Thank you for this, I take it the new API supplied by Ovid is its own proprietary search interface then.
I have not compared a seach of resources via the existing NLH search interface with results for individual resources in a systematic manner. I've just found when trying to show users NLH that using the existing NLH integrated search interface can be a little erratic in terms of not finding guidelines, shared content etc that I know exist in the discrete resources. The resources are found when a search is conducted using the interface for those discrete resources.
As you say, it would not make sense to do such a systematic comparison at present when the next generation of search is so imminent.
Paula Younger
Electronic Resources Librarian
Exeter Health Library
Royal Devon and Exeter NHS Foundation Trust
Barrack Road
Exeter
EX2 5DW
________________________________________
From: Andy Richardosn [[log in to unmask]]
Sent: 06 November 2007 11:49
To: Paula Younger; [log in to unmask]
Subject: Re: Ovid
Core Content are purchasing all the databases quoted by Claire Honeybourne from Ovid
NLH will be using the application programming interface (API) provided by Ovid to power it's own search interface. This should produce the same functionality as the Ovid interface plus enable us to target other resources - apart from just fulltext journal links - through the search interface - as the present version of NLH Search does now with guidelines and images etc. The version of Search on NLH now is not what will be available to test in beta from December so comparison against different search interfaces would not make sense at the present time.
For further details on Search 2.0 see http://www.library.nhs.uk/nlhdocs/search_2_0_newsletter_final.pdf
Andy Richardson
Digital Services Librarian
National Library for Heath
Paula Younger <[log in to unmask]> wrote:
I am also concerned that we will not be able to use the Ovid interface, so don't quite understand what the NHS is purchasing from Ovid.
A small scale study a colleague and I are currently writing up on 3 different interfaces for the same search on the same database has thrown up some extremely interesting results. All database interfaces are definitely not created equal, and in our study, intriguingly, it's actually DIALOG that threw up more relevant hits initially than either Ovid or EbscoHOST. Federated searching in general is a poisoned chalice as well as the Holy Grail, and even DIALOG as part of the Thomson corporation, with megabillions behind it, can't get its federated searching to work across more than one database consistently.
I haven't done any systematic study of the NLH search engine in the same way, but I do know that using the current search engine on the NLH site often does not retrieve results that I know are there if you search the various parts of the site individually. Perhaps I'm just using the wrong search terms.
Paula Younger
Electronic Resources Librarian
Exeter Health Library
Royal Devon and Exeter NHS Foundation Trust
Barrack Road
Exeter
EX2 5DW
|