Elaine Sloan, Boise Public Library · 1:30–2:30 PM · Houston Room · General Track · Wednesday, April 15
Speaker: Elaine Sloan
How an 11-library Polaris consortium in Idaho walked away from a six-figure OCLC contract, survived the collapse of BTCat when Baker & Taylor filed for bankruptcy, and landed on BookWhere Suite — a Z39.50 copy cataloging client that dropped their total cataloging costs to five figures. A candid, at times fiery session covering the practical realities of cataloging without OCLC: ILL trade-offs, staff retraining, record quality, macro limitations, and the legal flashpoint of OCLC record ownership.
Elaine’s consortium of 11 Polaris libraries in Idaho was paying six figures annually to OCLC. The cost was the primary motivation to leave. Their state does not have a statewide library lending service, which simplified the ILL question.
ILL was the biggest hurdle. The consortium libraries performed cost analyses and found that the number of patrons who actually used ILL was a much smaller portion of the cardholder base than expected. Most libraries in the consortium made the controversial decision to discontinue ILL entirely.
One library took a creative approach: instead of paying $5+ each way to mail items, they redirected their ILL budget into purchasing items directly — still serving those patrons, just buying the $20 book rather than spending $10+ to borrow it.
The consortium also has courier routes between all 11 libraries, so members can still borrow from each other without OCLC. The few libraries that kept ILL service search the online WorldCat database and send requests directly to holding libraries.
Once ILL was off the table, leaving OCLC became much easier.
| Date | Event |
|---|---|
| June 2025 | OCLC contract ended; switched to BTCat (Baker & Taylor) |
| Oct 2025 | Baker & Taylor announced cessation of library services operations |
| Nov 2025 | ~1 month to evaluate alternatives (SkyRiver, BestMARC, BookWhere) |
| Dec 2025 | Went live on BookWhere Suite |
| Mar 2026 | Baker & Taylor filed Chapter 11 bankruptcy |
BTCat had incredible macro functionality with conditional statements — something no other vendor has replicated. The consortium only used BTCat for about six months before Baker & Taylor shut down. The loss was clearly felt: “All right, baby cat” — an affectionate play on “BTCat.”
From the research that Elaine’s consortium did, no other vendor offers customizable macros the way BTCat did. It remains the feature they miss the most.
With roughly one month to decide, the consortium evaluated three products:
A web-based product. The consortium administrator preferred this option, but when staff tested it, they “kind of hated it” — they found it clunky, hard to use, and the hit rate wasn’t great. Most staff were still accustomed to client-based interfaces (Connexion, BTCat), so the web-based approach didn’t resonate.
SkyRiver hadn’t been actively selling until BTCat announced discontinuation. Its interface looks and acts like Connexion client, which was a point in its favor. However, several issues emerged:
BookWhere Suite had the best hit rate of anything tested — the speaker thinks it’s actually better than BTCat. Its client-based interface was familiar to staff who were already using Polaris client rather than Leap. And with Z39.50 access to thousands of libraries, it offered far more variety of records than SkyRiver’s limited pool.
BookWhere connects to approximately 2,400 libraries over Z39.50 (the vendor advertises 3,000+ targets in their current marketing). Searching all 2,400 at once is impractical — too many hits, too slow. Instead, the consortium created curated database groups:
The search interface “looks like it’s from the 1990s” — not pretty, but it works. You can search by virtually any MARC field, which is both extensive and daunting.
BookWhere ranks search results by RDA score and MARC 21 score, displayed as colored boxes alongside a numerical ranking. The consortium guideline is to choose records scoring 50 or above when possible. This was similar to BTCat’s ranking system, which eased the staff transition.
BookWhere Suite does have macros, but they lack conditional statements — the key feature that made BTCat’s macros special. BookWhere macros are basic “remove this” / “add this” operations, with notable limitations:
Coming from BTCat’s conditional macros, this was the most painful part of the transition.
The built-in editor is “as utilitarian as it gets.” It works, but editing leader fields is a nightmare — there are no positional guides, so it’s very easy to get a space in the wrong place and throw everything off. The recommendation is to export records into Polaris for leader editing, where the positional guides make it much clearer.
On the plus side, you can open multiple records simultaneously in the editor, drag-and-drop fields between them, and do comparison work. This is particularly useful for copy catalogers piecing together records.
Configuration is stored in XML files that can be distributed to workstations — staff drops them into a folder. It’s not as seamless as web-based administration, but at least each user doesn’t have to change individual settings. Managing this across ~50 workstations in a consortium is painful but workable.
