On Fri, 1 Dec 2006, peter murray-rust wrote:
> > SH:
> > The DOAJ definition of OA is better than the BOAI one (which contains
> > too many unnecessary as well as redundant requirements).
>
> If there are multiple definitions of Open Access it makes it
> difficult to operate. For science we need an operational definition
> of whether a given journal (sic) is Open Access or not.
The BOAI definition is not a definition of an OA journal; it is a
definition of OA, and as such it applies to a work (mostly to a journal
article), not to a journal. And the definition is not "operational,"
it's commonsense: A journal article is OA if it is accessible free online
(added later, as an afterthought: immediately, permanently).
(How many of a journal's articles need to be OA for a journal to be OA?
I'd say all the peer-reviewed ones.)
> If all members of this list feel that the BOAI definition (to
> which you are a signatory :-) ) has unnecessary and redundant
> requirements, then Open Access should be redefined.
It has been. But it's not such a formal, legalistic business. It's just
common sense.
"Proposed update of BOAI definition of OA: Immediate and Permanent"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/4420.html
> Personally I feel the phrase "crawl them for indexing, pass them as
> data to software, or use them for any other lawful purpose, without
> financial, legal, or technical barriers other than those inseparable
> from gaining access to the internet itself. " is absolutely essential
> for the definition of Open Access in a Scientific context. Unless I
> am guaranteed this explicitly, then publishers will deny us access to
> this right, whatever the journal (sic) is labelled.
To be able to "crawl them for indexing, pass them as data to software,
or use them for any other lawful purpose, without financial, legal,
or technical barriers other than those inseparable from gaining access
to the internet itself" comes with the territory, if the full-text is
accessible (immediately, permanently) free for all, online.
> I use the word journal as I need to be able to access and re-use tens
> of thousands or articles automatically.
I couldn't follow that. The OA movement's primary target content is
peer-reviewed journal articles -- all 2.5 million of them, as published
in the planet's 24,000 peer-reviewed journals, whether those journals
are OA journals or non-OA journals. OA is defined on the article, not on
the journal.
And "re-use" depends on what use you have in mind. Downloading (by
humans or robots) is fine; feeding to software analysis is fine;
printing off locally is fine; storing is fine.
But printing and distributing or selling multiple hard copies is not
(necessarily) fine, and not part of the meaning of OA, which refers only
to online access to the online version.
> That has to be done on the assumption that either (a) all "OA" articles
> (sic) have a machine-readable contract online (or else how do we know
> they are OA?)
You lost me again: You know they're OA because you can access them free
online. OA said nothing about a "machine-readable contract": that's
welcome, but it's certainly not necessary. (Of the annual 2.5 million
articles published, only about 20% are OA today, 15% via self-archiving,
5% via OA journals: let's not get gratuitously greedy, demanding free
access *and* a machine-readable contract -- before we at least have the
vanilla free access to the remaining 80%!)
> or (b) are published in journals that assert that all their
> articles (sic) are Open Access.
It's not what the journals assert that matters, but the articles you can
access...
> As an example I can take the statement from the Beilstein
> Journal of Organic Chemistry:
> Articles with this logo are immediately and permanently available
> online. Unrestricted use, distribution and reproduction in any medium
> is permitted, provided the article is properly cited. See our
> <http://www.biomedcentral.com/info/about/charter>open access charter.
Fine, but unnecessary, and not part of what OA means, or requires.
> Anyone is free:
> * to copy, distribute, and display the work;
"Anyone is free to access the work online" covers this. Multiple print copy
and distribution is not included (but with the online version freely
accessible, downloadable and printable by anyone, and the URL freely
distributable, who cares?).
> * to make derivative works;
Not part of OA, beyond the usual quoting and Fair Use.
> * to make commercial use of the work;
Not part of OA.
> Under the following conditions: Attribution
> * the original author must be given credit;
Trivial: standard scholarly practice for a copyrighted, original text.
Otherwise it's plagiarism.
> * for any reuse or distribution, it must be made clear to others
> what the license terms of this work are;
Irrelevant to OA.
> * any of these conditions can be waived if the authors gives permission.
A lot of irrelevant legalisms -- for OA. It's fine for CC, but CC is not
OA and OA is not CC.
> to give me the rights that I and my robots need to access and re-use
> any article in the journal.
You and your robots can do with OA articles, freely accessible online,
the same thing you can do with any other freely accessible online
content.
> >SH:
> >The best definition of an OA article (sic) is one whose full text is
> >free online >immediately and permanently upon publication.
