Well searching for "Wizard of Oz" on Google is about like searching for "hot teen sex" in that there is a lot of commercial incentive for Google to return results and returning something relevant is trivial given that Google's algorithms are tuned to return potentially monetizable results over a diversity of potentially informative results.
In other words, when was the last time something from the LOC appeared as a result of your Google search - such as the copyright data from Webvoyage? For example google("wonderful wizard of oz" copyright) and try to find this highly relevant information from the LOC [http://cocatalog.loc.gov/cgi-bin/Pwebrecon.cgi?Search_Arg=wo...]
Aside on MrSid - if you are archiving a map for reference as a map rather than clip art it is ideal, and archiving maps for use as maps is what the LOC does. On the flip side, Google's goal is to sell advertising for companies like "worldsbestmaps.com" and so returning useless (from a map making perspective) results increases the value of Google ads to those companies.
Sure they have http://www.loc.gov/library/libarch-digital.html
Which is "ok". But it's not really searchable, the interfaces are clunky, and I can't really just go find say...the Wizard of Oz series, or review historic maps of my state without going through dozens of clicks in an interface that's taking up about 1/6th of my screen real estate (and only downloadable as a jpg or an obscure geodata format called MrSid).
Ugh.
It's obvious that a ton of thought has gone into organizing the data, but not much into retrieval.
By comparison, it took me .16 seconds to find the Wizard of Oz books...and oh, if I care, a play in three acts, some song books, and other odds and ends.