RE: diglib: IR project

[Mark Watson <mrwatson@OREGON.UOREGON.EDU>]

Sara:

I don't think anyone is saying that the MSU wouldn't work
for the IR project.  It's just that the MSU Guidelines do
not currently address something of the IR ilk, nor does IR
exactly fit the definition of a digital library collection
as defined by the former DigColl group.  I think we are struggling
a bit with what it means exactkt to include something in the library's
collection.

Mark

-----Original Message-----
From: owner-diglib@lists.uoregon.edu
[mailto:owner-diglib@lists.uoregon.edu]On Behalf Of Sara Brownmiller
Sent: Monday, June 16, 2003 1:09 PM
To: JQ Johnson
Cc: Digital Collections Group
Subject: RE: diglib: IR project



It seems to me that the MSU may be a good place for a "pilot project"
since it would not require much investment in creating a storage system
for this project.  It's my recollection that some of the original
questions prompting this discussion was where to put the data that some
groups or departments are ready to deposit now, or soon.  Should this turn
out to be a wildly successful undertaking, then we would need to review
and evaluate the best means to offer and maintain the digital collections.

Could you elaborate on why you think our current MSU wouldn't work for the
IR projects?

Sara

Sara Brownmiller			University of Oregon Libraries
Director, Library Systems 		1299 University of Oregon
Women's Studies Librarian		Eugene, OR  97403-1299
					541/346-2368 (voice)
snb@uoregon.edu				541/346-3485 (fax)

On Mon, 16 Jun 2003, JQ Johnson wrote:

> I agree with both Mark and Faye, and as I noted before am REALLY glad that
> we're
> having this discussion.  I also believe very strongly that the world is
> changing rapidly, and that the one sure way that Bob would be right in
> predicting the demise of libraries would be if libraries didn't innovate
and
> adapt their vision to the changes that are occuring in the global
> information infrastructure.  The worst fate would be to continue to copy
> scrolls by hand and guard them in the scriptorium.
>
> One critically important point that Faye raises is that the IR is a pilot
> project.  We don't in fact know exactly how it will evolve in the next few
> months.  I think from my perspective that that's a key reason not to use
the
> MSU for it at the moment.  A secondary reason is that if the pilot project
> were wildly successful the full project could require a lot more disk
> resources than the MSU has available.
>
> Faye raises the question of long-term archival and preservation.  I think
> it's important to realize that this is not a black-white question.  We all
> realize that some items in our collection are eventually deaccessioned.
We
> realize that the balance between access and preservation should depend to
> some extent on the purposes of the collection.  We also all realize that
we
> don't yet understand all of the technical issues associated with long-term
> preservation of non-book artifacts.  Will those reel to reel tapes you
> collected in the 1950s still be playable in the 22nd century?  Will
scholars
> at that time be frustrated because you did a format conversion to CD in
the
> 1990s and in the process lost some of the richness of the originals?  The
> dspace model that the IR group has evolved (mostly borrowing from MIT
> Library) may be slightly different from the model espoused by some of the
> diglib group, but I don't think it's wildly different.  It assumes that
> there are multiple different formats of information, and that some are
more
> easily preserved than others.  It makes a strong committment to long-term
> archival of the "bitstream" (the sequence of bits in the submitted digital
> artifact) and to some metadata describing the putative format of that
> bitstream.  It only commits to preserving the intellectual content of the
> item if it is submitted in some subset of all possible formats, and we
> currently believe that we are best served in the IR project by drasticly
> limiting the set of formats for which we commit to format conversion and
> preservation in that stronger sense.  That's because we believe that most
of
> the IR collection is likely to be of greatest interest to scholars in the
> short run -- the first few years or decades, and hence that our efforts
are
> appropriately weighted towards providing intellectual access.  Longer term
> we, at Heather Briston's instigation, are unwilling to make pie-crust
> promises, and so what we think we must commit to is a policy where IF we
> were in the future to decide that we could no longer maintain the IR
archive
> THEN we would not just deaccession the works but would (a) attempt to
return
> them to the original submittor and (b) would reserve the right to assess
the
> works for potential inclusion in some successor archive.  In other words,
we
> think we've done a pretty thorough job of understanding the long-term
> archival issues, and that the diglib group will likely be able to learn
from
> our pilot project.
>
> JQ
>
>
>