On Mon, May 7, 2012 at 11:11 PM, Stevan Harnad <[email protected]>wrote:
> Richard, you are quite right that making research data open for all > to mine is not the same thing as making the texts of research articles > (Libre) OA for text-mining, and you are also right that there are > different problems associated with each. > > I should point out that Libre OA as defined by Suber-Harnad does not automatically give rights for textmining. Libre OA indicates that "some permission barriers are removed" - it does not indicate what those barriers are. For example by default an toll-access article may not be posted in an Institutional repository. The permission to post it is the removal of a barrier. However that permission does not allow text-mining as the copyirgh and re-use rights remain with other parties (publisher or repository or both). For this reason the phrase "LibreOA" is operationally meaningless - it may have political value. Wiley claims "fully open access" for its Gold hybrid and no doubt would label it as LibreOA but I can see no difference between it and publisher-supported Green OA other than that Wiley is 3000 USD better off and the research community is worse and that the authors can claim they have "Gold OA". I urge people to realise that textmining requires an explicit statement of rights of re-use. > Making article texts open for text-mining calls for Libre OA. Specifically it calls for BOAI-compliant LibreOA, not just "LibreOA". It also calls for clear licensing with something equivalent to CC-BY or CC0. > > But few publishers endorse Libre OA, for fear of 3rd-party free-riders. > BMC and PLoS do this enthusastically. I have seen no serious evidence of free-riders. > (Moreover, some flavors of Libre OA call for further re-use rights > that even some authors would not wish to grant.) > This statement is made without evidence and is typical of some of the casual and damaging inaccuracies made in this debate. I have no evidence that people fail to publish in PLoS and BMC because of their worry about re-use. > > So all in all, both data OA and Libre OA face problems that Green > Gratis OA does not face. > And they deserve careful, accurate discussion. P. -- Peter Murray-Rust Reader in Molecular Informatics Unilever Centre, Dep. Of Chemistry University of Cambridge CB2 1EW, UK +44-1223-763069
_______________________________________________ GOAL mailing list [email protected] http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal
