Re: BIBFRAME – draft data model for bibliographic data

Posting to Autocat

On 11/28/2012 09:59 PM, McDonald, Stephen wrote:

It is ludicrous to describe the _addition_ of data as _reducing_ access. My response if someone complained that they weren’t getting everything when they searched by “editor” is that a year ago they wouldn’t have been able to search that way at all. Yes, everyone here realizes that adding these fields to new records is not going to do much for the old records. What would you have us do instead? Would you have us not improve what we can? If we don’t start somewhere, we won’t have _any_ records where you can search by these facets.

Do you seriously believe that starting to add new access points is worse than not adding them at all? Do you think it was a mistake to start adding 776 fields for alternative versions, even though most previous records didn’t have them? Or to start adding genre headings?

For the public it is definitely the case that adding relator codes and searching on those codes reduces access. I guess you haven’t read or listened to my podcast, so let me try another tack.

Let’s imagine a card catalog with authors in one run of cases and subjects in another run of cases. I know that is difficult for some out there who have never worked with a card catalog, but please try.

There are a few hundred thousand cards in these runs. All personal names have been added for the last 100 years into the run of authors. Suddenly, someone says that people need to be able to search film directors separately, so that means moving the cards from the run of personal names into a new run of film directors.

Of course, that means a lot of work. So, a searcher comes up to the section marked “film directors”, sees very little and says, “I thought this library had a great collection of films. But there is nothing directed by John Huston.” What does the librarian say? The librarian says, “Well, we do have multiple versions of everything directed by John Huston but those cards are still in the other run.” The searcher says, “How long will it be before it’s fixed?” “Hard to say, but we are working on it. Probably at least a decade or so because we don’t have money for a special project and we can only work on it during our free time. So until it’s fixed you have to search both files.”

At least this searcher would have been able to talk to a librarian and find out that he had to search both files, but I am sure he would not be very impressed with the library. Everyone could also physically see the enormity of the task. How will it work today when very few people even ask reference questions? They will never know they need to look in both places. Should catalogers care that the catalog is giving wildly false search results? It seems as if we aren’t supposed to care because it is all done for the good for the users.

This attitude I utterly reject. As I mentioned in my podcast, the old attitude that people will just have to deal with it is completely backwards today. 20 years ago, the public had no real choice but today the public has lots of very attractive choices, where those in charge of those new alternatives really and truly care about the people who use their systems. Besides, people will find out pretty quickly and will they say, “Well, it’s no big deal that the search results are bogus. I don’t care about that anyway.”

I don’t think so. I think people will complain. And loudly.

So when you ask “Would you not have us improve what we can?” I reply yes, so long as it is a real improvement for the reality of the user. Not just in somebody’s theory or so that someone can point to it on a graph–but in everyday life. Improving things for the user does not mean hiding 90+% of everything that they can use. That is a misuse of language. Therefore, the reality of RDA and FRBR will definitely not be an improvement for the public and will definitely lead to decreased access. That needs to be understood and accepted if any progress is to be made.

Are there any alternatives? Of course–there are always alternatives, but first we need to open our minds from the stultifying influence of FRBR–as if that is all that we can aim for. Google doesn’t aim for it, Yahoo doesn’t, the rest of the information world doesn’t. Luckily for the cataloging world, for those who believe FRBR is so important, modern software allows it right now. Right now.

What can catalogers do? Once such a question is asked seriously, matters then turn interesting. I have come up with several suggestions and others have also, but the very first step is to find out what the public wants–not to try to force-feed everyone with predigested dogma that has never been seriously questioned or researched–but to find out what the public really wants. It may turn out that they want FRBR, but I seriously doubt it.

In fact, my own suspicion is that this is exactly the reason why there has been no research done on the public: everybody realizes the results would show that people want something entirely different from what FRBR or RDA envision. Just thinking about our own experiences searching Google is enough to show that.