Re: Copyright: Form, Content, and Prepublication Incarnations

From: Stevan Harnad <harnad_at_cogprints.soton.ac.uk>
Date: Wed, 24 Oct 2001 19:12:14 +0100

On Wed, 24 Oct 2001, Joseph Ransdell wrote:

> Suppose the copyright assigned to the publisher permits the author to:
>
> > > > "mount your version of the article on your personal World Wide Web home
> > > >page and/or that of your employer's, provided that you (a) cite the Journal
> > > >as being the original place of publication and acknowledge XXX as the
> > > >copyright owner, and (b) provide an electronic link from your article to the
> > > >Publisher's home page for the journal."
>
> And suppose the author has in fact taken advantage of this by mounting his
> or her version of the article either on his or her personal website or on
> the employer's website, complying with (a) and (b) as well.
>
> Question: Can the third party list the URl of that paper on his or her own
> website?

Of course! (Do you know of any legislation that dictates what [non-porno,
non-terrorist] URL anyone can list on anyone's website?)

> It would no doubt be best if the author's employer had an OAI-compliant
> Eprint Archive, but the number of such employers will almost certainly be
> quite small for quite a long time to come, and in the interim a portal
> website specialized to the author's professional field would accomplish much
> the same thing simply by listing it along with its URL.

I hope you are wrong that there will be few University Eprints Archives
for a long time to come!

But in any case, yes, the glue of interoperability allows all the metadata and
links to be harvested and recombined and portalled in any way one wishes
(apart from plagiarism).

> Moreover, the generic OAI search engine could be modified to include a
> separate (but cross-referenced) list of all known papers (with URL)
> available via URL on all such websites -- or, indeed, on any websites
> whatever.

This is getting into technical matters in which I am not sufficiently informed.
As far as I know, OAI search engines harvest metadata from OAI-compliant data
archives (Eprint Archives). It cannot harvest metadata from
non-OAI-compliant websites. (On the other hand, ResearchIndex, taking a
more activist approach to harvesting, can: http://www.researchindex.com/)

(Perhaps Steve or Lee could reply for ResearchIndex, and Herb or Carl
or Hussein could reply for OAI?)

> The papers so listed and thus made available would not have the special
> benefits that accrues to the paper as OAI-compliant, such as identifiability
> and retrievability via string or keyword searches, but it would still be
> identifiable from its title and retrievable by its URL.

You mean the <title> of the URL document (if it has one)? Because that,
plus the URL itself, and whatever google can invert and booleanize from
its contents, is all you can count on. No author-name, no
article-title, no publication-date, no journal-name, etc. (This is why
metadata tagging conventions like OAI are som important. Without them
a document may be buried in an unmarked common grave.)

> Is there any legal problem with this? i.e. with a third-party website
> listing of papers by title and URL, where the paper is archived in accordance
> with the permission specified in the quoted passage above?

It would astonish me if there were -- and would astonish me even more if there
were any way to enforce it even if there were... Isn't the public accessibility
and manipulability and re-resentability of URLs and their page contents the
principle on which search engines like google are based? Does it make any
difference what we call the symbol-recombinations that this generates (archives,
caches, search-results, etc.)?

> 2nd Question: What about the case where there is no such special permission
> mentioned in the copyright transference? Is there any legal problem in this
> case?

Not sure what you mean. Are you referring to papers that may have been
archived in violation of copyright? Hard to enforce, but I guess
enforcers could try to go after the original site, if they can find and
prove it. But all they can do with sites that have linked to or harvested
those URLs would be to ask them to drop the link or the copy (if they can!).

I detect a lot of inadvertent Gutenberg thinking behind these PostGutenberg questions.
It's another world we're talking about now.

Stevan Harnad
Received on Wed Oct 24 2001 - 19:12:39 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:16 GMT