On 20 Jul 2010, at 18:21, Karen Coyle wrote: > Quoting Jodi Schneider <[email protected]>: > >> There've been some interesting discussions on Wiki-research-l about >> citations lately, including a post today about using a centralized, >> semantic wiki as a repository for all the world's citations, using >> infobox-based citation templates, and expressing "cited by" relationships >> as backlinks. > > > First, I would like to know what folks mean by "citations" -- from the posts > it seems that they are talking about it in terms of 'Science Citation Index' > - which resources cite other resources?
Yes, if bibliographies were listed, a 'science citation index' (backwards AND forwards) could emerge -- from the wiki's backlinks. One motivation of the project, from what I've gathered, is to have a separate namespace for citations to be referred to throughout the web (something like OpenLibrary, except for everything). One consequence of this, if Wikipedia citations used such identifiers, is that we could query for all references in Wikipedia to a source -- and notice more easily when an unreliable source were used, propogating the "re-referencing needed" upward. Creating a metadata commons -- where the bibliographic data is free for use and reuse by all -- seems like another essential feature of the proposed project. > > I always have a hard time figuring out how citation and bibliography connect. > In libraries we create bibliographic data that has many of the same elements > as a citation, but not all (e.g. lacks the page number of the cited text). > Citations are mini-bibliographic records and haven't yet started to have some > key elements such as ISBNs/ISSNs. It seems that there should be interlinking > between citations and bibliographic data created for inventory and discovery, > but that is not the case today. It would enhance the citations as well as > allow for discovery in libraries or online. Yes, it seems to me that the vast bibliographic web could become denser in that way. The distinguish between abstracting & indexing and full-text databases has become harder to recognize -- because these functions start to merge in many modern online databases (which are often at least partly full-text). Citations are for finding and identification; bibliographies are for saying what you used and helping others find them later. And these are only a few of the *things* that are out there. > > I would caution against a single repository for 'all the world's citations' > but look to linking as a better solution. I would also caution against > limiting citations to academic textual materials. It would be good to know > where photographs, illustrations, maps, graphs, and data have been cited. To > include these one would need to have the expertise of those communities. This > leads me to conclude that we might have many communities of resource > description that interact with citations. I don't think there needs to be one repository (and of course if there were one it had ought to be mirrored)! But it needs to *act* like a single repository from a user perspective. A transparent linking infrastructure might be able to do that -- it would know "this is a map, oh, I'll look in the map directory", "this is a book, I'll try Open Library, and failing that, Library Thing", "this is a scientific article, so I'll try ..." -- would help. Users don't need or want to make these distinctions. But minting and maintaining identifiers is work. -Jodi > > kc > > -- > Karen Coyle > [email protected] http://kcoyle.net > ph: 1-510-540-7596 > m: 1-510-435-8234 > skype: kcoylenet > >
