Hi

In my role (online programmes) we get quite a lot of requests for help with article permalinks for use in reading lists, there can be some dissatisfaction with conventional methods which require intermediate pages. Whilst DOI provides a very direct approach this isn't supported for all journals nor where subscription is via a 3rd party database.

I was wondering if anyone had tried using Google Scholar to provide an alternative permalink solution (for articles which are searchable in Scholar but also accessible under institutional subscription), this kind of "permalink" is basically an automated Scholar search where the first result is accessed immediately rather than displaying all search results.  

For a University with subscriptions which are well represented in Scholar's index, this approach appears initially to work well, although there are potential pitfalls such as an inappropriate search result, potential changes in the Scholar index over time or lack of coverage for the specific subscription route.  Obviously a Scholar link created in this way would need to be tested on an individual basis.

An example of a Scholar "permalink" would look something like: 

http://scholar.google.co.uk/scholar?btnI=1&q=%22Behavioural+representation+and+simulation+of+design+concepts+for+systematic+conceptual+design+of+mechatronic+systems+using+Petri+Nets%22 

The above example leads directly to the article in T&F. The %22 characters are needed to force a specific phrase style search so the exact document title is located, rather than potentially display a textually matching but unwanted result.  As shown in the example above, it is also possible to pass Scholar through a proxy server, retaining the proxied version of the URL when visiting the search result.  

Some of the potential problems might include:

• Resolving an article page where a subscription is held locally, but where Scholar displays the article in a non-subscribed database (a potential solution may be to add additional terms alongside the title search, such as the database name/acronym thus forcing this result to be displayed instead of the default source shown in typical results).

• Future changes to search behaviour, e.g. if Scholar indexes a more “relevant” matching page.

• Changes/ deprecation of Scholar URL syntax as shown above, this is probably less likely given the simplicity of the URL.

• Loss of the Scholar Web service as it is currently provided (this is probably the most serious objection to this method vs. OpenURL, DOI etc.)


Some of the possible benefits of a Scholar perma link might include:

• Automation of URL updates within the source database, so if the target article's URL changes, this should be reflected in the Scholar index/search results automatically (potentially automating some aspects of link maintenance).  This principle would also apply if the journal shifts to a new provider with a new domain/URL.

• Relatively simple URL query (q=) which can be relied upon to work in the future, using only 1 search parameter in the Scholar search (btnl=1), if this single parameter changes and becomes unrecognised in Scholar this would likely default to a normal Scholar results page, providing at least some functionality, e.g. http://scholar.google.co.uk/scholar?btnI---rubbish---=1&q=%22Behavioural+representation+and+simulation+of+design+concepts+for+systematic+conceptual+design+of+mechatronic+systems+using+Petri+Nets%22 

• Additional access to WWW indexed content in Scholar such as scholarly reports which are not normally accessible using a conventional permalink like OpenURL

• Scholar is compatible with ezproxy and presumably other proxy services, allowing for authentication, e.g. 
 http://ezproxy.liv.ac.uk/login?url=http://scholar.google.co.uk/scholar?btnI=1&q=%22Behavioural+representation+and+simulation+of+design+concepts+for+systematic+conceptual+design+of+mechatronic+systems+using+Petri+Nets%22 

• The above approaches can also be replicated in the normal Google.com index, but can lead to less satisfactory results due to inclusion of reading lists, CVs and other odd variants of the target article.

• It is also easily possible to create a simple Javascript based Web form for staff contributors to build these kinds of links.


So I was just wondering if anyone has tried to use this method, what were the results or are there other potential benefits/ disadvantages of advocating this approach.

Rgds

Paul





Paul Catherall | E-Learning Librarian | Harold Cohen Library | University of Liverpool | P.O. Box 123 | Liverpool L69 3DA
Email: [log in to unmask] | Mobile: 0780710 6836 (internal: 6836) | Office: +44 (0) 151 794 4345 

Follow us: @LivUniLibrary | Like us: LivUniLibrary | Visit us: http://www.liv.ac.uk/library | Special Collections & Archives blog: http://manuscriptsandmore.liv.ac.uk

Paul Catherall | Llyfrgellydd E-Ddysgu | Llyfrgell Harold Cohen | Prifysgol Lerpwl | P.O. Box 123 | Lerpwl L69 3DA
Email: [log in to unmask] | Ffôn mewnol: 0780710 6836 (internal: 6836) | Swyddfa: +44 (0) 151 794 4345 

Canlynwch ni: @LivUniLibrary | Hoffwch ni: LivUniLibrary | Ymwelwch â ni: http://www.liv.ac.uk/library | Blog Casgliadau Arbennig ac Archifau:  http://manuscriptsandmore.liv.ac.uk

lis-e-resources is a UKSG list - http://www.uksg.org
UKSG groups also available on Facebook and LinkedIn
Follow us on Twitter:  https://twitter.com/UKSG