Stevan,
Unfortunately (or fortunately for poets, etc) most words have multiple meanings so simple Boolean search is often not sufficient and overly complex Boolean will miss relevant items because illogical human beings often call the same thing by different names. This is one reason why even the best automated translation systems can sometimes produce complete garbage. A subject tag (even a simple one like 'science') can often provide the necessary context to clarify the intended meaning of a search term.
So until we get intelligent systems that can clearly distinguish between orange the colour and orange the fruit we may need to have some form of subject tag. I think we will agree Google Scholar is a useful practical tool and it has simple subject categories to help disambiguate words.
Card catalogues? Haven't you been to a library for the past 30 years? :-) .
Regards,
John Smith.
> -----Original Message-----
> From: Repositories discussion list [mailto:JISC-
> [log in to unmask]] On Behalf Of Stevan Harnad
> Sent: 13 March 2008 12:17
> To: [log in to unmask]
> Subject: Re: Challenging assertions
>
> Dear friends,
>
> (1) Are individual journal *articles* classified in your library
> card
> catalogues (using any classification system whatsoever)?
>
> (2) Do your users often search the literature for all and only the
> articles on a particular topic published by your university, and
> your
> university only?
>
> (3) Do you really believe that in the online age, any
> prefabricated
> classification system could ever beat boolean full-text search?
>
> If the answer to these three questions is No, then why not leave
> search and classification (if any) to the joint harvester level
> rather
> than trying to microtag individual journal articles at source?
>
> A bemused, nonbiblionomic lurker,
>
> Stevan Harnad
>
> On Thu, 13 Mar 2008, Ian Stuart wrote:
>
> > Rebecca Parker wrote:
> >> I can't speak on behalf of UK universities, but I would
> challenge
> >> Arthur Sale's claim that most Australian university libraries
> use LC.
> >> In my experience, this is not the case at all.
> >
> > Before this particular thread runs off in yet another tangential
> direction,
> > the use of a "Subject Classification" within repository software
> is not one,
> > I think, that can be resolved easily.
> >
> > The very *original* question was about metadata fields within
> repositories,
> > as part of EDINA's quest to provide the most useful Transfer
> Service possible
> > from "the Depot".
> >
> > I know, from speaking to UK repository managers, than many of
> them actually
> > dispense with any LCC/Dewey/JACS/whatever classification (as the
> term is used
> > by the Librarian/Archavist community), and instead use a
> heirarchical system
> > that is more applicable to their Institution:
> school/faculty/department.
> >
> > So, although it would be wonderful (in the ideal world) for all
> Repositories
> > to use the same Subject Classification scheme, it just isn't
> going to happen
> > - so lets not side-track ourselves over this one.
> >
> > (and cross-walking Classification Trees is not something that
> "the Depot" has
> > on its investigative ToDo list :chuckle: )
> >
|