[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Libraries and archiving (Re: RE: If electronic is to replace paper)

> There are some serious technical and licensing problems with this. The
> electronic representation available to licensees will be in a display
> format such as HTML or PDF or more proprietary formats on CD publications.
> Many of these formats are subject to obsolescence over the years as
> software evolves and commercial fortunes change. The publisher holds (or
> should hold) a more durable representation of the data based on SGML/XML.
> Only if you have access to this can you be sure that in decades to come you
> will be able to gain electronic access to the works in a convenient form.
> What is the likelihood that in 20 years time we will have full backwards
> compatibility in web browsers (or whatever we are using then) to todays
> 'primitive' HTML markup?

The licensing problems are not serious -- they are problems of attitude
and policy and can be dealt with through education and negotiation.  The
technological concerns are serious, and I don't wish to discount them. But
they don't address what I think is the core question: where does it make
the most sense to locate the archival function: with publishers (who have
never served that function before) or with libraries (who have always
served that function)?  Once we answer that question, THEN we can discuss
how the technological problems will be addressed.  No matter who takes on
the responsibility, those issues will be more or less the same.  That
doesn't mean that we should lightly pass over the problems posed by
proprietary formats and the unpredictability of backwards compatibility --
on the contrary, I think it means that libraries must get more involved in
solving them.  This is our job.  If the primary function of a library is
to maintain a durable collection for its patrons, why shouldn't libraries
be the ones to develop a durable representation of the data to which it
has purchased access (based on SGML/XML or whatever other
language/context/technology comes up in the future)?


> It seems unlikely that publishers will licence their source data to
> libraries for archival purposes. 

I'm not convinced that licensing the source data would be necessary.  I'm
not a technology expert, but I can't imagine that there's no way to
translate the published manifestation of online content into a durable
electronic format.

> Why would libraries want to take on the
> role of maintaining current electronic publications from such data, even if
> they could get it? They would have to become publishers in their own right.

Libraries would obviously not be able to support indefinitely the exact
representation of content as originally published -- the bells and
whistles would probably disappear in an archival context.  But I think it
would be possible to archive the basic content (text and associated
multimedia files).  To the degree that libraries create content of their
own for public use (as they always have and always will), I suppose
libraries are "publishers" in the most literal sense of the term.
> Regardless of who does it, it will cost money.

Libraries have always been expensive, and always will be.

Rick Anderson
Head Acquisitions Librarian
Jackson Library
UNC Greensboro
1000 Spring Garden St.
Greensboro, NC 27402-6175
PH (336) 334-5281
FX (336) 334-5399

"Freudian apologists welcome his
objections as the Undead welcome
-- Frank Cioffi (on Adolf Grunbaum)