>
> This is useful. I assume that by "free" you mean that the reader
> and/or their institution do not have to pay, not that any rights are
> transferred to the reader.
Correct.
> > SH:
> > The rest comes with the (free online) territory and does not even
> > need to be specified. No special extra license needed. And it is
> > a mistake to conflate article archiving and its needs with data
> > archiving and its needs.
>
> I have not conflated anything. I have asked a simple question - is a
> particular journal "Open Access"?
OA is not defined on journals, it is defined on journal articles.
(And I *bet* that you *are* conflating what is needed for data archiving
and re-use with what is needed for OA article self-archiving and use.)
> If I understand your position an article is Open Access if a human
> can access and read it without payment and in perpetuity
A human or a robot, like googlebot. As I said, that comes with the
(free-on-the-web) territory. A human or robot can also crunch the bits
with software. What they can do with their output is another matter. A
human can re-use the *content* in other works, with attribution, but not
the words, apart from quotation. And they cannot in general "re-publish,"
online or on paper, without permission (though why anyone would want
to republish something online that was already available, permanently,
free for all, online, I cannot quite fathom: the hyperlink and citation
sound like all anyone would ever need, or want...)
> even if it is accompanied by the following statement:
>
> "users are NOT permitted to copy,
Go ahead and download and print off for personal use: ignore this
nonsense. Ignore it also if it tells you that you may not look at your
computer screen, or breathe the air in the apartment you've just bought:
they come with the territory.
> NOR to distribute,
If that means distribute multiple hard copies, I'd be careful, because
though it seems like an utter waste of paper (when distributing the URL
would do), it would be illegal. But that has nothing to do with OA,
which is only about online accessibility.
> NOR to search,
Nonsense. Tantamount to saying you may not download, which is a
contradiction with being freely accessible, because online access means
downloading. (That's how it appears on your screen when you click.)
> NOR to link to the full texts of these articles
Utter nonsense. (I've really never understood this: roughly equivalent
to publishing something and saying you may not cite it!)
> NOR to crawl them for indexing
If it's freely accessible, it can be crawled: that comes with the
territory. What the crawler may *do* with it is another matter.
A crawler is not a mind that reads. If it gathers the text into a
derivative work that it distributes, or makes accessible, that's not
necessarily allowed. (But, practically speaking, google does it with all
freely accessible content, so until/unless google needs to lose sleep over
it, OA article authors and users need not lose any sleep over it either.)
> NOR to pass them as data to software,
See above, regarding passing eyes over computer screens or breathing air:
No permission required. What you may do with the *output* of the software
is another matter, if it consists of the harvested texts themselves. I
fervently doubt any OA provider will ever object to indexing and boolean
search, a la google, so we are being platonic pedants if we waste any
time fretting about that. (Feeding it into a commercial database is another
story, in which I have no interest and is likewise not an OA matter...)
> NOR to use them for any other lawful purpose, without financial,
> legal, or technical barriers other than those inseparable from
> gaining access to the internet itself.
Translation: everything that comes with the (free online) territory, but
no more.
> The authors MAY NOT retain copyright of this work"
And your point is...?
> If this is your position, then we understand each other and can make
> useful progress. I assume that you would argue that since we can read
> Molbank it is Open Access.
I have no opinion on what counts as an OA journal. It's not an issue.
What matters is what counts as an OA article. (Intuitively, an OA journal
ought to be a journal all of whose articles are OA, but who knows? It is
no secret that I am not particularly interested in OA journals at this
time. I used to think, foolishly, that they were part of the prerequisite
for OA, and then I realised that not only was that not true, but that
trying to convert all journals to OA is merely a distraction at this
time, when we should really be self-archiving all journal articles to
make them all OA.)
> However if some authorities take your definition of Open Access as
> definitive and others take BOAI then it possible there may some
> confusion in future.
Confusion there is, but not because there is anything nontrivial at
issue. OA merely means free online accessibility. The next step ought
to be researchers making all their articles freely accessible online,
by self-archiving them -- and their institutions and funders seeing
to it that they do, by mandating self-archiving. All this fussing
about the formal definition of OA and OA journal is merely one of many
distractions that have been slowing our progress in reaching the optimal
and inevitable. (I have no theory of why we tarry so; I'm just doing my
level best to get it all done and over with so we can move on and start
reaping the benefits of a 100% OA research corpus at long last...)
Stevan Harnad
Received on Sat Dec 02 2006 - 02:35:30 GMT