MUSEUM-L Archives

Museum discussion list

MUSEUM-L@HOME.EASE.LSOFT.COM

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Reply To:
Museum discussion list <[log in to unmask]>
Date:
Wed, 7 Aug 1996 13:22:48 -0500
Content-Type:
TEXT/PLAIN
Parts/Attachments:
TEXT/PLAIN (92 lines)
No need for asbestos underwear, Paul - I agree completely. I'm inherently
interested in computers and the digital age and have been doing some
reading on the subject. In my view, the big problem with moving toward
digital archiving is that computer technology isn't stable for long
enough. The media is stable, sure - perhpas even more so than analog media
(paper, film, etc.) - but there is another factor to be considered when
moving to digital storage: FILE FORMAT.

Say you choose to save your images in TIFF format (a very common format)
on a read-write CD ROM. In ten years, when you want to access that file,
will your programs be able to read TIFF files? Maybe. In twenty years? I
doubt it. OK, so whenever a new format seems to be taking over, you buy a
program that can convert all your files from TIFF to the new standard.

This is expensive. Very expensive. Perhaps prohibitively expensive for a
museum. Sure, the program may be cheap, but someone has got to spend the
hours doing the conversion. (Believe me, I know. The lab where I work
stored all its 1980-1985 research data on reel-to-reel tape and when I
joined the lab in 1991 people were discussing the idea that since hard
drives for our VAX were getting so cheap, perhaps we should transfer all
those files onto the VAX, allowing us to dump the old tapes and get some
more storage space. Doing so required changing the format slightly, but we
had a program that would do that. When our sysadmin was approached by
this idea, he rebelled, pointing out that it would take him days, maybe
weeks, to do the transfer - even longer if there were any difficulties
with the media (and there always are, when you are talking about old
technology meshing with new technology).)

And that brings up another problem. Will you have the hardware to read
your CD-ROM in 10 years? In twenty years? In my desk drawer at work I
have a stack of 5 1/4 DOS disks that contain data. I wanted to take a
look at the data, but couldn't find a single computer in our lab that had
a 5 1/4 inch floppy drive. Finally, our sysadmin resurrected an old
computer that nobody had touched in several years that had both types of
drives and I was able to copy the data onto 3 1/2 inch disks. So, you can
mitigate this problem by keeping your old equipment, right? Well, you
also have to keep that old equipment working (not an easy task) and
connected or connect-able to your current system (also not an easy task).

All of these ideas were shamelessly stolen from Clifford Stolls book,
_Silicon Snake Oil_ (see Ch. 11). And before you dismiss me as a Luddite,
let me just say that I love computers and spend much of my day using them.
Wouldn't have it any other way. I just think that there is too much hype
and not enough real thought behind a lot of decisions involving
computers/digial information these days.

So don't get sucked in by the extreme coolness of being able to do quick
and easy searches of your archives or by the idea of storing images of
your entire collection on a set of six CD-ROMS. Seductive, yes.
Practical, you gotta decide. Don't let your source of information be the
people who are trying to sell you their system - make sure you read the
critic's p.o.v. (P.S. to Tim: any salesperson telling you that you're
gonn use floppies to store your information is clearly out of his/her
mind. Smile politely but run. If the salespeople aren't up on this field,
is their software gonna be?)

I could go on and on about this (some of you may think I already have)
but let me give one more example about how a technology that sounds great
can have serious drawbacks.

<<rant mode on>>

Digital card catalogs. I love certain aspects of them - I don't have to
stand up, move around, I can find out if they library has the book I need
from my lab, and I know if the book is checked out and when it will be
returned. But last week I was in one of Yale's main libraries and had to
wait for 10 minutes before I could get a terminal in order to find a call
number. Don't get me wrong, they have a lot of terminals - at least 35.
But the same amount of space that those terminals employ holds the entire
card catalog. I'd estimate that at least 100 people could simultaneously
use the real card catalog. Sure, there are times when you might need the
same drawer that another person is using. But chances are that the person
won't need it for a whole 10 minutes. When people do online searches they
can easily use the terminal for a long time - 20 minutes or more,
effectively stopping anyone from using that slot to access the entire card
catalog.

Of course, the argument is *yeah, but next year we'll have 20 more
terminals and this won't be a problem*. Right. So instead of supporting a
method that has worked well (for patrons, at least) for many years, I've
been put into digital purgatory always waiting for the next installment of
new machines, faster processors, new versions of software. It is hard to
say if things are better or worse, sometimes.

<<<rant mode off>>


Margaret A. Martin
Yale University

[log in to unmask]

ATOM RSS1 RSS2