Posting to Autocat
On Sat, 23 Oct 2010 02:40:01 -0400, Hal Cain wrote:
>On Fri, 22 Oct 2010 13:50:49 -0500, Mike Tribby wrote:
>>As to Amazon’s willingness to make changes in their procedures, they have definitely been apprised of the shortcomings in their search results that we have discussed on this forum as well as the problems their publishers have getting edition and ISBN changes into their system. I don’t know of any complainants who have received more than perfunctory answers about these issues, but that doesn’t mean no one has.
>This also means they can’t (or can’t truthfully) say that “nobody has complained” about their inadequacies.
People always complain–they have complained about the library catalog for years for all kinds of reasons with few results, and the fact they have been abandoning it for other options speaks volumes.
But returning to FRBR: has anybody ever seen any research or evidence that people want FRBR-type displays? Remember, OCLC found that less than 20% of everything in their database has some kind of variant that would make any difference at all with FRBR. I wonder what percentage of that is for obsolete textbooks that almost no one will ever look at again? In the complaints I have read about the metadata problems in e.g. Google Books, they discuss accuracy of data, but I have never seen a word about not being able to get works, expressions, manifestations and items.
Additionally, I have never seen any attempts that have shown that FRBR-type displays cannot be made with the records we have right now. I personally see no reason why they could not be generated from records as they are now.
I still maintain that above all, we should be working to provide what people want. If research shows that people say they need FRBR-type displays so badly, then OK: figure out how to do it in the most efficient ways possible, but I honestly doubt if the results would come out this way.