Sunday, May 2, 2010

IMLS progress report

IMLS Interim Performance Report Narrative
Creating Collaborative Catalogs: Using Digital Technologies to Expand Museum Collections with Indigenous Knowledge
Grant number: LG-24-09-0106-09
May 1, 2010
PI: Ramesh Srinivasan, UCLA –

We had four major goals during this initial phase of the grant, and we have been able to accomplish much within these goals.

1. Planning, selection of collections, digitization (Oct. 2009-Aug. 2010)

A. Planning Meeting Decmber '09:
We were able to accomplish several important steps in terms of project planning during this initial phase of the grant. All but one of the project partners were able to meet at Zuni for our kick-off meeting in December (this partner met us via teleconference), where we discussed many of the broad strokes of how the Collaborative Catalog system and larger project will be set up and how it will function when implemented. Our meeting discussions centered around three primary topics: the system setup and architecture (and a non-technical discussion of how we will get the different systems to talk to each other); evaluations, scale, and the distilling of best practices (and what we want success to look like); and intellectual property and protocol issues.

System setup and architecture:
The back-end of our system is going to be built using an emerging tool called Web Hooks, which is a web-based mini-program designed to pay attention to ‘events’ within a database. When a certain event occurs that changes the overall state of the system -- such as the 'remarks' field in an object record at a partner museum being updated with new information -- this mini-program knows to take a specific action -- in this case, sending the updated information along to update the collaborative catalog records. The upshot of this is that as information is updated, both in the databases at the partner museums AND the Zuni system, the systems will be able to synchronize automatically. The goal of this project again as mentioned in the proposal is to build this Collaborative Catalog System as an interface between a local tribal museum system (that can be scaled to any tribal museum) and ARGUS-based digital museum systems.

There are three major engineering/software tasks that are fundamental to this 'collaborative catalog': 1) a script to format the content coming via the Collaborative Catalog system to the Zuni system from outside museum partners; 2) The local Zuni system, which will be housed on a server at the A:shiwi A:wan Museum and Heritage Center (AAMHC) in Zuni, which will have a simple front-end interface allowing users to view museum objects and comment upon them (as well as add their own digital objects (video/photo/text) to the system); and 3) the Web Hooks 'pub-sub' protocol (publish-subscribe) that pushes information back and forth which is fundamental to the Collaborative Catalog System. We expect that all three of the aspects of our project that will be the most portable and adaptable to applications in other museums and museum consortia are actually all three of the above. The idea again as articulated in our IMLS proposal is one of building a Collaborative Catalog layer that can articulate knowledge across networks between a set of digital museums that use ARGUS and a local Zuni system.

Evaluations, and what we want success to look like:
We identified a number of primary activities where outcomes-based evaluation can be used, and we brainstormed around the best means of achieving these outcomes: 1) designing creating the system at Zuni, and monitoring its use within the community; 2) creating the Web Hooks scripts/ protocols for the Collaborative Catalog, and monitoring the number of collections records that were updated and improved using this; and 3) the leadership workshop that we will host at the end of year 3, where we present our system and methods to interested parties from other museums and tribes, and where we will monitor attendance and outcomes of what workshop participants do with what they've learned.

While we were discussing evaluations, the question of "what would success look like?" was put on the table. Jim Enote, the director of the AAMHC, talked about the desire to allow this system to truly be impactful of life at Zuni, including sharing the system and engaging schools and artists to use the system, and to be able to expand and extend it to other places and institutions (such as the National Parks & Monuments). He also spoke eloquently about the sense of hopelessness that people feel at Zuni when thinking about Zuni objects in other places, since the objects have been absent from everyday and contemporary life and community members are disconnected from this tangible patrimony, not knowing where the objects reside in many cases. He said that this contributes to sometimes negative feelings tribal members have about museums. Projects like ours are a really important part of closing the loop and enabling the museum to be a relevant continued space to underserved publics and source communities. This is ideally instantiated via our creative deployment of this system.

The protocols of sharing - ideas and discussion
An important part of our meeting (and project as a whole) was our discussion about the protocols of sharing information, using the appropriate protections for Zuni intellectual property. The goal is to create an architecture of the system that works with the different notions of property, knowledge, and information circulation as they stand with the multiple partners, and be respectful to Zuni practices and local social and cultural expectations. Therefore, the team agreed these intellectual property protections & protocols are an important element of our project, but exactly what form they will take and what they will look like must be worked out. At our kickoff meeting, we had a lengthy discussion and came up with several ideas that we would like to put into practice, and see if they work.

What emerged from our discussion was a 'moderated' sort of system where the appropriate religious leaders would review the object information coming into the database at AAMHC and make a determination about whether specific objects can be accessed by all Zunis, only certain groups, or only religious leaders. Many local especially tribal communities hold such local protocols which are critical to the continued presence of cultural traditions. For the objects that are accessible by everyone at Zuni, the system will allow comments, feedback, and corrections (this includes one idea, suggested by Museum of Northern Arizona director Robert Breunig to have a section called "things the museum would like to know", which we thought could also be called a "set the record straight" tab for Zunis who have information that they feel they wish to share that might contrast with something they have seen in the existing digital museum catalogs).

The steps in the process:
1. cultural advisors determine which objects are best to be shared at Zuni, and they set permissions for access to objects
2. Zuni users come in to AAMHC & interact with catalog, adding text, audio, video, etc.
3. individual set levels of access to their comments
(ie. Zuni community only | museum staff only | staff and public (just ideas) )
4. advisors get to review information
5. Zuni's info goes out to museum via WebHooks script

Our idea was to have a "parking lot" / "holding area" / "filter" where Zuni cultural advisors can monitor information coming into and out of the database. The question about how much work this filter will be was raised, and not exactly answered. We will want to phase into this, to see how it works and to see whether we need to rethink this setup.

