[school] Help a sick MLIS student out
Here is what I have for my paper so far (a lot of it is very rough). To reiterate, the point is to take a stand on automatic indexing (ala Google) and manual cataloging (ala traditional library science). Please to critique, comment, say anything that might get the wheels turning again in my cold-addled brain.
Google Book Search aims to “organize the world’s information and make it universally accessible and useful” by allowing users to “search the full text of books to find ones that interest you and learn where to buy or borrow them.” While this is an innovative and exciting development, I don’t believe that the Google search mechanism should (or will) replace the manual cataloging of materials. Rather, these two powerful roads to information should merge to provide multiple access points for users.
Every Google user has experienced both the power and the failure of the results that it returns. A well-crafted string of keywords can result in links to specific, relevant information. Let’s talk a bit about what a “well-crafted query” means. The more precise and detailed you make your query, the better the results. This begs the question, “Do all users know how to construct a solid query?” If they don’t, are they left behind in the search world? What about users who are searching for information but aren’t sure of the keywords for which to search?
Keyword searching is very useful to users performing a known item search. It becomes less useful when trying to locate information about which little is known.
Consider my midterm paper on the Huntington Library’s art collection, which was a critique of the search functionality of a Web resource. The results of that research showed that Web sites that don’t have good search functionality are frustrating and in some cases unusable. An example that I used in my paper was a search for the keyword “bronze” on the Huntington Library Web site. My intent was to find sculptures housed at the Huntington Library that were made of bronze. The Huntington’s Web site provides only full-text search for keywords (as Google Book Search does). My search resulted in works made of bronze and works that had the word “bronze” in the title or description. There was no way to browse the collection. Had I been able to browse, I could have navigated through a hierarchy of Sculptures > > Bronzes rather than performing a keyword search for “bronze.” This would also have shown me the materials used for all of the sculptures in the Huntington’s collection. Perhaps I was looking for all sculptures made of metal and all I knew was bronze. The discovery of other metallic materials may not be possible without those “multiple organization layers.”
Deanna Marcum contrasts the “multiple organization layers”2 of library online catalogs versus the immediate results that are returned using Google. Marcum feels that quick results are more important to users than precise results that require some work from the user. This is the key issue of the differences between automatic indexing (Google) and manual indexing (traditional cataloging). Without having those multiple organization layers, users can’t browse collections. If users can’t browse collections, they may never discover resources that meet their needs. They may never connect with keywords that would help to refine their search results. That Marcum views these organization layers as barriers to information seekers reveals a level of misunderstanding that is astounding for someone who holds the position of Associate Librarian for Library Services at the Library of Congress.
A lot of manual, cataloging-type work goes into Web sites that feature superior keyword searching functionality. Variant terms are accounted for and associated with preferred terms. Metadata is added and edited regularly to account for the ways in which users are searching for information. While Google does a great job considering the sheer volume of sites on the Web, it is not as usable as a clearly defined collection that is manually serviced. Google Book Search is such a collection. Many users are going to want to use GBS like they would a virtual library. A level of functionality is going to be expected that isn’t going to be there in Marcum’s vision.
Another argument that supports the need for manual cataloging is the contrast between users who are looking for scholarly resources versus those looking for quick information. Google is a wonderful resource for answering a quick question, such as getting information on a specific law. It becomes less usable when researching for detailed, factual information, such as finding past cases that establish precedence for a current case.
While being able to “search every word from a book’s dust jacket to its back-of-the-book index” would certainly be userful, it does not eliminate the need for “intermediate-level descriptions.” 2 This would present a flat search environment to users and ignore the value of a hierarchical browsing structure.
Another troubling assertion of Ms. Marcum’s is that it is not surprising “that many students just go Googling instead of to the library, virtual or physical, and use whatever turns up first in the key-word search.”2 This seems to be an acceptance of sub-standard research methods. One goal of a college education is to learn how to locate accurate, factual information.
I also wonder about the work that has already been done in terms of cataloging resources. Are these being shared with Google? If not, they should be. A partnership between libraries and Google would best support the needs of all users. Ms. Marcum is doing a disservice not only to libraries and users, but also to Google Book Search. Google needs the advice and direction of librarians in order to make this virtual library usable. Instead of providing that level of assistance, Ms. Marcum is putting the decisions that are best made by professionals into the hands of those that have no training in the discipline.
