*Update* Please see Stevan’s comment below which clarifies some points I didn’t get right!
Open Access (OA) publishing means that the research paper or other information may be accessed by a reader without payment, unlike much scholarly research which is published behind a paywall. Paywalls are often invisible to university members, as they can click though without logging in if they are on-campus (much paywall access is mediated by IP addresses), but the such subscriptions are still paid by their institutions and they can be pretty expensive!
The main argument for OA is that much research is paid for by taxpayers through government grants, so it is argued that the end product of publicly-funded research should be freely available for citizens/taxpayers to read. Furthermore, academics in many fields often wish to continue to keep up-to-date in their subject after they have retired. They may still contribute to professional organisations and contribute papers and letters, but if they are no longer current members of a university, they lose institutional access entitlement to scholarly resources and thus their experience and expertise is lost to that field of study after they cease paid work for that institution.
Jackie Wickham (Nottingham University) spoke about research repositories and two routes to open access (OA) publishing: the more established gold model and the emerging green model.
The gold route is also known as the ‘author pays’ model, and it means that the publication of an article in an OA journal is usually paid for by the author’s institution or included in their research grant. Increasingly, UK Universities have established publication funds e.g. Nottingham and Birmingham. There are two further options for the gold route: OA publishers (PLOS, BioMedCentral, Hindari) or traditional publishers with OA option (Nature, Elsevier, Springer).
The green route involves self-archiving the article or conference paper in a repository of published research. This may be done by subject (e.g. PubMedCentral, arXiv, Repec) or by institution (also providing a way for universities to showcase their research and to preserve it). There is no charge for depositing the article or paper; the costs of running the archive are met by the institution.
The green route is usually used in addition to publishing in a journal, which may be OA, subscription or a hybrid of the two.
A problem with the green route is that the commercial journal publisher may impose an embargo on the publication of the article anywhere else but in their journal. As I understand it, the solution to this is that a very similar but not identical version of the article is presented to the commercial publisher and to the repository.
Laura's Dark Archive 
Pingback: CILIP Umbrella conference 2011 « Laura's Dark Archive
GOLD AND GREEN ROUTES TO OA (OA ≠ OA PUBLISHING)
Laura, good summary. Just a few suggestions:
(1) It is not the Gold and Green routes to open access (OA) publishing, it is the Gold and Green routes to open access (OA).
(OA publishing is Gold OA. The goal of the OA movement is OA, not necessarily Gold OA — although it is likely that Green OA will eventually lead to Gold OA.)
(2) Yes, some journals still embargo Green OA self-archiving (of the author’s refereed, revised, accepted final draft). But the majority of journals (over 60%, including most of the top journals) endorse immediate Green OA.
And authors who decide to abide by publisher embargoes can deposit their embargoed article in their institutional repository just the same, setting access to it as “Closed Access” instead of OA during the embargo. Via the repository’s semi-automatic “eprint request” button users can still request and authors can still provide “Almost-OA” to tide over research usage needs during any embargo.
Sale, A., Couture, M., Rodrigues, E., Carr, L. and Harnad, S. (in press) Open Access Mandates and the “Fair Dealing” Button. In: Dynamic Fair Dealing: Creating Canadian Culture Online (Rosemary J. Coombe & Darren Wershler, Eds.) http://eprints.ecs.soton.ac.uk/18511/
(3) But, most important, the worldwide research community (researchers, universities, research institutions, and research funders) are in a position to provide at least 60% OA (plus 40% Almost-OA) immediately by mandating Green OA self-archiving.
(In contrast, Gold OA cannot be mandated, is currently at about 12% and at its current growth rate will not even reach 60% for at least another decade.)
http://openaccess.eprints.org/index.php?/archives/821-guid.html
Stevan Harnad
EnablingOpenScholarship
http://www.openscholarship.org
Dear Stevan,
Thank you so much for the clarification – I appreciate it!
One concern I have about green OA and I would be interested in other’s thoughts because I rarely hear it voiced is the potential confusion of multiple somewhat different versions of manuscripts that can be disseminated by the process. For example I recently submitted a manuscript with a colleague for consideration by a journal that allows green OA. Since I wanted to get the information out, I posed a preprint clearly labeled as such under consideration by the journal. If accepted it will undoubtedly require some changes and I will post a revised version as accepted and eventually the actual published version by the journal will also be disseminated with possibly a few more minor changes and paginated somewhat differently. If it is not accepted, and submitted elsewhere there would also be multiple versions floating around.
Even with versions clearly labeled, it seems to generate a real potential for confusion. I think the benefits of green OA far outweigh this issue but it does seem a real issue to me. I would appreciate other’s thoughts on this.
Dear David,
The EPrints repository software tags versions and identifies the most recent one. And the question to ask: would an access-denied user rather have access to multiple green versions or access to no version.
Here’s the pertinent passage from the (12-year-old) self-archiving FAQ:
http://www.eprints.org/openaccess/self-faq/#23.Version
Cheers, Stevan
23. Version control
“I worry about self-archiving because there may be many versions and there is no way to be sure which is which, and whether it is the right one.”
There will be self-archived preprints, revised drafts, final accepted, drafts (postprints [ but not necessarily the publisher’s proprietary PDF), updated, corrected post-postprints, peer comments, author replies, revised second editions. OAI-compliant Eprint Archives will tag each version with a unique identifier. All versions will be retrieved by a cross-archive OAI search , and the “hits” can then be identified and compared by the user to select the most recent, official or definitive draft, exactly as if they had all been found in the same index catalogue.
Thanks for the feedback.
Pingback: The Finch Report – media roundup « Laura's Dark Archive