[LibX] LibX with subscription WorldCat catalog?

John Bickar johnbickar at stanford.edu
Tue Apr 15 11:40:01 PDT 2008


Thanks Godmar and Carol!

The URL from the "Just a Bunch of Catalogs" edition needs to have the : 
escaped as a %3A like so:

http://firstsearch.oclc.org/dbname=WorldCat;FSIP;query=%SWITCH{%type1}{Y:kw}{t:ti}{a:au}%3A%term1

Other than that, it runs like a top!

Thanks again,

JB


Godmar Back wrote:
> I added this configuration to the "Just A Bunch Of Catalogs" edition.
> Try entering "firstsearch.oclc.org" into the auto detection box in the
> Catalogs & Databases tab, and then import the hit you'll get from
> edition 20678963.
> 
>  - Godmar
> 
> On Tue, Apr 15, 2008 at 1:20 PM, Carol Ou <Carol.Ou at coloradocollege.edu> wrote:
>> Hey John,
>>
>>  Coming into this conversation late, but for a worldcat keyword search,
>>  try:
>>
>>  http://firstsearch.oclc.org/dbname=WorldCat;FSIP;query=kw%3A%Y
>>
>>  If you're trying this from your own browser, substitute the %Y at the end
>>  with a test search word.
>>
>>  -Carol
>>
>>
>>  --
>>  Carol Ou
>>  Systems Librarian
>>  Colorado College Tutt Library
>>  (719) 389-6895
>>  carol.ou at coloradocollege.edu
>>
>>
>>
>>
>>
>>
>>
>>  -----Original Message-----
>>  From: libx-bounces at mozdev.org [mailto:libx-bounces at mozdev.org] On Behalf
>>  Of Godmar Back
>>  Sent: Tuesday, April 15, 2008 11:03 AM
>>  To: John Bickar
>>  Cc: libx at mozdev.org
>>  Subject: Re: [LibX] LibX with subscription WorldCat catalog?
>>
>>  I used my old Stanford account to see what you see. You're right that
>>  firstsearch.oclc.org leads you directly to a search form for keyword,
>>  author, etc.
>>
>>  Closer examination reveals that this service does not appear to
>>  support RESTful access. Instead, it needs to establish a session
>>  first, then pass the session id on a subsequent request.
>>  Unfortunately, we don't support this model "out of the box" in
>>  Bookmarklets. However, it is possible to implement a custom catalog in
>>  JavaScript that supports it.
>>  If there are any takers with JavaScript skills out there, I'd be happy
>>  to provide guidance on how to implement such a custom catalog (please
>>  also see the links on the Edition Builder FAQ web page)  We have the
>>  infrastructure in place such that, once implemented, the custom
>>  catalog implementation can be shared among editions.
>>
>>  It also seems that this product may be OCLC's "Sitesearch", a product
>>  with which I am not familiar. Quick research shows a 2001 article in
>>  which OCLC announces the discontinuation of SiteSearch. Since then,
>>  they apparently open sourced the original source code, it is now a
>>  Sourceforge project:
>>  http://sourceforge.net/mailarchive/forum.php?forum_name=opensitesearch-mai
>>  n
>>  though it doesn't appear to be very active.
>>
>>   - Godmar
>>
>>  On Mon, Apr 14, 2008 at 7:16 PM, John Bickar <johnbickar at stanford.edu>
>>  wrote:
>>  > Thanks for the information, Godmar. Converting the POST form to GET
>>  >  failed with a 404 error for me. I will see if any of the folks here who
>>  >  handle the OCLC subscription have any documentation that I can use.
>>  >
>>  >  Also FWIW, our link to WorldCat is
>>  >  http://firstsearch.oclc.org/fsip?dbname=WorldCat&done=referer, which is
>>  >  what led me to believe that our access was IP address-based (and
>>  >  therefore I could set it up as a custom catalog in LibX).
>>  >
>>  >  -John Bickar
>>  >
>>  >
>>  >
>>  >
>>  >  Godmar Back wrote:
>>  >  > To answer this question, let me step back for a minute.
>>  >  >
>>  >  > LibX, when installed on a client's computer, will contact a
>>  particular
>>  >  > resource directly using the URL and method configured.  The server
>>  >  > hosting the resource will directly communicate with the client and
>>  >  > assign appropriate credentials. Many resources restrict access by IP
>>  >  > address - in this case, the server will see the IP address of the
>>  >  > machine the client is currently using. So in most cases, there are no
>>  >  > issues.
>>  >  >
>>  >  > When configuring LibX and using the auto-detection box, however, our
>>  >  > server at libx.org (which is really a VT-owned machine, aka
>>  >  > libx.cs.vt.edu, with a VT IP address) will try to contact the
>>  resource
>>  >  > for you. At this point, it's treated like a VT machine. It will then
>>  >  > try to infer the URL and method to access this resource.
