Alexander Johannesen wrote:
> To put it a bit cheeky, most of them would say Google can't do fielded
> searching, which is true, but you can simulate it, and what's more, in
> order for fielded searching to work the contents of those fields must
> be accurate and effective. And, well, MARC meta data has a few global
> mailing-lists dedicated to fixing errors, explain peculiarities and
> work with it. Most librarians show examples of stuff they already
> know, stuff that wouldn't fail in a kind of straw-man way.
>
> Put differently, if you let Google have 100% access to everything in
> your catalog, which would be better for research?
Naturally, there are many mistakes in what human beings do, but it doesn't follow that if a computer does it, there are no mistakes either. More to the point though, we should keep in mind that Google very well may have 100% access very soon to everything in our catalogs through the Google Book Search. Already it is changing the way people find materials in our collection: I tell my students to search Google Books by keyword and if it's not available for free, to see if the book they want is available here in Rome somewhere. It is one tool of many that can be used, including the catalog.
But it still needs to be proven that Google can do a reliable search for something by Tolstoy or Confucius. This is child's play in a catalog, but that is because so much work has gone into it behind the scenes. Yet, this type of search is simple compared to finding U.S. soldiers' memoirs of WWI. Before we onclude that it's just not that important to get reliable results--that whatever Google throws out is OK--there should be a much larger discussion among lots of interested groups.
> >>�> But the point is that no
> >>�> library is interested in collecting free e-books.
> >>
> >>�Why not?
> >
> > First reason for not collecting them (at least I believe), free e-books
> can disappear tomorrow
>
> Isn't the point of collecting them so that if they disappear in one
> spot you still got a copy?
I didn't understand the first time. Yes, libraries should do this or at least work with the Internet Archive much more closely than they do now. That way, they could avoid many of the technical problems. I've never really understood why libraries haven't done this.
> > But I think the main reason is that it bypasses normal library workflows.
>
> And with that I think the word "Exactly!" shall close my emailing for
> the day. If it isn't clear enough, I think this is the worst possible
> reason ever! Exactly! :)
I wouldn't dismiss this so quickly. I think that if a system could be worked out that allowed selectors to be notified in some sort of way, in essence replicating their normal workflows (and I'm sure this would be possible), there would be much more interest than there is now. I think Intute and Infomine would be an excellent starting point and could show the future in many ways: an item selected once by an expert, a record made once--and well--by another expert, available to everyone. With more selectors (including scholars), and catalogers working in unison from around the world, plus adding Web2.0 powers, I think it would be a fabulous model for the future library community.
Jim Weinheimer
Received on Mon Mar 02 2009 - 07:01:05 EST