Dorothea Salo has a good rant about stupid OPACs and makes a number of good points, a number of which have been bothering me in some form, in particular:
Why are articles omitted in filing order? Do a browse search in a library catalogue for “The Iliad” and you won’t find anything by Homer, or rather, you shouldn’t. Try “Iliad” and you should get more than enough. Especially in the good old days of the card catalogue, it might make some sense to collocate under significant words: the brain is more likely to look for the Iliad under I rather than T. It is also consistent with the eminently sensible decision to ignore and de-capitalise initial articles for corporate bodies (e.g. “Annals of the Royal Society” as a title and “Royal Society” as a heading). However, it does have the drawback in library catalogues of creating multiple entries because Uniform titles generally miss off the article (“Iliad”), and normal titles use filing indicators in the MARC record to tell the computer to index, say on the 4th character:
100 0_ a Homer.
240 10 a Iliad.
l English
245 14 a The Iliad /
c Homer.
Both the 240 (uniform title) and 245 (title) will file in the index under Iliad but will produce two different headings. The most damning argument, however, is Dorothea’s: A kind OPAC, especially in an academic library, doesn’t demand exact titles. Or titles that start with the first non-article word; human brains don’t use MARC indicators.
In other words, people don’t expect them.
Secondly, why do librarians have such a strange attitude towards Google, thinking always that it is librarians versus Google? I’ve been to cataloguing conferences (can you imagine?) where Google is seen as an enemy: Do we give in to Google or do we try to improve the way things are catalogued and indexed?
. Dorothea gives the example of the difficulty of finding a book on the library catalogue compared with Amazon, although with more complex material I would expect the advantage to go the other way. The problem, I suppose, is really with users’ reliance on Google where other sources are available, hence articles like this. There is also the interface problem: OPACs are really not nice to search on the whole: they don’t tell you exactly what power you have, their navigation is not normally very easy, and the precise nature of the results you get is a lottery (brief lists, full records, some catalogues AND keywords together, some OR them, others treat them as a phrase). I think libraries could learn a little about usability and how real people actually think from commercial and search services and these services could certainly learn a little about compiling the data itself. If Google works, we should look at why it works.