Re: The Green and Gold Roads to Open Access

From: Susan Payne <>
Date: Sat, 10 Apr 2004 17:08:06 +0100

This may be a fairly dumb question, but recently I've read some posts
about publishers who are blue or gold or some other color. I'm finding
myself very confused by all this color business. Is there a standard
list that describes what the various colors represent? Is it fairly
new? I've been reading about it quite a bit recently and wondered how
long it has been around and what its potential staying power is.

Susan L. Payne, Librarian for Science and Engineering, Johns Hopkins University

Moderator's Reply:

The color code is extremely simple, and reflects the specific distinctions
with which the Open Access (OA) initiative is concerned:

    A GREEN publisher (or journal) has given its official "green light"
    to its authors to self-archive their papers (i.e., make them OA by
    depositing the full-text on a toll-free, publicly accessible website).

The green color comes from the original Romeo project, which listed publisher
policies on author self-archiving:

Because one can self-archive either the unrefereed "preprint" or the
peer-reviewed "postprint," green can come in two "shades": pale-green for
preprints and bright-green for postprints (or both). But the distinction
between the shades of green is much less important then the distinction
between publishers (or journals) that are or are-not green at all.

In the original Romeo color-code, non-green was coded as "white": i.e.,
a publisher that has not yet given its green light either to preprint
self-archiving or to postprint self-archiving. (Because white is often
the background colour of a page, however, I have recently proposed that
"non-green" be coded as gray rather than white. I hope this change will be
adopted. In any case, green vs. not-green is what has entered into general
parlance. "White" publishers have not been explicitly so-called much,
so not much would be involved in agreeing to call them "gray" instead!)

In the original Romeo color-code, "blue" was the code for a publisher who
gave the green (sic!) light to preprint self-archiving only or postprint
self-archiving only, but not both. It is now obvious that these are
really two shades of green, not, confusingly, another color. So I have
proposed dropping "blue" altogether, using pale-green for preprint-only,
and bright-green for postprint (as well as for both postprint & preprint,
since the postprint is what OA is really all about).

In the new SHERPA/Romeo, still more unnecessary colors have been
introduced, but the new color code is still under discussion and I am
hoping that economy and functionality will prevail, and the new SHERPA
colors will be dropped.

The new SHERPA colors would have been: green (both), blue
(postprint-only), yellow (preprint-only), white (neither). That would
have left us with green publishers, blue publishers, yellow publishers
and white publishers. I think the only distinction between publishers
that needs to be given a color-code insofar as self-archiving policy is
concerned, is whether or not they give their green light to self-archiving
*at all*: If yes, they are green. If not, they are not. The two shades
of green are only for those who are specifically interested in preprint
vs. postprint policy, and the shades need only appear as a code in the
entries in the Romeo list. They need not be used as a general descriptor
for publishers unless one is specifically interested in highlighting
preprint/postprint policy differences.

There is one prominent distinction among green publishers, however,
that *does* deserve a color-code of its own, and not just a different
shade of green, and that is whether or not a green publisher is also
an Open Access (OA) publisher: OA publishers not only give the green
light to both preprint and postprint self-archiving by the author,
but the publishers themselves archive all their articles publicly. Such
OA publishers are called "gold" publishers and their journals are gold

It will be noted that just as bright-green (postprint self-archiving)
is "dominant" over pale-green (preprint self-archiving), in that we code
it as bright-green whether the green light is for postprint-only or for
postprint+preprint, whereas the pale-green code is for preprint-only,
similarly, gold (OA journal) is "dominant" over green, in that if a
journal is gold, it is implicit that it also gives the green light to
author self-archiving.

This kind of asymmetric coding, in which one of the binary values does a
double-duty, coding both a particular value and a generic value whereas
the other the codes only a particular value is called "markedness"
(q.v.) and it is a very general property of natural language. (Test
it out by noting the difference between asking how *long* a line is
vs. asking how "short" a line is: one inquires only about generic length,
the other further implies that it is short!)

The advantage of asymmetric codes and markedness is that they are more
economical and intuitive than exhaustive rote codes that assign an
arbitrary name (or color) to every combination.

It is for this reason that the distinction between a GREEN (green-light
to self-archive) and a GRAY publisher (no green light to self-archive)
is a transparent and easily understood and remembered one, and then
so is the sub-difference between "pale-green" (preprint-only) and
"bright-green" (postprint or both), whereas the difference between a
YELLOW (preprint-only), BLUE (postprint-only), GREEN (both), and WHITE
(neither) publisher is not transparent, nor easily understood and remembered.

The GREEN/GOLD distinction is also easily understood and remembered once
one knows the GRAY/GREEN distinction: Green and Gold are the two roads
to OA. Via the Green road authors provide OA by publishing their article
in a green journal and also self-archiving it. Via the Gold road, authors
provide OA by publishing their article in a gold journal and the journal
archives it.

Stevan Harnad
Received on Sat Apr 10 2004 - 17:08:06 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:26 GMT