“If such an expenditure produces great benefits for the Library of Congress, libraries across the country, and others around the world, then we can justifiably argue that the forty-four million is well spent. But in the age of digital information, of Internet access, of electronic key-word searching, just how much do we need to continue to spend on carefully constructed catalogs?” I find it interesting that Ms. Marcum questions whether cataloging is worth the cost with the explosion of information architecture in the private sector. Information architects specialize in creating facets and hierarchies that are built on principles first established by libraries. The cost of building and maintaining Web sites is enormous and a large amount is earmarked for just the types of cataloging that Ms. Marcum is addressing. Businesses realize the benefits of supplying users with multiple access points through such constructs as site maps, browsable views, and facets, in addition to keyword search. A quick look through the Findability section of Boxes and Arrows (an online information architecture resource) reveals articles that deal with modes of information seeking (not just search), hierarchical metadata and taxonomy, sitemaps, site indexes, synonym rings, authority files, facets, and controlled vocabularies.
With all the time, money, and effort that has already gone into cataloging the world’s books, isn’t it just as costly to abandon the effort? What if Ms. Marcum is wrong in her assumptions and every digitized book has to be revisited to add the appropriate metadata that libraries used to provide? Won’t that be even more costly?
“Providing leadership in the cataloging and classification of book collections is no longer regarded as necessary at the top levels of LC administration. Cataloging and classification are themselves regarded as replaceable by keyword searching, relevance ranking, and user-referral mechanisms.” Every person working in a technology related position knows that keyword searching on its own produces very little benefit for end users. This is why so much effort goes into developing controlled vocabularies, authority files, and other ?? to connect users with the information that they need. It’s interesting that the business world seems to value traditional library goals while they are being abandoned by the very institution that developed them.
Private businesses are putting a lot of money into creating faceting and hierarchies on the Web and in internal databases that come directly out of library science. It’s ironic that the private sector is adopting library practices that Marcum seems to want to relinquish to Google.
Deanna Marcum addresses the issue of “how digital-era students work.” She states repeatedly that college students do not want to go to physical libraries, and instead they use Google first (and often only). Her use case is built around an undergraduate student doing research for a term paper. I would argue that Ms. Marcum is not addressing all users of the library. Does she feel that the needs and practices of undergraduate college students should dictate library and Web usage for everyone? What about lawyers, doctors, and other professionals who rely on research in order to do their work? One would hope that doctors aren’t Googling and using the first hits to diagnose and treat patients.
Google Book Search does not have every book that is in existence scanned. With the copyright issues that they face as well as the vast number of books that exist, it is questionable as to whether this goal will ever be achieved. Books that are not included are non-existent to users of the system.
Another goal of Google Book Search is to point users to resources in their local libraries. Ms. Marcum advocates storing
While the Internet is certainly one avenue, there are many others that may be of more
use.
Google Book Search aims to “organize the world’s information and make it universally accessible and useful” by allowing users to “search the full text of books to find ones that interest you and learn where to buy or borrow them.” While this is an innovative and exciting development, I don’t believe that the Google search mechanism should (or will) replace the manual cataloging of materials. Rather, these two powerful roads to information should merge to provide multiple access points for users.
Every Google user has experienced both the power and the failure of the results that it returns. A well-crafted string of keywords can result in links to specific, relevant information. Let’s talk a bit about what a “well-crafted query” means. The more precise and detailed you make your query, the better the results. This begs the question, “Do all users know how to construct a solid query?” If they don’t, are they left behind in the search world? What about users who are searching for information but aren’t sure of the keywords for which to search?
Keyword searching is very useful to users performing a known item search. It becomes less useful when trying to locate information about which little is known.
Consider my midterm paper on the Huntington Library’s art collection, which was a critique of the search functionality of a Web resource. The results of that research showed that Web sites that don’t have good search functionality are frustrating and in some cases unusable. An example that I used in my paper was a search for the keyword “bronze” on the Huntington Library Web site. My intent was to find sculptures housed at the Huntington Library that were made of bronze. The Huntington’s Web site provides only full-text search for keywords (as Google Book Search does). My search resulted in works made of bronze and works that had the word “bronze” in the title or description. There was no way to browse the collection. Had I been able to browse, I could have navigated through a hierarchy of Sculptures > > Bronzes rather than performing a keyword search for “bronze.” This would also have shown me the materials used for all of the sculptures in the Huntington’s collection. Perhaps I was looking for all sculptures made of metal and all I knew was bronze. The discovery of other metallic materials may not be possible without those “multiple organization layers.”
Deanna Marcum contrasts the “multiple organization layers”2 of library online catalogs versus the immediate results that are returned using Google. Marcum feels that quick results are more important to users than precise results that require some work from the user. This is the key issue of the differences between automatic indexing (Google) and manual indexing (traditional cataloging). Without having those multiple organization layers, users can’t browse collections. If users can’t browse collections, they may never discover resources that meet their needs. They may never connect with keywords that would help to refine their search results. That Marcum views these organization layers as barriers to information seekers reveals a level of misunderstanding that is astounding for someone who holds the position of Associate Librarian for Library Services at the Library of Congress.
