July 31, 2007

Every book ever published in every language

Writing about web page http://news.bbc.co.uk/1/hi/magazine/6924022.stm

The BBC article raises some interesting issues about the Open Library project, contrasting it with Google’s own library project and the way information about books on the web often links back to Amazon.

It will be interesting to see if the concept of the Open Library takes off. I think there are dangers of malicious editing & spam as with any collaborative project, and there is a possibility that the owners of Open Library will try to make profit from it in some way in the future. But then if they invested in the database design as the article describes, and they are having to police it and protect it, then maybe they deserve some of the profit. Which leads me back to the question of why we don’t just use Google’s project or Amazon anyway?

I think it would be far better to allow people to edit library catalogues that already exist, to contribute their own reviews and tags on top of the professionally created information, but not in place of it.

Library catalogues already adhere to an international standard of machine readable bibliographic data (known as MARC) and are therefore theoretically cross-searchable. All you would need to achieve the aim of collating all the books ever published is a cross-searching platform, and a selection of libraries wide enough to cover every book ever published… and an eternity to wait for the results of your search :-) or else a metadata harvesting tool with access to the library catalogue records (via the OAI protocol) and a super-huge database to store all the records in, on a super-fast machine to return results to you quickly.

Not that I would want to search the records of every book every published in every language anyway. Imagine the time it would take you to find what you were actually looking for. Imagine how overwhelmed you would feel once you got the results set. Only someone with a very precise query and the advanced information skills to express it accurately would be able to handle such a search with any degree of accuracy.

It’s not ever likely to be necessary for someone to search every book ever published, surely? That is why we have small branch libraries and subject libraries, etc and ways of selecting what goes into a library in the first place. The library acts as a filter for you, and its content reflects the interests and needs of its patrons… which is why I like the idea of putting users’ reviews and tags onto library records.

Wouldn’t you rather read a review from someone who has studied on the same course as you, than from a random person on the Amazon website or Open Library?

If you weren’t finding what you needed from one library, you could then look for another library with a different specialty or focus. So a search for a library, perhaps based on the libraries’ own collections descriptions (Libraries have standards for those, too) would be a good place for you to identify which catalogues you could/should be searching (whether separately or through a cross-searching platform). What you would need alongside that is access to those other specialist libraries and their content. Which is when reciprocal visiting/loan arrangements and digitisation initiatives become interesting.

If Open Library or Google were ever to succeed with creating a collection of every book ever published in every language, who would use it and how? In order to simplify the search process, and in order for people to handle the number of results returned, someone somewhere is making decisions for you about what you should find… and in Google’s case not being very open about how they do it. Wouldn’t you rather that that someone was your friendly librarian who you can speak to, who can explain how they chose a particular book or collection, than someone you’ve never heard of doing techie things with algorithms that you don’t understand?

- 2 comments by 2 or more people Not publicly viewable

  1. Chris May

    Some of your concerns sound a bit like the ones that were voiced about search engines pre-google, to me. In the days before google the web was full of link directories, hand-maintained and optimised by people, which had very high quality results but limited scope, and went out of date rapidly. No-one believed that you could do away with the people who maintained those directories, because a computer wouldn’t be as accurate.

    What Google have shown (in Web search, not in book search yet) is that if you can be convenient, fast, and more-or-less right, you’ll tend to win out over a service (like a human being) which is less convenient, less fast, but more right.

    What google don’t yet have (as far as I know) for book search, is any equivalent to PageRank – a means of getting a range of opinions on the value of a piece of content (“I found this page useful enough to link to it”). Whether they can (or have) come up with anything comparable for books, I’m not sure.

    31 Jul 2007, 14:28

  2. Jenny Delasalle

    I suppose the equivalent to the ranked search would be when Amazon picks up on the titles that others have bought, and recommends books to you on that basis. So Amazon has an advantage over Google and Open Library in that sense.

    Google does make searching the web very simple and efficient, which is no bad thing. Often in the academic sphere, however, the most appropriate source is part of the hidden web: ie it is not indexed by Google or one of the major search engines. It is surprising how few students have even heard of Google Scholar or tried to use the Google advanced search form, which are more appropriate to academic needs, even if still not reaching those hidden web resources.

    As both fast food and fine restaurants exist and thrive, so there is a place for both Google and human-compiled directories on the web. Library subscribed resources reach beyond the needs of the masses. The number and type of human compiled directories are growing, with increasing Web 2.0 enabled services. I like using del.icio.us when looking for web pages rather than Google, especially looking for pages that my network have tagged. Indeed, the way Google works involves human input (someone clicked on the results, and someone created the links to that site), its just that we don’t know who from or how it influences the results!

    A library serves and involves a community of people. Web 2.0 is all about user participation, and it strikes me that this is the opportunity for such communities to really make libraries their own & a part of the web, rather than a chance for another huge information source to become all-powerful, prodigious and profitable – especially if it is off the back of those who contribute to it.

    However, I wait to see whether any of these “everything ever published” endeavours are successful… they need to create something fast and convenient first. Even if they manage that, there should still be a place for a Web 2.0 library community, its just that I won’t be the first to contribute to Open Library!

    31 Jul 2007, 15:12

Add a comment

You are not allowed to comment on this entry as it has restricted commenting permissions.

July 2007

Mo Tu We Th Fr Sa Su
Jun |  Today  | Aug
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31               

Search this blog

Most recent comments

  • Are you guys still pursuing the VLE route? There has been some advance made by the open university o… by Alejandro Chiner Arias on this entry
  • Well, yes, there are all sort of fascinating stories about the human factors in the fate VLE develop… by on this entry
  • estimating the different costs involved in each particular case Is there anyone anywhere capable of… by Robert O'Toole on this entry
  • Thanks for the link to Shared Copy. I've been looking for a site like that. The sites I've found bef… by Emma on this entry
  • I read about these "knol"s recently, and I'm not too sure what they will do. As with all these techn… by Jenny Delasalle on this entry

Blog archive



UK Copyright Swicki

check out the UK copyright swicki at eurekster.com
Not signed in
Sign in

Powered by BlogBuilder