[Eril-l] questions about ebooks: dda, etc., catalogue/other marc/holdings management

Heather Shipman heather.shipman at cornell.edu
Mon Jan 22 11:04:13 PST 2018


Hi, Melissa,


A.     DDA:

a.      We’ve had a myiLibrary DDA since 2009 or so, without any weeding until recently. MyiLibrary is merging onto ProQuest Ebook Central; when we migrated our DDA profile to PQEC, we weeded out the unpurchased titles published >3 years ago.

Three is a nice round number, but we also chose it based on some research Melissa Goertzen did at Columbia University (https://melissagoertzen.wordpress.com/portfolio/e-book-study/) – one of the things that stuck out at me from her report is that it takes about 3 years for a newly published ebook’s usage stats to really start to reflect its long-term usage. Riffing off that finding, we figure that if nobody’s triggered the title in the past 3 years, it’ll probably continue to have low usage forever, so we might as well ditch it from the DDA and if somebody actually wants it, they can submit a request and we can firm-order it.

Whether we’ll weed on a yearly basis is an issue we haven’t talked about yet. It’s probably a good idea? It’s probably easy enough? But we’re not at that decision point yet. I guess I should run a query to see how much money we’ve spent in the past on the types of books we might’ve weeded.


b.      We de-dupe DDA records during their load into the catalog, every load – although not loading them into the catalog is not the same as removing them from the DDA profile on the platform itself. We’ve occasionally had purchases trigger for titles that patrons must have run across from a source other than the catalog. Not a lot, but not zero. That’s something we’re hoping to figure out a solution for, post-migration.

We also have a daily query that checks for duplicate ISBNs, and someone goes through that to de-dupe all sorts of stuff, not just the DDA. But that’s a point at which we might delete a DDA record from a catalog. That’s also not perfect – do you want to delete a DDA record if the only other access is subscription? I dunno. We’ve ignored the complexities and just zapped them, I think.

At the moment, our DDA profile(s) are all on ProQuest Ebook Central, where we do have a significant ebook subscription. PQEC automatically removes the subscription titles from the DDA list. Our profiles are managed by Coutts/OASIS, and we upload our holdings to them on a regular basis. (Maybe weekly? Monthly?) In theory, this allows Coutts to help us with the de-duplication process. I haven’t actually studied how well that part’s working; my impression is that a lot of the automated processes involved there are still evolving.


B.     and

C.     We load all our ebooks into our primary catalog, regardless of source or license model. We make exceptions when we can’t get our hands on quality MARC, though – if all the records are crap, we just make one database-level catalog record for the whole collection.

DDA-specifically, our Ithaca campus uses a separate catalog from our NYC Med campus. From the larger, Cornell-wide DDA plan, I believe the Med campus only loads records specific to their own subject areas. But Med also has a small DDA profile of hand-picked extremely expensive books – we don’t load the unpurchased ones to the Ithaca catalog, to minimize the chances that Ithaca’s larger population might spend quite a lot of Med’s money. However, it’s easy to do that because it’s a separate profile – we would have to take an extra action to load those records. If they were just included in the main DDA profile and mixed in with all the rest, we probably wouldn’t bother trying to keep them out.


D.     We both get MARC from Serials Solutions/Intota and from various other sources; we have a mixed policy on how many links go on a record:


a.      The SS/Intota records have multiple links to different platforms by default; that’s cool, we keep that.


b.      For locally-loaded batches we obtain elsewhere, it’s a nightmare to try to put all the links on a single record, especially if some of them are subscription and vulnerable to withdrawal.  We do not try to put those on the same record. Any record that has been loaded in batch, and may need to be updated/deleted in batch, is off-limits for adding links to other platforms.


c.      For ebooks purchased manually title-by-title, we keep multiple links to perpetual access titles on the same record; it’s easier for staff and patrons both.

Very, very rarely, we might add a perpetual access link that originally came from (a) or (b) to the manually-created record, if we notice that patron confusion exists – for example, if long ago we bought a 1-user ebook for a title that we now also own in unlimited-user on another platform, and we’re getting huge turnaways on the 1-user ebook. We’ve really only had to do that a handful of times, though.

[Thumbnail2]
Heather Shipman
E-book Acquisitions and Management Specialist
110 Olin Library, Cornell University
Heather.shipman at cornell.edu<mailto:Heather.shipman at cornell.edu> ; 607-254-1499


From: Eril-l [mailto:eril-l-bounces at lists.eril-l.org] On Behalf Of Melissa Belvadi
Sent: Monday, January 22, 2018 11:38 AM
To: eril-l at lists.eril-l.org
Subject: [Eril-l] questions about ebooks: dda, etc., catalogue/other marc/holdings management

Hi, all.

We're still struggling with managing our increasingly complex collections of ebooks under various purchase models, and are wondering whether we're trying too hard to be "perfect".
So I'd like to take a very informal survey to get a sense of what other academic libraries are doing.

A. If you have one or more DDA programs:

1. do you periodically weed these or just let the total list keep growing? If you weed, what are your criteria, e.g. latest 3 pub years only, or something like that?

2. If you also have a major ebook subscription package, do you routinely make an effort to remove DDA records when those same ebooks are added to the subscription package? If so, do you do that with every "additions" load or only on some less often periodic cycle?

B. We have heard that some libraries don't load at least some ebook collections into their catalogue at all but just load them into their discovery service.

Which of the following would you NOT load into your catalogue and where do you make them available in that case?
1. large aggregate subscription ebook packages (like the 150K+ Proquest and EBSCO ones)
2. smaller publisher-specific, often subject-specific collections
3. EBA packages

C. Does your decision not to load some ebook records into your catalogue hinge on ownership versus subscription, or some other factors?

D. When you do have a particular ebook from multiple platforms and are loading the records into the catalogue, do you attempt to merge them into a single MARC record, or keep them separate?

Thanks for consideration and sharing your input on these questions!


Melissa Belvadi
Collections Librarian
University of Prince Edward Island
mbelvadi at upei.ca<mailto:mbelvadi at upei.ca> 902-566-0581
my public calendar<http://www.google.com/calendar/embed?src=mbelvadi%40upei.ca&ctz=America/Halifax&mode=week>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.eril-l.org/pipermail/eril-l-eril-l.org/attachments/20180122/be1725ef/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 2094 bytes
Desc: image001.jpg
URL: <http://lists.eril-l.org/pipermail/eril-l-eril-l.org/attachments/20180122/be1725ef/attachment-0001.jpg>


More information about the Eril-l mailing list