A lot of manual, cataloging-type work goes into Web sites that feature superior keyword searching functionality. Variant terms are accounted for and associated with preferred terms. Metadata is added and edited regularly to account for the ways in which users are searching for information. While Google does a great job considering the sheer volume of sites on the Web, it is not as usable as a clearly defined collection that is manually serviced. Google Book Search is such a collection. Many users are going to want to use GBS like they would a virtual library. A level of functionality is going to be expected that isn’t going to be there in Marcum’s vision.
Another argument that supports the need for manual cataloging is the contrast between users who are looking for scholarly resources versus those looking for quick information. Google is a wonderful resource for answering a quick question, such as getting information on a specific law. It becomes less usable when researching for detailed, factual information, such as finding past cases that establish precedence for a current case.
While being able to “search every word from a book’s dust jacket to its back-of-the-book index” would certainly be userful, it does not eliminate the need for “intermediate-level descriptions.” 2 This would present a flat search environment to users and ignore the value of a hierarchical browsing structure.
Another troubling assertion of Ms. Marcum’s is that it is not surprising “that many students just go Googling instead of to the library, virtual or physical, and use whatever turns up first in the key-word search.”2 This seems to be an acceptance of sub-standard research methods. One goal of a college education is to learn how to locate accurate, factual information.
I also wonder about the work that has already been done in terms of cataloging resources. Are these being shared with Google? If not, they should be. A partnership between libraries and Google would best support the needs of all users. Ms. Marcum is doing a disservice not only to libraries and users, but also to Google Book Search. Google needs the advice and direction of librarians in order to make this virtual library usable. Instead of providing that level of assistance, Ms. Marcum is putting the decisions that are best made by professionals into the hands of those that have no training in the discipline.
“If such an expenditure produces great benefits for the Library of Congress, libraries across the country, and others around the world, then we can justifiably argue that the forty-four million is well spent. But in the age of digital information, of Internet access, of electronic key-word searching, just how much do we need to continue to spend on carefully constructed catalogs?” I find it interesting that Ms. Marcum questions whether cataloging is worth the cost with the explosion of information architecture in the private sector. Information architects specialize in creating facets and hierarchies that are built on principles first established by libraries. The cost of building and maintaining Web sites is enormous and a large amount is earmarked for just the types of cataloging that Ms. Marcum is addressing. Businesses realize the benefits of supplying users with multiple access points through such constructs as site maps, browsable views, and facets, in addition to keyword search. A quick look through the Findability section of Boxes and Arrows (an online information architecture resource) reveals articles that deal with modes of information seeking (not just search), hierarchical metadata and taxonomy, sitemaps, site indexes, synonym rings, authority files, facets, and controlled vocabularies.
With all the time, money, and effort that has already gone into cataloging the world’s books, isn’t it just as costly to abandon the effort? What if Ms. Marcum is wrong in her assumptions and every digitized book has to be revisited to add the appropriate metadata that libraries used to provide? Won’t that be even more costly?
“Providing leadership in the cataloging and classification of book collections is no longer regarded as necessary at the top levels of LC administration. Cataloging and classification are themselves regarded as replaceable by keyword searching, relevance ranking, and user-referral mechanisms.” Every person working in a technology related position knows that keyword searching on its own produces very little benefit for end users. This is why so much effort goes into developing controlled vocabularies, authority files, and other ?? to connect users with the information that they need. It’s interesting that the business world seems to value traditional library goals while they are being abandoned by the very institution that developed them.
Private businesses are putting a lot of money into creating faceting and hierarchies on the Web and in internal databases that come directly out of library science. It’s ironic that the private sector is adopting library practices that Marcum seems to want to relinquish to Google.
Deanna Marcum addresses the issue of “how digital-era students work.” She states repeatedly that college students do not want to go to physical libraries, and instead they use Google first (and often only). Her use case is built around an undergraduate student doing research for a term paper. I would argue that Ms. Marcum is not addressing all users of the library. Does she feel that the needs and practices of undergraduate college students should dictate library and Web usage for everyone? What about lawyers, doctors, and other professionals who rely on research in order to do their work? One would hope that doctors aren’t Googling and using the first hits to diagnose and treat patients.
Google Book Search does not have every book that is in existence scanned. With the copyright issues that they face as well as the vast number of books that exist, it is questionable as to whether this goal will ever be achieved. Books that are not included are non-existent to users of the system.
Another goal of Google Book Search is to point users to resources in their local libraries. Ms. Marcum advocates storing
While the Internet is certainly one avenue, there are many others that may be of more
use.