http://wwmm.ch.cam.ac.uk/blogs/murrayrust/?p=439

So it seems clear that we have to have licences. I shall take the following position:

* Any publisher or author who exposes a CC-BY or Open Knowledge Foundation licence I shall call “OA-BY”. This permits full data re-use.
* Any publisher or author who exposes a CC-NC or CC-ND or similar I shall call OA-NC or OA-ND. This does not permit full data re-use but does permit some. We may have to kludge some of the worst “conditions” like “you may post this on your web site but not in your institutional repository”
* Any publisher or author who posts a paper that I can read I shall call OA-FREE.

An excellent suggestion by Peter Murray-Rust.

On July 5 I wrote to Peter Suber:

PeterS: "What he could have added is that most OA journals do use a CC license or equivalent. Hence, it usually is safe to assume that OA journals expressly permit scholarly uses beyond fair use."

KG: I think this is in no way right. Nearly all journals in DOAI have the usual copyright notices. There are very important CC licensed
journals (PLoS, BMC, Hindawi) and a lot of other e-journals with
similar licenses but this is a small minority.

On July 6 I wrote him the results of a little test:

KG: I have checked ALL 27 new journals in DOAJ. I have found 3 CC-licensed journals:
one BMC journal = by
SC-JR = by
one French mathematics journal cc-by-nc-nd
one cc recommendation (but no cc content found), theological journal

The rest has "all rights reserved" or "personal and classroom uses" or no notice (a lot of journals). Due to language difficulties it is
possible that I have overlooked some notices but I do not think that I have overlooked CC licenses or clearly marked similar licenses.

It's not a proof but an impressive random sample.

On September 3, 2004 I commented in this weblog an entry of PeterS:

I do not agree with the conclusion that BBB "doesn't require removing barriers to commercial re-use".

BOAI says: "The only constraint on reproduction and distribution, and the only role for copyright in this domain, should be to give authors control over the integrity of their work and the right to be properly acknowledged and cited." In my opinion that clearly allows commercial re-use. I do not see that "for any responsible purpose" in the Berlin declaration (and the Bethesda definition) excludes commercial re-use.

Suber has avoided mentioning the fact hat PLoS has chosen a less restrictive Creative Commons license (attribution, derivative works and commercial re-use allowed).

I would like to reduce BBB to one B: Berlin. Berlin is the broadest (and latest, basing on Budapest and - mainly - Bethesda) consensus of the Open Access community. Berlin clearly allows derivative works. We should not say that policies which forbidd derivative works are'nt "true Open Access" but we can say that they are definitively NOT compatible with the Berlin declaration.

And, by the way, we can definitively say that the Lund criteria for Open Access Journals (http://www.doaj.org) are NOT compatible with the BBB definition. Most of the listed journals allows only free access (journals with an embargo period are not listed) and have copyright reservations - no permission barriers are removed. I have called this "Open Access LIGHT". The Open Access Community praises the Lund directory (although the EZB is much more better) but will not see the fact that the Lund criteria are exactly in the same way misleading as the diluting publisher's use of "Open Access".

What I have called OPEN ACCESS LIGHT is Murray Rust's OA-FREE.

There was also a suggestion in 2005 mady by Charles W. Bailey Jr.:

http://www.escholarlypub.com/digitalkoans/2005/05/13/the-spectrum-of-e-journal-access-policies-open-to-restricted-access/

To fully understand where things stand with journal access policies, we need to clarify and name the policies in use. While the below list may not be comprehensive, it attempts to provide a first-cut model for key journal access policies, adopting the now popular use of colors as a second form of shorthand for identifying the policy types.

1. Open Access journals (OA journals, color code: green): These journals provide free access to all articles and utilize a form of licensing that puts minimal restrictions on the use of articles, such as the Creative Commons Attribution License. Example: Biomedical Digital Libraries.
2. Free Access journals (FA journals, color code: cyan): These journals provide free access to all articles and utilize a variety of copyright statements (e.g., the journal copyright statement may grant liberal educational copying provisions), but they do not use a Creative Commons Attribution License or similar license. Example: The Public-Access Computer Systems Review.
3. Embargoed Access journals (EA journals, color code: yellow): These journals provide free access to all articles after a specified embargo period and typically utilize conventional copyright statements. Example: Learned Publishing.
4. Partial Access journals (PA journals, color code: orange): These journals provide free access to selected articles and typically utilize conventional copyright statements. Example: College & Research Libraries.
5. Restricted Access journals (RA journals, color code: red): These journals provide no free access to articles and typically utilize conventional copyright statements. Example: Library Administration and Management. (Available in electronic form from Library Literature & Information Science Full Text and other databases.)

He has repeated the color codes in his well-known text
http://www.digital-scholarship.com/cwb/OALibraries2.pdf

Now it's more complicated because there are publisher's experiments called HYBRID JOURNALS with Partial OA (payed OA and publisher-sponsored free articles).

It is not clear which color code journals have which are operating with changing free sample issues.

There are three independent criteria for the "openness" of journals (let aside the DOAJ criterium "peer-reviewed"):

1. time of free access (the embargo-question)

This is the only question Stevan Harnad is interested in.

If we can call the OA-FREE journals of DOAJ "OA" we should also can call on the article level free accessible articles after an embargo "OA".

2. Is OA permanent/irrevocable?

OA journals can change their publishing model to TA, and journal websites can disappear from the internet.

All what we can say on this topic is relative. We don't know the circumstances in 50 years for journal publishing.

3. Are there licenses/labels?

This is an essential question not only for the data re-use PMR wants.

The DOAJ entry should mention the license status of each journal!