RE: diglib: RE: thinking through maintenance of digital resources andbuilding communication methods

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

... from talking with Normandy, I gathered that one of
the problems facing Special Collections is the need to
curate exhibits over time.  In other words, it would be
unusual for Special Collections to mount something that
amounts to a static exhibit because its "exhibits" are
almost always featuring collections.  This poses a problem
from the standpoint of access to the sub-sites.  My initial
thinking is that Special Collections ought to keep and
maintain its sub-site pages, irrespective of whether they
also function as exhibits.

Mark

-----Original Message-----
From: owner-diglib@lists.uoregon.edu
[mailto:owner-diglib@lists.uoregon.edu]On Behalf Of Normandy Helmer
Sent: Monday, July 28, 2003 11:35 AM
To: David McCallum; Diglib; Juanita Benedicto; Cara List
Cc: Ted Smith; ayogi@OREGON.UOREGON.EDU; tamarav@OREGON.UOREGON.EDU;
mmohr@OREGON.UOREGON.EDU; llong@OREGON.UOREGON.EDU;
hyatt@law.uoregon.edu; Andrew R. Bonamici; Kellie Ann Garsed-Donnelly;
Andrew R. Bonamici
Subject: diglib: RE: thinking through maintenance of digital resources
and building communication methods


Many thanks to all who are helping us think about these issues. It is big
stuff that won't be resolved overnight and it does require broad
participation. Any procedures and policies that do emerge will have to
continue to evolve as our needs and abilities change.

First, let me emphasize that I DON'T want to see a strict regulatory agency
established in any of these areas, as our culture doesn't support that and
we don't have the resources for it either.

I would like to see a system of mentoring and monitoring that keeps all our
Web-based stuff accessible and appealing, while accommodating developing
skills as well as innovative ideas. I think we need that for design, for
routine upkeep, and for long-term maintenance of content, function and
discoverability (which includes metadata in appropriate formats as well as
functioning links). The goal is to be proactive in establishing viability,
instead of trying to clean up established messes.

We're good at delegating responsibility but not authority, and any systems
we develop need to take that as a given. Also, limits to the time commitment
we can reasonably expect from over-expended people. Failure to comply will
eventually have to go up the chain of command for resolution. Generally,
we'll get compliance through consultation, informed innovation, and
consensus. That's how we built a great library catalog.

For graphic design I think we need at least two people who are trained
graphic designers, preferably not from the same unit, reviewing proposed
designs. And a list of designers who can help put together good looking
pages. Having parameters written up front will help avoid extensive design
efforts that simply won't fly.

I'm NOT suggesting we codify aesthetics. We could even bring in people from
elsewhere on campus sometimes, to ensure that we welcome evolving ideas in
design principles. Tastes are personal and they change, so it's impossible
to achieve design that delights everyone for all time but we ought to be
able to find some kind of community standard. And the standard will depend
on the audience for a given sub-site--great design for freshmen will not
look much like great design for senior faculty doing research in historical
sources.

For technical design (i.e. interactive stuff), we desperately need to make
sure everything gets documented so we can maintain functionality when the
creator moves away from the site.

Cara's point about a "press" identify is a good one--the spine of the book
shows the publisher but every page does not. So maybe the home page of a
sub-site needs to reference the "skeleton" graphic id, but not every page.
That would work for digital projects too. (By "skeleton" I mean a little
consistent graphic that is visually coherent with Libweb design and also
provides navigation back to the home planet from the sub-site. If there's a
better term that we can all relate to, please let me know.}

My point about sunsetting is that if we are NOT going to maintain a
resource, exhibit or otherwise, it should either be removed or the home page
should clearly state the date that we stopped altering it (as we did for the
card catalog). The Blues exhibits has a prominent and complete date, but the
Track exhibit just says "month of April."  Exhibits are a little easier to
set dates for because they generally have less need to evolve than the
sub-sites that are core finding aids for uncat collections, or for digital
sites that are expected to continue to grow. However, the public doesn't
care what server holds the material, they are going to blame the Library as
a whole whenever information is out of date or pages aren't functioning as
expected.

Normandy


-----Original Message-----
From: Cara List [mailto:clist@uoregon.edu]
Sent: Friday, July 25, 2003 10:34 AM
To: Normandy Helmer; Juanita Benedicto; Cara List; Diglib
Cc: Kellie Ann Garsed-Donnelly; Andrew R. Bonamici; Tedsmith@Darkwing.
Uoregon. Edu; clist@OREGON.UOREGON.EDU; hyatt@law.uoregon.edu;
llong@OREGON.UOREGON.EDU; mmohr@OREGON.UOREGON.EDU;
tamarav@OREGON.UOREGON.EDU; ayogi@OREGON.UOREGON.EDU
Subject: Re: thinking through maintenance of digital resources and
building communication methods


I've read through Normandy's email once, and would like to review her
thoughts further with the exhibits committee, but I'm going to add a few
thoughts to the conversation off the top of my head.  My thoughts concern
web exhibits only-- I am not referencing other sub-sites where
considerations may be quite different.

