Discussion 8 Post
Written in 1986, Svenonius’ article anticipates concerns still relevant in 2017 regarding: “the ability of users to verbalize search requests”, a definition of “aboutness”, and “semantic structures” of existing human knowledge in particular domains (p. 339). After surveying the three eras of free-text vs controlled vocabularies and examining the evidence from variance studies and draws some conclusions about where vocabulary control should reside – with users or designs – and suggest areas for research to empirically “contribute to a rational basis for the design of retrieval tools” (p. 338). And to judge by the standards of 2017, it is still often the case when using search tools that “too little”, “too much” and “not everything” is retrieved in most searches. Clearly, these are either intractable problems or insufficient imagination, work, money or a combination of all three has been applied to their amelioration. As Spencer’s 2006 article demonstrates, the future promised by technologists always seems to reside just over the horizon. The promise of web search and the unlimited availability of information hasn’t quite lived up to early optimistic assertions. After detailing methods for finding database resources in the deep web, he concludes, “until computer scientists develop a metasearch engine that will instantly search all of the web’s speciality databases simultaneously, reference librarians must continue to source the internet for the relevant, speciality tools that will best meet our patron’s particular needs” (Spencer, 81). Sadly, this situation doesn’t seem any better than 2006 as a brief search for deep web databases will quickly reveal. Many sites are dead links, many are still human curated (and probably still relying on Spencer’s worksheet). But you can bet your next iOS or Android upgrade that a solution is just months away. And in the wake of these false promises, venerable print sources like the Guide to Reference (published since 1902) have disappeared – even after making the move to the web in 2008. I can hear Elaine Svenonius sigh as I conclude this post. Plus ça change, plus c'est la même chose!
References: Spencer, B. (2007). Harnessing the deep web: A practical plan for locating specialty databases on the web. Reference Services Review, 35(1), 71-83. Svenonius, E. (1986). Unanswered questions in the design of controlled vocabularies. Journal of the American Society for Information Science, 37(5), 331-340.