewarned!). (About the perennity of URLs, see above.)

(2) Those re-uses that do not come with the online territory --
republishing rights, for example -- are *not necessary* for open access,
and insisting on them (e.g., by forcing self-archiving authors to first
negotiate republication rights with their toll-access publishers) in order
to meet the definition for "open access" would simply create a monopoly
on the term "open access" for (some) open-access publishers. (Even
a publisher that continues to sell a toll-access version, on paper or
online, with rights transferred, but *also* makes it permanently (sic)
accessible online toll-free (and ungerrymandered) would not, according to
this gratuitous restriction, qualify as an open-access publisher! That
probably already eliminates a number of the "gold" journals indexed by

Renegotiating rights is not a prerequisite for OA. The only prerequisite
is the right to self-archive one's own full-text online toll-free
(immediately, and permanently).

Stevan Harnad

NOTE: A complete archive of the ongoing discussion of providing open
access to the peer-reviewed research literature online (1998-2004)
is available at the American Scientist Open Access Forum:
        To join the Forum:
        Post discussion to:
        Hypermail Archive:

Unified Dual Open-Access-Provision Policy:
    BOAI-2 ("gold"): Publish your article in a suitable open-access
            journal whenever one exists.
    BOAI-1 ("green"): Otherwise, publish your article in a suitable
            toll-access journal and also self-archive it.
Received on Fri Jan 09 2004 - 13:09:10 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:15 GMT