We discussed for some time whether these comments should also be reviewed before being sent along to the outside museums, since we wouldn't want sensitive information to be inadvertently made public through our system. On the other hand, the task of reviewing (possibly) hundreds of comments about hundreds of objects is pretty intimidating. We do not yet know if people will make mistakes with sensitive information, but one decision we made was to make it as clear as we can that a user's comments can be either "Public" or "For Zunis Only", and each user will have to establish who can access their comments -- letting the community self-regulate. The most appropriate solution has yet to be determined, and it is likely that we will not know until we begin testing the system with a preliminary group of users.

B. Selections of collections and digitization:
Much has been done towards this end during this first phase of the grant. Most of the grant partners have digitized their relevant collections now (with the exception of the Denver Museum of Nature & Science), and previous collections reviews and/or consultations between the partner museums and the A:shiwi A:wan Museum and Heritage Center have yielded smaller sub-sets of objects that might be of the most interest to Zuni catalog users.

For those partners still needing to digitize their Zuni collections, some progress has been made in this first six months of the grant period. The Denver Museum of Nature & Science has worked to lay the groundwork for its participation in the online catalog. Specifically, the Curator of Anthropology has collated and refined the list of objects to be included, and worked with the Anthropology Collections Manager, Image Archivist, and Museum Photographer to create a process to begin photographing the objects and embedding documentation information within the digital images. On March 8-10, Zuni members and AAMHC staff Jim Enote, Octavius Seowtewa, and Curtis Quam visited the museum, making a preliminary examination of objects, vetting the object list, and providing guidance to the Museum Photographer. To date, some 160 objects have been photographed, created 359 images (many objects are photographed from multiple angles) for the project.

And since almost all of the collections from the other project partners were digitized prior to the start of the grant, the work is focused on the issue of extracting the relevant data from their systems and loading that data into our prototype system (see below).

2. Collaborative Catalog system development (November '09 - October '11)
The system development for Creating Collaborative Catalogs has been in happening in two parts. The first part has been to develop a local catalog system at the AAMHC in Zuni to accumulate and make use of the diverse catalog records and images from the other museums. The second part has been to develop the pubsub/WebHooks system for communicating between the institutions.

The first part, the development of the catalog at Zuni has been completed to the first prototype stage. A working system exists, in FileMaker Pro 11, and 9582 records have been uploaded from the four partner museums including 1257 images. The current prototype includes the ability to add Zuni knowledge to each record; add new images, sound files or videos; request new images or information from the holding institution; and to create relationships between the object records. In addition, there is an easy to use search feature that allows the searching of record data, image data and/or Zuni information.

The second part, that of the pubsub/Webhooks interface is also going well. The project is now working with Josh Frazer, a developer in Boulder, Colorado, who is also one of the original developers of the Google Code Pubsubhubbub hub, as well as the developer of the Wordpress Webhooks plug-in. Josh will be helping us over the summer to finish the development of a Webhooks hub and interfaces between the separate museum systems. Another issue with this part of the development has been ARGUS, a proprietary Collections Management System (CMS). Several of the museums use ARGUS as their CMS and, as a very limited and rigid system, it has proved very difficult to interface with. However, at the end of March, Questor Systems, the developer of ARGUS, was bought out by SydneyPLUS. Dr. Boast contacted SydneyPLUS in the middle of April and has had a very positive response from the company. As a result, Questor Systems and Dr. Boast are now in discussions as to how they can build a WebHooks module into ARGUS. Regardless of this challenge we have discovered some mechanisms by which we can work with the ARGUS systems, at least when they export information into readable spreadsheet and other database formats

3. Data transfer (partners to AAMHC) (March '10 - June '10)
Initially, we met with some frustrating challenges in getting data out of the systems of the partner museums in order to start working with it in the Collaborative Catalog. The ARGUS system has a very restrictive way of managing data (as noted above), which meant that we had to develop a creative workaround to start getting data into the new system. The partner museums each created an Excel spreadsheet with all of the relevant metadata from their catalogs, which were then integrated into the Collaborative Catalog. This is a preliminary solution that is less than ideal, since it lacks the important characteristic of ongoing, automatic updates. But as noted above, it is likely that this will not be a problem for too much longer, since the ARGUS system may undergo a significant redesign in the near future. The preliminary data transfer has been completed, and as we make progress on the system development, the ongoing nature of data transfer will start taking place.

4. Catalog launch, use, and evaluation (Present – June 12)
As we are working on the finalization of the data transfer issue, we have yet to complete the system’s initial deployment. But our timetable is allowing us to launch the prototype system with users by June of this year, which will coincide with the opening of a new exhibit at the AAMHC about the Zuni Day School, and using additional collections from the Maxwell Museum at UNM, a new potential partner to the project (many other museums have also expressed similar interest in our efforts!). The Maxwell as mentioned has expressed interest in the work that we are doing, and they have agreed to allow us to integrate their data into our Collaborative Catalog. Therefore, we believe that launching our catalog at the opening of this exhibit is timely and appropriate.

As mentioned, we have worked to make great progress (better than even expected!) on the four major goals held during this initial phase of the grant, and also are placing our communications and updates on our project blog:

We thank the IMLS again for its most generous support of this important effort and are available to answer any further questions from IMLS staff. Dr. Srinivasan, PI of this effort, can be reached easily at

Thank you,

Professor Ramesh Srinivasan
UCLA, Dept. of Information Studies