At 03:58 PM 7/7/2003 -0400, [log in to unmask] wrote:

>Two illustrations:
>1) I am trying as best I can to restore a c. 1870 house here in Oshawa,
>It was built with some typical gothic revival bric-a-brac which was removed
>the early 1940's. It would be a simple task to reconstruct it...IF I knew
>it had looked like! Over the last 130+ years, it's likely the house was part
>of any number of casual snapshots...but it's also equally likely that most
>these have long since been discarded. When the last one "expires" (if it
>already) I have no way of ever knowing what I'm recreating!

A photo archive of every house built ca. 1870 would be prohibitive. An
alternative is to have a modest set of records on selected houses in that
era and to have a capable designer synthesize a solution in the right style
which suits your needs.

>2) Suppose some historian, future or even present, is studying how and why
>the Ku Klux Klan became so powerful in Indiana in the post-WWI period. Will
>this person find useful data in articles in large newspapers, or the
>of politicians? Probably nothing more tham implications, if that. Where
>would the information be? In things that recorded the opinions and views
>of everyday people of the time...local newspapers, letters, diaries and
>suchlike. The exact thing we are told to discard!

Suppose one were to know that a collection of all communications to and
from Muncie, IN, was available in cartons in the basement of the local
library. Without a catalogue and suitable indexing, the information is of
no practical value. Amassing the material, preserving it and making it
accessible indiscriminately would be worthy but prohibitive. In the real
world, the collection must be culled at its source; if a small city like
Muncie is a poor example, consider a county or Indianapolis instead so that
the base would be large enough. However, choosing such at random would
overlook the regional variations - Indianapolis is nothing like Fort Wayne,
for example. So one should save all the correspondence of the state ... but
I trust you get the idea.

I am arguing that we need a practical solution to preserve not only what we
know to be of value but a representative subset of the rest. The size of
that set is constrained by the real world.

>I think that the more we know about history, the more we can undersand the
>present and predict the future...and the more we save from our history the
>more we can know about it!

There's no argument here, but having all without preservation or indexing
accomplishes little.

[log in to unmask]