[WEB4LIB] Re: Google Print as the library's mission

[arhyno at uwindsor.ca]

The signal to noise ratio in indexing full text book materials makes me 
wonder whether the hype has outdistanced the reality in terms of practical 
information retrieval. The value of Google print may be more as an 
augmentation to physical media or other systems, finding a passage in 
something you have previously read, for example, or jumping into the full 
text after being led there by carefully crafted metadata. Even as  "print 
on demand" and display technologies improve, most digital representations 
of content still lose their edge as the quantity of content grows. Karen's 
point about Walmart is well-taken (I can even picture the sappy 
commercials) and I think the Walmart analogy also works well on a lot of 
nitty gritty technical layers (it's easier to park near a Walmart than 
most small stores in a downtown in the same way that Google is often 
closer and much more fault-tolerant on the network than most library 
services), but I wonder if Google Print is far more audacious in concept 
than it will be in execution.

It seems to me that Google Scholar could have a much more profound impact 
than Google Print  because researchers so often demand quick turnaround 
time, and Google has made investments in network infrastructure that even 
a company like Elsevier is not going to be able to match. I also tend to 
think, in the academic world at least, that we greatly overestimate how 
much research skills are applied in information seeking among those who 
otherwise have some dependence on research for a living. One factor that 
kept our library databases relevant to the Computer Science department on 
my own campus was that CiteSeer used to be plagued by performance 
problems. Google isn't going to let that happen, and with some of the same 
brain trust that was behind CiteSeer working on Google Scholar, it worries 
me that the best that most of our databases can do is reach out with IP 
authentication at a time when a majority of our users access library 
content from external networks. 

So to go to Ranti's question, if "our faculty found the collections 
outside of our own holding and outside of of the subject area we have in 
our institution, what will that mean to us in term of supporting their 
research interest and needs", I would say that we need to get our rights 
management, interlibrary loan, and authentication hooks up to snuff, and 
find a way to plug into Google Scholar and other third party suppliers of 
licensed content at the points where our user community uncovers desired 
sources. I am not quite sure how best to do this, but if I had access to a 
budget for implementing a federated searching engine right now (and I 
don't), I would be sorely tempted to redirect those funds into flexible 
implementations of technologies like shibboleth and making interlibrary 
loan a more viable option. 

art