I agree that standards and protocols for exhibits should be established,
and the exhibits committee will be working on this-- hopefully in
cooperation with Juanita and the WAG group.  One thing that I have started
to ask exhibit designers to do is to create exhibits within a single folder
with internal page links that are designed to be sustainable even when the
folder is moved.  This is a fairly easy step to take.

I hesitate to insist that all pages within an exhibit should have a
"skeleton" that identifies them through a "graphic identity" as a part of
the UO Libraries site.  I think that exhibits should be considered more
like the publications of a particular press-- identified as a product of
that press, but not necessarily bound together by a particular look.

The idea of assigning a guardian is a good one in most cases, however, web
exhibits are again more like monographic publications than, lets say,
subject guides, and my feeling is that exhibits should be built by the
original creator to be as sustainable as is _reasonable_.  If the original
author wishes to maintain a relationship with the exhibit and make sure
that external links and such continue to function-- that is terrific, but
if that author should leave the UO, should we "assign" a new guardian to
actively monitor the site?  My feeling is that we should encourage exhibit
designers to limit their external links to stable sites, but that we should
not edit the content in the future.  We could write some kind of caveat on
the main exhibit archives page stating that the exhibits are products of a
particular event-- the actual exhibit-- and that external links are not
maintained.  On the other hand I do like the idea of having a general
exhibits "guardian" who would be a contact person for any visitors who
found broken links that could be easily repaired ie. the guardian might be
contacted by a visitor who says www.xyz.com  has moved and is now
www.pdq.com.  I'm a bit loath to start burdening the committee with the
need to actively maintain exhibits created in the past (of the future) by
a  group of individuals who may well be outside of the library staff.

Regarding this point:
2. Standards for acceptable design parameters. We do want to encourage
creativity and design that really responds to the content of a given
sub-site; we also want design that is functionally accessible for
middle-aged librarians as well as Gen Z-ers. We want design that will not
embarrass us in six months, either from sheer ugliness or from lack of
sustainable functionality.

I absolutely agree about creating standards that go as far as possible to
ensure sustainability, but I do not want to insist on the king of design
standards that will limit exhibit designers in their expression re. style
or content.  Sheer ugliness would be unfortunate, but surely that is a
matter of individual preference and should not be determined by committee.

I certainly do not want to sound negative about these concerns-- I just
want to make sure that we do not limit our exhibitors, nor that we burden
ourselves too heavily with excessive responsibilities for what may in the
future be a large archive of past exhibits.  I am sure that I do not want
to include a "sunset" clause for exhibits in the same way that I do not
believe that we should weed our monographic collection of old books.

