[Eril-l] link checking best practices

[Patricia Pang]

Hi all,

I use a Python script so I can run a KBART list or any title list with URLS from the same publisher though and it exports a csv file to let me know which titles aren't working. The script works by scraping the HTML of the page and looking for elements on the page that are unique to having access. For example if my institution has access to an Ingenta title there will be a small  icon so I ask the script to tell me if it see the icon to return "Right on!". If the page doesn't have the icon, the script will output "Look into this" and I can filter those titles and investigate.

The script requires each publisher platform to be defined individually and unfortunately can't check entitlement years but it has been so much better than spot-checking for access. If we purchase or subscribe to a collection of hundreds of ebooks or journals I can run the list through and ask the publisher to restore access to titles we should have. The University of Victoria uses Serials Solutions so I export title lists from their knowledgebase.

The code for the one I'm maintaining can be found here https://github.com/UVicLibrary/KrakenAccessChecker and it was taken and modified from another access checker here https://github.com/telezoic/Inquisitor-Python-Wx I knew nothing of Python or XML before I started looking into this access checker so it is not difficult to get running, although it takes a bit more knowledge to troubleshoot error messages and write the publisher definitions. I'm happy to answer any questions or elaborate. I like this method because I can control the results and refine my criteria. I find with a lot of commercial link checkers there are too many false results or they only check for 404 errors.

We also have a different workflow for our individual subscriptions which we have been incorporating in our annual renewal process. Before approving a subscription for renewal, Acquisitions staff check the records for each title listed with our subscription agent and determine if receipt has been okay for print, note change of publishers and price increases, and if there are any other reasons to hold our renewal. The last two years we've had 5 staff members check a bit under 2000 titles in the summer on top of their other work and it took around 3 months.

Last year I asked staff to check the online titles when going through this annual renewal process. They searched each title in our catalogue to make sure it was represented and that we have access to the latest issue. This year I'm requesting staff to also check each print title in the Serials Solutions knowledgebase and see if there is online access available anywhere. This is to catch free open access titles for our print subscription titles, titles we might want to move online, and if the title is included in an online package deal.

Cheers,
Patricia


[LIBR_comb_v_4c_rgb.jpg]

Patricia Pang, Electronic Resources
William C. Mearns Centre for Learning-McPherson Library Acquisitions
3800 Finnerty Road PO Box 1800 STN CSC, Victoria, BC  V8W 3H5 Canada
P: 250-721-8246| F: 250-721-8240 | ppang at uvic.ca<mailto:ppang at uvic.ca> | uvic.ca/library<http://www.uvic.ca/library>



From: Eril-l [mailto:eril-l-bounces at lists.eril-l.org] On Behalf Of Beth M. Johns
Sent: Friday, June 15, 2018 5:24 AM
To: ERIL-L
Subject: [Eril-l] link checking best practices


Hello,



I am developing procedures for link checking e-journals, both individually subscribed titles and links within aggregators. We would like to have our student workers do this for us on a regular basis.



If you do link checking of your e-journals, how often do the student workers / staff do this?



Do any of you have guidelines / best practices that you can share?



Thank you.



Beth





Beth M. Johns, MLIS

E-Resources Librarian

Saginaw Valley State University

Melvin J. Zahnow Library
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.eril-l.org/pipermail/eril-l-eril-l.org/attachments/20180618/76733766/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image003.jpg
Type: image/jpeg
Size: 1927 bytes
Desc: image003.jpg
URL: <http://lists.eril-l.org/pipermail/eril-l-eril-l.org/attachments/20180618/76733766/attachment.jpg>