FRBR statistics

Attending the FRBR Workshop at OCLC today, one of the most common questions/assertions is that the minority of bibliographic records need FRBR to group them.  That's an easy impression to get from some of the statistics we've released (e.g 78% of the works in WorldCat have a single manifestation), but at least slightly misleading.  Here's a statistic that is closer to what a typical user sees:  63% of the holdings in WorldCat are associated with works with two or more manifestations.

Update: Sorry, I got that wrong.  Here are the latest figures:

Out of 54,830,689 manifestations we found 43,794,883 work sets
88% of the works in WorldCat have a single mainifestation
55% of the holdings are in works with more than one manifestation

--Th

Comments

[cliff]

I think that some of the raw numbers, while still potentially misleading, might also have an effect in terms of the way that people view FRBR. For example, if 78% of the works in WC have a single manifestation, that still leaves roughly 4.5 million records (22% of 59 million) that represent works with more than 1 manifestation. While technically a minority, 4 million records is nothing to sniff at.

Perhaps a better way to handle this assertion in some cases is to agree with it. I'm thinking specifically of identifying rare items/last copies/etc. FRBR can be used as a valuable tool in weeding out items (manifestations) that seem to be unique - e.g., a single record with a single holding - but that in actuality are part of a widely-held group of items (work).

More importantly, FRBR provides a conceptual model for reasoning about these kinds of things that I don't think was present prior to its existence. That's my two cents...

~cliff

[Zoltan Tomory]

All this number stuff is fine and well, but if a work does not yet have a second manifestation, this is not an adequate predictor of whether there will be a second or subsequent manifestation by an arbitrarily chosen date in the future. I think there is a big difference between where we are and where we will be.

Consider for every title that now exists there will someday be a version from IDC, Chadwick-Healey, a free version on the web, a version you have to pay for on the web, a reprint from Bishen Singh Mahendra Pal Singh, editions from dozens of publishers, the photocopy "edition" and a file with it in ASCII text from Project Gutenberg.

If we do not prepare ourselves now by adopting rules, records and online systems that make our lives simpler and make the mess easier for patrons to navigate and simplify getting a copy to patrons for ILL people, we will be inundated by a maelstrom of paperwork, and FEMA will not be there to pick up the pieces.

The best way to judge where on the interstate you will be is not by peering out at the cornfield over your shoulder! Unless your car is stopped, of course. However, the world we live in is very much a car in motion--speeding for that matter.

The sky is not falling, but the database is exploding. Better deal with it gracefully and now as we seem to be accelerating.