[pd-discuss] PWD calculator - first attempts at consumption and a few queries ...

Thu Oct 21 10:03:18 UTC 2010

Hi folks,
         Just to introduce myself, I'm an Arcadia Fellow at Cambridge 
University Library this term. I'm taking a look at digitisation-on-demand 
services. (http://arcadiaproject.lib.cam.ac.uk/projects.html) One of my 
deliverables involves test intergration a copyright calculator into a 
library catalogue, as potential component in a delivery service. Rufus 
suggested I post my findings to this list!

I've made an initial run on incoporating the existing PDW API into a 
Cambridge catalogue, its a very hacky attempt but already useable.

We are using Aquabrowser, a fairly standard commmerical interface, so right 
now I'm bascially routing people through to an iframe holding a rough 
script that displays CC status. I've been able to marry up our record data 
API with the OKN PDW calculator API and at least get some results to a 
screen.

Here are some that work fine (look for the check copyright status link):

http://ab1.lib.cam.ac.uk/?itemid=|depfacfmdb|334346
http://ab1.lib.cam.ac.uk/?itemid=|cambrdgedb|606512
http://ab1.lib.cam.ac.uk/?itemid=|collandb|372537

Here are some more interesting results:
http://ab1.lib.cam.ac.uk/?itemid=|collandb|593096
http://ab1.lib.cam.ac.uk/?itemid=|cambrdgedb|3379711

(Or try a search and find your own. Please note that this is on a staging 
server that also runs a few hefty jobs, so response time will be a bit 
slack.)

A few points / questions at this stage, If I may ...

* is recording supported as a work type? I noticed a clause in the 
pdw/pd/uk.py file regarding it, but any queries to the service seem to 
result in a 500.

* Within certian confines, the calculator works fine examining a work in 
general. It will obviously fall down on certian editions, i.e. 1 1998 
reprint of pride and prejudice that will have annotations, introductions in 
a record. Do you have any plans to explore this area, or alt least generate 
a warning? Libraries do hold this kind information but it is piecemeal and 
not always easy to inteperet.

* Libraries also hold information on corporate authors and editors. Is 
there any provision for this in the calculator? Ditto with Crown as a 
publisher in the UK.

Happy to share code / more examples if people are interested. I'm sourcing 
our bib data via this API:

http://www.lib.cam.ac.uk/api/voyager/bibData_emc.cgi?bib_id=606512&bib_database=cambrdgedb&format=json

The php integration script also has a rough debug mode if you want to know 
what it is using:

http://www.lib.cam.ac.uk/ab_support/crCheck.php?bib_id=606512&bib_database=cambrdgedb&debug=true

Needless to say, its a fantastic concept and API access like this is 
invaluable for this project. IMHO libraries could make great use a service 
like this. Many thanks to all of those involved in its creation.

Best 

Ed  