Licensing requires a per-PC key that must be deactivated and reactivated when hardware is replaced. Individual licenses run in the hundreds of dollars; site licenses in the thousands. The total consortium cost dropped from six figures (OCLC) to five figures — significant savings.
BookWhere Online (the web version) is a completely different product from the Suite — less functionality, no macros, different interface. The speaker was emphatic: “the online version is truly awful.” Crowdsourced advice from catalogers on Facebook confirmed this. If you’re evaluating BookWhere, look at the Suite first.
BookWhere has not been sued by OCLC despite being in business for 25+ years. Their strategy, whether intentional or not, is legally sound: they save nothing and have no record repository. They only facilitate Z39.50 connections between libraries. BTCat, by contrast, had its own “community records” database, which contributed to OCLC’s case against them. BookWhere’s position: “there’s nothing to sue.”
The speaker noted that OCLC record quality has gone down, particularly in the last five years. Because of this, the additional vetting required when using BookWhere isn’t as dramatic a change as it would have been a decade ago.
One practical issue: records retrieved via Z39.50 include more localized fields that OCLC and BTCat used to strip out automatically. The consortium had to beef up their Polaris import profiles to handle fields they “never would have expected to come in.”
The speaker does original cataloging in BookWhere’s MARC Notepad editor, piecing records together. Others in the consortium do it directly in Polaris.
The downside is clear: original records stay local. You can’t share them back to WorldCat the way you could with Connexion. “That was a bummer for me for leaving OCLC,” the speaker acknowledged.
The upside: anyone with your Z39.50 connection open can access your records. “Bibliographic data is not proprietary and should be shared,” the speaker said — a philosophy that informs their willingness to leave their Z39.50 server open for other libraries.
This was the most heated part of the session.
When using BTCat, the consortium ran a macro to remove 035 fields containing OCLC numbers before importing records. After switching to BookWhere, they didn’t remove existing OCLC numbers from their catalog — “maybe we should have.”
The speaker asked Marshall Breeding on Monday (at the conference) about this. His assessment: even removing OCLC numbers wouldn’t matter because the records are still “marked as proprietary.”
The speaker’s response, on the record. She expects the consortium may receive a cease-and-desist at some point, and expressed concern that individual libraries don’t have the resources to fight OCLC in court. OCLC has already prevailed against BTCat and MetaDoor.
Her position: “For people that have stopped using OCLC, those records really don’t belong to them. We should still be able to share them, no matter what OCLC says.”
An audience member suggested starting an alternative: “NOclc” — which got a good laugh.
Loss of familiarity. Staff had been on OCLC for 20+ years and had never used anything else. The other hurdle was retraining staff to evaluate bibliographic record content holistically — many had relied on the presence of an OCLC number as a proxy for quality, which “obviously wasn’t always the case.”
Staff time has gone up overall. It increased when they moved to BTCat (new system) and again when they moved to BookWhere (different system + more careful record evaluation). However, staff actually settled into BookWhere faster than BTCat, likely because the client-based interface felt familiar.
Not exactly. The macros only do basic add/remove operations, and you have to enter remove commands multiple times per field. It’s not a true import profile in the OCLC sense.
The speaker’s library uses Collection Manager for their O’Reilly database — it’s part of the O’Reilly subscription, separate from OCLC, so it still works. “OCLC doesn’t love it because they’re losing part of your subscription, but it is possible.”
The consortium uses Backstage Library Works and has for years — that didn’t change when they left OCLC. “If any of you need authority work, check out Backstage. They’re the best.”
Occasional individual library connections go down, but with thousands of alternatives, it’s never been a real issue. However, setting up client-to-client connections across a consortium can be an IT nightmare — non-standard ports, firewall rules that need to be configured per library. The speaker’s assistant administrator Brad managed this across the 11 organizations.
The consortium looked at MarcEdit but decided it was too complicated for copy catalogers, especially given their tight evaluation timeframe. An audience member noted that Polaris’s built-in Z39.50 client is another option, but the speaker didn’t like that records save immediately to the system with no staging area.
No. BookWhere is purely a connection service — “they don’t offer full cataloging. They are just a connection service.”
Audience discussion: Amazon vendor records “have gotten progressively better, and they’ve gotten better faster than I ever expected.” But the consensus was they’re still poor quality — an audience member’s description was more colorful.
An academic library commenter expressed significant frustration with OCLC: turnaround time on paperwork is “abysmal” (5+ weeks), their county council red-lined OCLC contract terms, and OCLC refused to negotiate. The commenter questioned why they even maintain a contract.
Additional context gathered during the session to supplement the speaker’s presentation.
