RE: [-empyre-] Who decides and what to preserve

["Paul Koerbin" <pkoerbin@nla.gov.au>]

Some interesting replies to this pertinent question by Sharmin. As a
mere practitioner and public servant I will not venture out of my
territory into the philosophical issues of whether to archive or not.
Only to say, we are not specifically focused on expressive culture
(although that it part of it) but also dull documents like, lets say,
reports of royal commissions, technical documents, scientific working
papers, refereed e-journals, government publications - things that there
is a high probability will need to be accessed in the future for further
research and analysis.

Anyway, I really wanted to address the main question in Sharmin's post.
How the curatorial organisations make selection decisions is indeed
interesting I think. Firstly of course there is the whole issue of why
select at all. And some institutions choose not to and attempt whole
domain archiving; or make selections based on other analytical
techniques, e.g. frequency of publication. So why select at all? Well,
as others posting to this list have pointed out, the large 'harvest all'
archives face the problem of not being able to do much to ensure the
quality and completeness of the content. The reason for selection is
that we recognise that some documents are important enough to want to
harvest and archive them in as complete and functional a state a
possible - and currently this requires a fair degree of manual work to
check the quality and fix problems. 

The problem with this selective approach is that we have to make hard
decisions: archive this but not that. And as Sharmin has pointed out, we
cannot really know for sure what would be of use in the future. As I
pointed out in my opening post, the NLA has been a major collector of
print ephemera for many years because we recognise the value in
ephemeral publications for future research. The fact that we are highly
selective in archiving online resources is more pragmatic than
ideological. As Margaret Phillips mentioned, we will be working towards
doing a whole (or partial) Australian domain crawl in the near future.
This is something we have wanted to do really since the inception of
PANDORA since we recognise that the ideal approach (for now, given the
current technology available to us) would be a mixture of selective
quality assessed archiving and more automated whole domain harvesting.

In regards to the mechanics of selection at the NLA for PANDORA, this is
based on our selection guidelines
http://pandora.nla.gov.au/selectionguidelines.html
These guidelines point to our priority collecting areas. However we do
aim to get a broad coverage as well and we do identify areas where we
are weak and see if we can address that by some sampling. For example,
while we have handful of blogs archived already, this year we hope to
better represent blogs in PANDORA by selecting a broader
'representative' range of blog for archiving, so as at least to have
some in the Archive.

Selecting in the way we do, howerver, raises another interesting issue
in that the activity of selection in fact defines the publication for
archival purposes. By that I mean we could (and do) select anything from
a single file to a whole web site. We could (and do) select a part of a
web site and THAT for the purpose of the archive (and associated
metadata creation in the form of a MARC catalogue record) is the title.
The selection process seems to necessarily deconstruct the web for the
archival purpose. It is this process that can cause the curators a great
deal of angst - but as ever it is a process driven by the exigencies of
available resources.

Our selection decisions will no doubt frustrate future researchers, but
then it will be the task of future researchers to make the most of the
resources that come down to them, as it has always been. The fact that
archives such as PANDORA and the Internet Archive have taken initiatives
and have resources dating back to 1996/1997 I think will be valued (and
now doubt our respective approaches and results will be criticised). But
many other curatorial institutions have done very little to date that
has provided practical results. Perhaps their research and deliberation
will eventually produce better methods and outcomes - let us hope so -
but in the meantime how much has been lost? how many years of web
publishing not represented in archives? As the Director-General of the
NLA has said, "if it is worth doing, it is worth doing badly"!

Paul

-----Original Message-----
From: empyre-bounces@gamera.cofa.unsw.edu.au
[mailto:empyre-bounces@gamera.cofa.unsw.edu.au] On Behalf Of Sharmin
Choudhury
Sent: Monday, 7 February 2005 5:58 PM
To: empyre@gamera.cofa.unsw.edu.au
Subject: [-empyre-] Who decides and what to preserve


I am not sure if anyone has brought this up but this is something I
think is
important. PANIC does not specify what to preserve and does not rate
digital
objects by content. We leave it to the curatorial organisation to make
that
decision. Yet how a curatorial organisation would come to a decision
fascinates me, because often what we consider not worth saving is
exactly
what the future generations might consider as being important. As a case
in
point, the ancient Egyptians did not think it very important to record
the
lives of the ordinary folk, the workers who built the pyramids and so
worth.
Yet one of most important discoveries in recent times have been camps
for
the said workers where the workers have left their mark and bits and
pieces
from their daily lives. Anyone have any comment in this regard?


Sharmin Tinni Choudhury
Research Engineer
DSTC PTY LTD

_______________________________________________
empyre forum
empyre@lists.cofa.unsw.edu.au
http://www.subtle.net/empyre