--Cara

At 05:04 PM 7/24/2003 -0700, Normandy Helmer wrote:
>There have been discussions going on in various places related to sub-sites
>and exhibits. Aspects of these discussion concern Library Web pages (thus
>WAG), maintenance of valued digital resources (thus DigLib) and Web
>resources related to Library exhibits (thus the Exhibits Committee). We
have
>expertise and concerns and ideas coming from many quarters but we haven't
>brought everyone to the same table to leverage our resources and set good
>policy so we can move forward.
>
>I'm sending this to DigLib and asking that Cara and Juanita forward it to
>their constituents so we can have all the parties involved talking together
>at the same time. We can add people to the DigLib list so they can
>participate directly in the discussions, as well as viewing the archive.
>
>Issue 1: Graphic design of sub-sites and their graphic relation to the
>Libweb standards.
>
>Discussion on this has generally leaned in the direction that exhibits and
>sub-sites can appropriately have designs that do not conform to the Libweb
>graphic standards nor use the Libweb templates. The rational is that, like
>an exhibit within a museum, the most effective presentation provides you
>with a visual space that is consistent within the exhibit gallery, no
matter
>what the rest of the museum looks like. Additionally, the Libweb design was
>tailored for text-based presentation with a little ornament, which is not
an
>effective way to present other kinds of materials.
>
>However, the Library wants all its Web-based resources to function
>effectively and display good design principles, and give some indication
>that there is a relationship between the sub-site and the Library. As we
>have Web-based materials on multiple servers (such as Boundless, the
>ContentDM server), this doesn't fall neatly into any one jurisdiction. To
>the public, it's all one big Web site, and we need to have some common
>ground under our feet that will also support creative design.
>
>I'd like to see us develop:
>
>1. A mandatory "skeleton" that appears on every page and connects the page
>to the Library. I say "skeleton" meaning a minimal graphic entity that
>doesn't substantially alter the design of the sub-site but does brand it as
>Library content and provide a link to a central location.
>
>2. Standards for acceptable design parameters. We do want to encourage
>creativity and design that really responds to the content of a given
>sub-site; we also want design that is functionally accessible for
>middle-aged librarians as well as Gen Z-ers. We want design that will not
>embarrass us in six months, either from sheer ugliness or from lack of
>sustainable functionality.
>
>3. Review process to facilitate developing and implementing a good design.
>That way anyone asked to develop a design for a sub-site will have
>guidelines and some assurance that the design process won't be derailed in
>midstream.
>
>Issue 2: Maintenance and preservation of sub-sites
>
>By sub-sites I mean sets of pages developed for one purpose, usually to
>present a finding aid for a particular collection
><http://libweb.uoregon.edu/speccoll/mss/burgess/index.html>; an exhibit
><http://libweb.uoregon.edu/exhibits/archive/track_and_field/>; or a digital
>collection <http://e-asia.uoregon.edu/>. These often have audiences beyond
>the campus community that most of Libweb is really aimed at.
>
>The Spec Coll site has dozens of sub-sites that were developed over the
>years (often by a student employee), that met no standards, were unknown to
>people elsewhere in the department, and were never maintained but stayed
>visible to the public. Getting those into adequate condition has been a
>nightmare so I am keenly aware of the burden of maintenance these sites can
>create. Maintenance of Web sites, even small ones, is a time-consuming
>proposition that requires a routine time commitment as well as the tools to
>keep the code as good as the content. Exhibits are another example of
>sub-sites that were OK when developed but usually didn't have anyone
>dedicated to keeping them maintained. While the Exhibits Committee can now
>help establish new sub-sites, they won't necessarily have the subject
>expertise or the technical expertise and time to maintain old exhibits.
>
>One aspect of maintaining sub-sites is the routine update of content and
>code; another is communication with people managing other related sites.
>When pages get moved around links into the sub-site get broken, and right
>now we're counting on the public to report link problems to us instead of
>developing a pro-active communication system between sub-site managers.
When
>a cataloged site gets rearranged, catalogers have to fix the links in the
>bib record too.
>
>Another aspect is the long-term preservation of the resource represented by
>the sub-site. Our thought on that has generally been: We put a lot of
>resources into it; let's maintain it to get our money's worth. However, do
>we preserve sub-sites through a snapshot of the original site or through
>maintenance of the evolved site? How much do we care about documenting the
>evolution of a given sub-site?
>
>We don't have a department charged with managing content or code of Web
>pages for us, and we don't have staff whose positions are dedicated to that
>purpose. So we've used a distributed model, which sort of works given our
>resource limitations, but can leave people with big Web maintenance burdens
>and no support group to make life easier. Juanita can help, but she can't
do
>all the Web work for us any more than I can do all the digital library work
>by myself.
>
>I'd like to see us develop:
>
>1. Policy on how we preserve and maintain sub-sites, what resources we
>allocate, and how we assign responsibility.
>
>2. An appointed guardian for each sub-site who will be responsible for the
>actual maintenance work. If that person is reassigned or can't continue to
>maintain the site, then a new guardian is chosen. The guardian's contact
>info goes on the home page of the sub-site. The guardian has the computer
>tools, the subject expertise, and the time available to fulfill the
>maintenance requirements.
>
>3. Communication system within the guardians (to include everybody who
>manages a lot of Web pages) to facilitate that sites continue to interact
>effectively despite reorganization and substantial maintenance activity,
and
>that knowledge of cool tricks is shared along with unsolved problems.
>
>4. Review process to ensure that every sub-site site either gets an active
>guardian or has an enforced sunset date. We can keep an archive copy but it
>won't be on-line. In the case of exhibits the Exhibits Committee will be
>vetting new exhibit sites that relate to physical exhibits, but we don't
>have a similar process for sub-sites that don't fit that mold.
>
>==============================================
>
>I think these are big issues that cross many boundaries and affect many
>people and constituencies within the Library. But they all come down to how
>we ensure a professional and coherent Web presence to our public, and we
>really need to start addressing them in an inclusive forum.
>
>I'm trying to start this forum by presenting some of my ideas about ways to
>address these issues, but MY ideas may be stupid or may not meet the needs
>of other people or other groups. Please share this message with other
people
>that you know will want to be involved in these discussions, please
>contribute additional issues to be discussed or ideas to be considered, and
>post them to the DigLib list so we can all think about what you have to
say.
>
>Normandy


Please notice that my email address has changed!
My new address is clist@uoregon.edu

Cara List
Art and Architecture Librarian
Architecture and Allied Arts Library
5249 University of Oregon
Eugene, OR 97403-5249

541-346-2200
clist@uoregon.edu