Stephen, Google is only one way to create a graph, and their graph is
based on keywords. There are many other ways to create graphs of
knowledge, including using classification, identities (like name
authorities), etc.
Here are three instances of graphs based on library data:
http://www.worldcat.org/identities/lccn-n79-129204
http://openlibrary.org/subjects/philosophy
http://agris.fao.org/openagris/searchIndex.action?query=fisheries
They all display differently, use different data, but they are all based
on graph concepts. (Well, maybe except for WorldCat identities which is
more of a data mined set, but the result is graph-like.) The Agris
search includes information from outside of the library in its results
-- it pulls in data from other information resources, like geographical
and scientific databases and displays this information alongside the
search results and individual articles. The LIBRIS system at the Swedish
national library works similarly, linking out to data sources like
DBpedia using graph technology. The BBC also uses this in some areas of
its web site, pulling in current information from linked science databases.
I can't imagine see the information/creative world as anything but
boundless. Drawing the line at what is in the library makes no sense.
That's an inventory function, not an information function. Sure, you do
want to know what the library has, but that's not the same as "what is
known about this topic?" Especially today.
kc
On 8/3/13 7:46 AM, Stephen Paling wrote:
> Karen,
>
> Sorry, but I'm not buying your explanation. The king of "an unending graph of relationships" is Google, the subject of endless carping by librarians. Look at this thread. Google has been described as predatory, riding the backs of content providers, of only running an algorithm over text (And an OPAC does ~what~, pray tell?), etc. And Google's graph is not limited by physicality.
>
> Now take a look at what librarians tend to suggest. They tend to want to use the same classification and subject heading tools that they've used for physical books. In one of your earlier posts you argued that Google doesn't actually organize information because they don't classify. But what makes a search engine like Google so powerful is that their, yes, algorithms, ignore the boundary of the information object and index the content itself.
>
> So, yes, the organizational tools of librarianship were originally meant for physical libraries and not much else. So why do we want to apply them to another context, a context for which they were never intended? What would you use to build this unending graph?
>
> Steve
>
>
>
> On 08/03/13, Karen Coyle wrote:
>> Stephen, I still disagree. The organization of the LIBRARY is the signature contribution. You can use the library without the catalog, and most users do. That said, the organization of the library has been limited in the past by its physicality. We should be able to go beyond this today, providing users with a map of topics that is an unending graph of relationships.
>>
>> kc
>>
--
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet
Received on Sat Aug 03 2013 - 11:09:21 EDT