Embracing the Cloud: Real Life Examples of Library Cloud Implementation
Below are listed questions that were submitted during the February 8, 2012 webinar. Answers from the presenters will be added when available. Not all the questions could be responded to during the live webinar, so those that could not be addressed at the time are also included below.
- Erik Mitchell, Assistant Professor, School of Information Studies, University of Maryland College Park
- Lynne Jacobsen, Associate University Librarian for Information Resources, Collections and Scholarly Communication, Pepperdine University Libraries
- Charlene McGuire, Technology Consultant, Southwest Kansas Library System
Feel free to contact us if you have any additional questions about library, publishing, and technical services standards, standards development, or if you have suggestions for new standards, recommended practices, or areas where NISO should be engaged.
NISO Webinar Questions and Answers
1. Did OCLC create LHRs for your entire range of journals & periodicals? If so, how long did this take?
Lynne Jacobsen: OCLC did create LHRs for all of our materials, including journals. Our data was migrated over a period of a couple of months.
2. Could you share any information about the cost of this service?
Lynne Jacobsen: Costs depend on what you subscribe to as far as OCLC services. Your OCLC rep could give you a quote. You will experience savings in hardware, software, and labor.
Charlene McGuire: You would have to talk to the vendor about costs. We had a special deal because we were a consortium. I can say that the price was very reasonable and that a hosted contract was not much more than a non-hosted.
3. Did OCLC use existing records to create the holdings, create the LHRs from scratch, or some combo?
Lynne Jacobsen: OCLC created the LHRs from our Voyager holdings and item records. They did the mapping of required fields from Voyager to WMS.
4. Do speakers think cloud computing is appropriate for medical libraries?
Lynne Jacobsen: I think cloud computing would be appropriate for medical libraries as long as data security and data privacy are provided.
Charlene McGuire: I agree with Lynne.
5. Has any speaker faced pushback from paraprofessional staff? How was the problem solved?
Lynne Jacobsen: We avoided pushback from staff by communicating with them often about what we were doing and why. We also asked for their input about workflow issues. We discussed the limitations of our current system and what a new system would offer. At the same time, we were respectful of the old system as some people worked with it for over 10 years. We addressed staff concerns and questions at every point.
Charlene McGuire: We did not have a problem with any of the staff. We made sure to get library input along the way and always, always have a positive attitude about the process. We do regular trainings and have a wealth of printed info with step by step instructions. One of my personal goals was not to lose any staff because of conversion or new installs. We did not lose anyone. Our oldest staff person turned 90 this week and she does all of the cataloging, circulation, and ILL functions. When we started the process five years ago, she barely used the computer so I think the process is very doable.
6. What, if any, training was provided for librarians, paraprofessionals, and administration about migrating data to the cloud?
Lynne Jacobsen: OCLC handled our migration. Since we were an early adopter, we all learned how to migrate data to the cloud at the same time. Now that many libraries have moved to the cloud, OCLC has most likely developed training for migrating from different systems.
Erik Mitchell: Training is a big issue in moving to PaaS and IaaS systems. When we migrated at WFU we spent about six months testing different solutions and coming up with a cloud architecture and management plan that we could work with. Outside of the technical team training we also spent a good amount of time making sure that everyone in the library understood what impact cloud services would have (e.g. budgeting and administration, system access and automated processes for technical services).
Charlene McGuire: Our vendor handled the migration based on library input. We had a team of three people that worked with the libraries to prepare data files and do some pre cleanup. Then we had a two-day training with the vendor right before startup. I followed with a two-day setup and training at each library. We have user’s group meetings twice a year and incorporate training into each one.
7. Sounds good, but how do the speakers handle a data breach, data exchanged among clients, or other disaster plans?
Lynne Jacobsen: OCLC employs multiple layers of defense against external attack. They have a Security Team that operates 24x7x365. They are equipped to respond to natural disasters as well as data breach events. There is a white paper that describes their rigorous security practices in detail.
Erik Mitchell: Before getting into any cloud environment your organiztion should 1) understand the terms of the service level agreement, particularly who is responsible for data security and system management, 2) Audit the information services to be provided to ensure that putting them in the cloud does not compromise other agreements or regulations and 3) Come up with a plan that ensures some way of tracking and managing security issues. In an SaaS environment you can put these questions to your vendor and get them to sign customized agreements (e.g. non-disclosure agreements) if there is sensitive data involved. In PaaS and IaaS environments you often have to incorporate the service support levels into your own processes. In an IaaS environment this is no trivial task!
8. If a library is not automated at all, how quickly can they get their records online?
Charlene McGuire: It depends on the size of the collection. Most of our libraries that had never been automated took about two years to do their retrospective.
9. Which data is shared and which data among all OCLC WMS and which is unique to each customer?
Lynne Jacobsen: Vendor records are shared, but there is space allocated to specify local data such as a sales rep name, phone number, and email which isn’t shared. Bibliographic records are shared, but Local Holdings Records are unique to each customer. Knowledge Base e-resource collections are usually shared, but if you create them, they can be shared or private. We look forward to even more shared data such as serial pattern data and ways to conduct collection comparison/analysis.