>>  >  >
>>  >  > This setup works peachy when a resource looks only at the IP address
>>  >  > to determine eligibility for a service and when the URL and method to
>>  >  > access a resource are the same from anywhere (and if, of course,
>>  >  > Virginia Tech has a subscription to the resource being probed.)
>>  >  >
>>  >  > However, it can break down in a number of ways: for instance, a
>>  >  > resource could recognize a client's IP address, then redirect to a
>>  >  > different URL: say, for instance, firstsearch.oclc.org/vt when I came
>>  >  > from a VT address but firstsearch.oclc.org/stanford for Stanford
>>  >  > users.  If we detect firstsearch.oclc.org/vt from our server, and you
>>  >  > include this in the configuration, your users will be going to the
>>  >  > wrong URL that won't work for them.
>>  >  >
>>  >  > Another issue are resources that encode session ids in the URL. In
>>  >  > those cases, the URL that's detected would be the current session URL
>>  >  > assigned to our server during the probing. This URL likely won't work
>>  >  > for you at all, or at least not for long.
>>  >  >
>>  >  > These are all inherent limitations in the auto-detection. To address
>>  >  > them, we'd have to develop and install a plugin on the client
>>  computer
>>  >  > before users could configure a LibX edition. Then this client plugin
>>  >  > could connect to the resource during the auto-detection process. This
>>  >  > is, however, future future work for now.
>>  >  >
>>  >  > What you can do now is trying to find out if there's a URL template
>>  >  > that can be used to search this resource.
>>  >  >
>>  >  > - Install the web developer extension:
>>  >  > https://addons.mozilla.org/en-US/firefox/addon/60
>>  >  > - go in your browser to the search form you'd want users to search
>>  >  > from the LibX toolbar.
>>  >  > - choose the option "FORMs: POST -> GET" in the web developer
>>  extension
>>  >  > - do your search using search terms xyzzy
>>  >  > - copy and paste the URL to which the search led you; replacing xyzzy
>>  with %Y
>>  >  > - set this up as a bookmarklet for a "Keyword".
>>  >  >
>>  >  > FWIW, FirstSearch is linked to as
>>  >  > http://newfirstsearch.oclc.org/topic=all;done=www.lib.vt.edu/;FSIP
>>  >  > from the VT library site.  Going there, however, leads me to a page
>>  >  > where I need to choose a database first.
>>  >  >
>>  >  > Maybe there are some librarians on this list who know the "deep
>>  >  > linking" syntax for this service?  Maybe OCLC provides documentation
>>  >  > to libraries who have subscribed to it?
>>  >  >
>>  >  >  - Godmar
>>  >  >
>>  >  > On Fri, Apr 11, 2008 at 12:22 PM, John Bickar
>>  <johnbickar at stanford.edu> wrote:
>>  >  >> Hi,
>>  >  >>
>>  >  >>  I've been asked by one of the librarians here to add our
>>  subscription
>>  >  >>  version of WorldCat to the LibX toolbar. When I use the
>>  >  >>  firstsearch.oclc.org that pops up in the Auto-detect, I get
>>  thwarted by
>>  >  >>  an authorization and password screen.
>>  >  >>
>>  >  >>  I've got the public WorldCat working pretty well, but the
>>  subscription
>>  >  >>  one is defying me.
>>  >  >>
>>  >  >>  Is it possible to get LibX to search a subscription catalog, or
>>  does it
>>  >  >>  need to be publicly available? And if it's possible, how would I
>>  get
>>  >  >>  started doing that?
>>  >  >>
>>  >  >>  Thanks,
>>  >  >>
>>  >  >>  John Bickar
>>  >  >>
>>  >  >>  --
>>  >  >>
>>  >  >>  John Bickar, Ed.M.
>>  >  >>  User Services Technology Specialist
>>  >  >>  Cubberley Education Library
>>  >  >>  Stanford University
>>  >  >>  johnbickar at stanford.edu
>>  >  >>  650.725.1131 (office)
>>  >  >>  650.796.4637 (mobile)
>>  >  >>  _______________________________________________
>>  >  >>  Libx mailing list
>>  >  >>  Libx at mozdev.org
>>  >  >>  https://www.mozdev.org/mailman/listinfo/libx
>>  >  >>
>>  >  _______________________________________________
>>  >  Libx mailing list
>>  >  Libx at mozdev.org
>>  >  https://www.mozdev.org/mailman/listinfo/libx
>>  >
>>  _______________________________________________
>>  Libx mailing list
>>  Libx at mozdev.org
>>  https://www.mozdev.org/mailman/listinfo/libx
>>  _______________________________________________
>>  Libx mailing list
>>  Libx at mozdev.org
>>  https://www.mozdev.org/mailman/listinfo/libx
>>


More information about the Libx mailing list