On Tue, 3 Jun 2003, Bob Parks wrote:
>sh> There is no need to generate PDF. All the Eprint archives require is
>sh> one text version, screen-readable and harvestable by full-text inverters.
>
> Well, then it is on US to be sure that the requirements are known, and that
> they are easy requirements, and that Eprint software requirements are
> known.
They are in the Eprints documentation. Every Eprints Archive need only
inform its depositor community.
http://software.eprints.org/docs/php/configarchive.php?printable=1
The ``document'' dataset
By default eprints requires at least one of ps, pdf, ascii or html
to be uploaded before an eprint is valid. You may change this list
in ArchiveConfig.pm - any more complicated conditions will have to
be checked in the eprint validation subroutine.
> For example, where is it written that the document must be
> full-text invertible?
That need not be written anywhere. All that need be written is that
there must be a screen-readable text version.
> If old versions of Word are a problem, then WE (the preachers) ought
> to remove those impediments to getting it into archives.
WE preachers are few, whereas THEY (authors) are many (about 2,000,000
peer-reviewed journal articles are published annually).
> I will offer to convert any Word or WordPefect document, up to 10 per month.
This is very generous of Bob, but alas neither realistic (how many
articles are there in economics worldwide, annually?) nor scaleable.
> MS-Word-generated HTML might not be available (if the Word
> version is old enough), and I suspect that this author would worry
> about formatting problems in the ASCII version (or even the HTML version).
Surely this problem can be offloaded on the authors and their
institutions. Perhaps their university's digital librarians are the ones
who could help:
http://www.eprints.org/self-faq/#libraries-do
Stevan Harnad
Received on Tue Jun 03 2003 - 23:25:47 BST