Posting to RDA-L
On 13/01/2015 23.39, Kadri, Carolyn J wrote:
Mr. Weinheimer wrote: ” Catalogers haven’t started adding links yet (for some reason that escapes me)”
It is highly likely that the reason for this is that most of us catalogers in the U.S. don’t really know anything much about linked data let alone how to apply it”. It was enlightening to learn that the data is not as important as the “link”. Thank you.
Glad I can help. In response to Thomas, those are some good examples you give. One of the sites that has the most meaning for me is the revamped AGRIS database http://agris.fao.org/agris-search/index.do. Today, if you do a search e.g. “maize italy” (http://agris.fao.org/agris-search/searchIndex.do?query=maize+italy) and you scroll down, there are documents from Europeana and links to nature.com. If you click on an individual record (http://agris.fao.org/agris-search/search.do?recordID=IT2005602164) it really opens up, giving you results from Google, World Bank, dbpedia, and so on.
I don’t question that this provides something never really seen before, and it makes searching several resources very simple. It’s cool. I know it was a lot of work to build. But I just wonder that for those people who are interested in this article (I am not one of them, so I am unhelpful) “The genetics of virus resistance in maize (Zea mays L.)” finds the information linked here as helpful, unhelpful, or confusing. The best thing I see is that by searching Google for the exact title, you find a link to the actual paper. Of tremendous help would be (in my opinion) knowing that in Google Scholar, you can see that this paper was cited 24 times. (http://scholar.google.com/scholar?hl=en&q=The+genetics+of+virus+resistance+in+maize+%28Zea+mays+L.%29)
In reply to David Bigwood: yes, linked data sites can go down, but that is why mirror sites become crucial. They not only serve to backup the data, but to backup the entire network if one part goes down. If it all crashes however, modern websites are supposed to have “fault tolerance” which means that if one part dies, it doesn’t all die, and the user may never even realize there is a problem.