>> So, instead of linking to American Memory, mirror it locally?
>
> Yes, quite possibly, if your collection policy warrants that direction.
IMHO, both of these approaches are fundamentally flawed. These methods
presume a centralized information model which has no future.
The catalog pays lip service to electronic resources, but catalog
records are built around a publishing model optimized for a few
specific types of resources. Shoehorning descriptions of how dynamic
resources outside this publishing model appeared during a moment in
time is an exercise in futility -- there will never be enough labor to
identify and process more than a minuscule percentage of what is new
even if nothing old is maintained.
Consider the OCLC's Intercat project in the late 90's. The Web was far
simpler then and many libraries contributed considerable staff
resources to making it happen. The idea was worth a shot, but it was a
spectacular failure that resulted in a database of dead links which
would have represented mostly ephemera (an academic euphemism for
garbage) had they worked.
Mirroring is equally hopeless. Ignoring that the resources required
are far beyond what we will ever have, resources can only be mirrored
if what you see is what they are. This is only the case with certain
types of materials that aren't dependent on the platform used to
deliver them, and the majority of those that we would want to mirror
involve considerable legal issues.
The catalog is quite good at what it does, so the trick is to figure
out how to adapt it so that it makes sense in an environment where
information is distributed and patrons are in whatever environment
they are in.
We should avoid using it for things it is poorly suited for such as
keeping track of Web resources. Otherwise, we detract from the value
it contributes and relegate it more quickly to irrelevance.
kyle
Received on Tue Jul 29 2008 - 12:18:25 EDT