> | It offers some features not natively supported by the Web and imposes
> | a much stronger hierarchy on the documents it stores.
> | Gopher's hierarchical structure provided a platform for the first
> | large-scale electronic library connections.
In Gopher, a thing you retrieve from the server is either a menu or a
file of some sort. (Searches produce menus.) During the heyday of
Gopher, HTML wasn't particularly common (or well supported in clients,
iirc) in Gopherspace, though it was possible to deliver HTML. You _can_
make the sort of mesh-like links in Gopher that are the primary mode of
the WWW, but since you can't really practically deliver content _and_
navigation in the same fetched thing, organization tends to be a lot
A lot of us had extensive discussions in our own shops and at the early
Gophercons about how to deal with the fact that every person has a
different mental map to the world than. We actually had a group of
librarians at the first Gophercon (specifically to keep us honest) who
were heard to mutter about it being like library school kindergarten,
watching us flail about.
There are a few reasons why pure hierarchy doesn't work very well. The
above noted mental map thing is one. The fact that knowledge is messy
is another. Another is that bald monkeys don't seek information very
rationally, and user interface design is basically the art of dealing
with that. Back then, finding the way to take a desired action (as
opposed to information seeking) was not as front and center as it is on
the web today, but it strongly affects user interface design as well.
The WWW is a multidimensional mesh (mess? ;)) where anything can link
to anything else. Early hypertext systems envisioned links being
bidirectional, but didn't anticipate the democratic nature of the WWW,
where it'd be difficult to get all the reverses set up. You _can_ use a
mesh-like technology (HTML links) to build a hierarchical organization,
but the user interface aspects mean its rarely done.
Real libraries deal with the information seeking side in several ways.
Call number systems are a way of creating serendipity: related stuff
lands close together on the shelf. Cataloging makes it possible to find
things via multiple paths (for libraries, titles, authors, various kinds
of subject groupings, etc) and to find related things that can't be
shelved in two places at once. (Electronic resources are easy to
"shelve" in multiple places, books are not.)