Getting Started

[kerim friedman]

Hello. I'm an anthropologist and documentary filmmaker who recently
got drafted to be an archivist. Having no experience with running an
archive, but lots of experience with WordPress, I was very happy to
learn about Scriblio. I've successfully installed Scriblio on my site,
but I'm unable to figure out how to get started using it. Since I
imagine I'm going to need a lot of help along the way, allow me take a
little time to describe the project and how I imagine Scriblio fitting
in with that project.

For the past four years I've been working on a film about India's
Denotified and Nomadic Tribes (DNTs):

http://hoochandhamlet.com

A short film (15min) we made early on is available on Google Video via
a link from this page, which also has a study guide:

http://der.org/films/acting-like-a-thief.html

In working on the film we've been closely affiliated with an Indian
NGO, the Bhasha Research and Publication Center, who works on behalf
of DNTs:

http://www.bhasharesearch.org.in/Site.html

They have an extensive archive of materials relating to the entire
history of the DNT Rights Movement which started back in the 1990s.
This includes correspondences, reports, legal documents, etc. much of
it is already in digital form. I've uploaded some sample documents to
a Box.net archive:

http://www.box.net/shared/l7kz743eby

I've agreed to help Bhasha create, host, and maintain a freely
available digital archive of this material. That's where Scriblio
comes in.

I've created a Scriblio install here:

http://kerim.joyeurs.com/dntarchive/browse

As it is just a test install it is password protected:
Username=dntarchive Password=budhan

But now what? I don't understand how to create records, or facets, or
anything of the sort. None of the documentation seems to explain these
basic steps. I even tried running the walkthrough, but got stuck at
importing MARC records because I didn't know how to create a proper
prefix and I was presented with a series of errors when importing.
I've not continued with the import because I'd rather work with my own
data, if I can only figure out how to do so.

Like I said, I'm not an archivist, so perhaps there are some things
obvious to archivists which are not to a novice like myself. I would
very much appreciate any help I can get.

Thanks!

kerim
____________________________________
P. Kerim Friedman, Ph.D.
Department of Indigenous Cultures
College of Indigenous Studies
National DongHwa University, TAIWAN
Tel: +886-3-863-5795
http://kerim.oxus.net/
____________________________________

[Steve Toub]

kerim friedman wrote:
> This includes correspondences, reports, legal documents, etc. much of
> it is already in digital form.
>

For these documents, has a librarian or archivist described them in any
way, or are they just PDFs and Word files with no "card catalog" or
"finding aid"? If there is any description, can you post examples of how
these items/collections are described?

> I've agreed to help Bhasha create, host, and maintain a freely
> available digital archive of this material. That's where Scriblio
> comes in.
>

If you don't have existing MARC records, I'm not so sure Scriblio is the
right tool for the job. If you can describe the functionality you want
to have in your digital archive and give a better sense of how these
materials are currently described, I may be able to point you to other
tools that may be more appropriate.

--SET

[kerim friedman]

As I said, I'm the archivist for this project. We'd love to have a
professional if someone was willing to volunteer, but at the moment
there is just me. So the answer is no - we don't have any other
information other than what you see in the box.net folder. If someone
could tell me how to produce the proper meta-data I'd be happy to do
so.

thanks,

kerim

[Casey Bisson]

Actually, a source of MARC records isn't required. The Beyond Brown
Paper image archive contains all original cataloging:

http://beyondbrownpaper.plymouth.edu/browse/

Photos are being uploaded to flickr, where they're tagged, then a
Greasemonkey script sends them to Scriblio. I've not released the
flickr importer because that site is running old code and it's not
compatible with the current version of Scrib, but I'm under the gun to
get the site upgraded and will have a new version very soon.

Additionally, a feature that hasn't been discussed is the ability to
add or change cataloging metadata inside Scrib. More on that later.

--Casey

[kerim friedman]

Yes. This seems to be a major source of my confusion. I just assumed
it would be possible, indeed easy, to add or change metadata inside
Scrib. - since I assumed it would be like adding tags or categories to
posts. I now understand that most people only import data from other
sources. Still, I'm glad to know it is possible to do so, and I look
forward to learning more. Casey gave me some info on how to run the
import of the test data and I will play around with it some more.

Cheers,

kerim

[Casey Bisson]

Kerim,

As noted previously, the Beyond Brown Paper archive might be a useful
model for you.

http://beyondbrownpaper.plymouth.edu/browse/

The cataloging is being done in flickr, where metadata is entered as
tags in this form:

fieldname=value

So a photo of a bunch of logs might have a tag that looks like this:

subject=logs

To bring it all together I wrote a very simple importer that reads the
tags from a flickr photo and parses them into the kind of metadata
Scriblio expects.

Your question also made me finish work on a feature in bSuite[1] that
allows you to enter and edit tags in WordPress/Scriblio in that form.
I call them "machine tags"[2] because that's a term flickr uses. All
the metadata inside Scriblio is actually stored as machine tags, so
this feature allows you to fix records or do original cataloging right
inside Scriblio.

You don't have to declare or configure a fieldname before you use it,
it's simply a matter of typing it in. On the other hand, what you type
is what you get, so any typos in the fieldname will create a new facet
in Scriblio (though you can fix it simply by correcting that typo).
And there's no vocabulary control or other enforcement of data
standards.

There are, in fact, a lot of reasons why this is a bad solution for
anybody attempting to create a project such as you describe (or the
one I named above), but my experience is that the "right" way to do
something often doesn't ever get it done.

Our comfort in this process with the Beyond Brown Paper archive has
been that we can pretty easily export the data in MARC or MODS or just
about any other format (in case we find better software). And flickr
offers some surprisingly powerful tag manipulation tools that help us
fix data as we get better about defining our vocabulary for the
collection.

For your project, you could think of it as WordPress with better
metadata. You'd create posts that represent the archive objects and
add machine tags to make them more findable. Then the users can link
to and comment on them (which is when an archive gets exciting to me).

--Casey

[1] bSuite is required for Scriblio, but it's also used in general
WordPress installations. http://wordpress.org/extend/plugins/bsuite/

[2] More info on machine tags and their implementation in bSuite: http://maisonbisson.com/blog/bsuite/machine-tags

[kerim friedman]

Casey,

Thanks. Given the difficulties I might face training NGO staff to
enter data, this does not sound like the right solution for me at this
time. I really appreciate your help, and I'll keep an eye on the
project as it develops.

Cheers,

kerim

> WordPress installations.http://wordpress.org/extend/plugins/bsuite/