Francis Jayakanth <fr...@ncsi.iisc.ernet.in> writes > Hello List, I have a question about the inclusion of all the > publications of an author in an institutional repository (IR). The > actual scenario is the following: > > Author A works in Institute X for 'n' of number of years. Institute > X has an IR and all the works of A till the time he was associated > with institute X are available from the IR of institute X. Author A > then relinquishes his position at institute X and takes up a > position in Institute Y and henceforth all his works are available > from the IR of Institute Y and so on. > > For a prospective research student or an end user, it becomes > difficult to get a full picture of the nature work being done by > author A from single source.
I agree. We have been dealing with this in the RePEc digital library by having authers build their profiles. Each author does it for him/herself. We have over 20,000 economists who have registered. We use the profiles to build rankings. As an example, for Ben Bernanke, we have http://ideas.repec.org/e/pbe55.html That system has been growing for 11 years now. > My question is, what should be the policy of the IRs with respect to > the scattered works of an author and how should it be handled? I am working on a system that implements an author claiming system for all disciplines. This is the AuthorClaim system at http://authorclaim.org. The software was funded by the Open Society Institute (OSI). The AuthorClaim system runs since 2008. Within the AuthorClaim system, I am currently reading data for IRs. I don't want to do the harvesting myself. I have been relying on the kind folks of the Bielefeld Academic Search Engine, http://www.base-search.net. I have been getting from them data about documents that could be used for author claiming. Clearly not all IR-stored items are suitable. For example, a mediaeval manuscript would be unlikely to be claimed by a living author. Also I am not taking archives of student work as a first priority but I will probably relax that later. Also some data I know for sure is duplicated. Say PubMed central paper are all in PubMed, so I only take one description of it from that source. The data that I have from selected from BASE is documented here http://wotan.liu.edu/base You may want to check that your IR is included. If you don't see it please conduct searches in BASE http://www.base-search.net to see if you find your documents there. If you don't see them contact, contact the Master Aggregator of BASE, Friedrich Summann <friedrich.summ...@uni-bielefeld.de>. If you do see them in BASE but not in http://wotan.liu.edu/base contact me with a sample record you found in BASE and I will see what I can do for you. At the moment I am still reading the BASE data. Today my records show that I have 2.7 million of 12 million read. Overall are over 100 million authorships that can be claimed in AuthorClaim. All profiles are available at ftp://authorclaim.org for easy mirroring and reuse. The data is licensed as CC0. All profile changes are instantly recorded in the ftp tree. It should be technically straightforward to import all profiles in AuthorClaim into any IR. If a local paper has been claimed (which can be found be looking at the handle) you can build a profile page that contains detailed information about the local paper and brief links to remote papers. Then if your local users complain that the profile page only shows brief descriptions for the remote papers tell them to upload. If you are running EPrints, we have some software that was written as part of the OSI project that allows a tighter intergration between AuthorClaim and an IR, including author name input aids and automatic AuthorProfile update when an author uploads a new paper. But discussing details here would be too technical. Cheers, Thomas Krichel http://openlib.org/home/krichel http://authorprofile.org/pkr1 skype: thomaskrichel
