As part of the Source Code Analysis, we are looking at a few selected open source and commercial collections management systems, to help us understand the landscape and customers we hope to serve. This analysis focuses on the user model, functionality, etc.
This is a set of PHP files building upon a MySQL database to support various aspects of museum and archives collections management. This code supports a close alignment of end-user functionality, however, it is not clear whether and how much the current implementation can be leveraged to support the planned services.
Dan Sheppard and Carl Hogsden at Cambridge are working on an analysis of the code, their findings are here: OpenCollection Source Code Analysis.
A SOA based system supporting museums, archives and collections management. Their scope is well beyond ours, including content/knowledge management, but they have a "heritage management suite" that makes a nice case for their solution. They also provide hosted services (SaaS) for smaller clients. They are a commercial venture, and so at least on the surface have developed a reasonable business/sustainability model. They have an intriguing client list, including museums, archives (some business and public archives), universities, etc..
Notes from Demo.
This has only some overlap with our functional domain, however it does have some very strong documentation we might use (e.g., complete functional specificationsand metadata crosswalks for common standards, etc.).
Some initial comments:
- The work in name authorities looks quite good, and the requirements and some of the data model could probably be leveraged.
- The work on locations is worth reading, but explicitly does not support movement, and is designed to support permanent location.
- The accession and deaccession requirements are well documented and worth reading. They appear to be a subset of what we have talked about in the workshops.
- The description area (general metadata for objects) has some really good ideas, and nice requirement descriptions (i.e., I (PLS) really like the clarity and the form, and think there are some good ideas in there as well, including the model for import export to common standards, hierarchical organization and description, etc.).
- The AT team recently conducted a user survey to help identify enhancements and updates. The results were recently posted.
- The AT team has also published a features matrix comparing Archivists' Toolkit to other archives management applications.
Patrick prepared some notes from a demonstration from the AT team.
This is a system for managing natural history specimen collections. It is a bit far afield in some ways, but has a lot of overlap in some areas. The new (v6) version is soon to be released, and in at least some notes is described as open source. I (PLS) have a question in to the Specify dev team about this.
Update: the version 6 release has slipped to "Fall 2008". They have release an updated DB schema in both text and graphical forms. These reveal quite a bit about the functionality, and provide some interesting points of comparison. They have made some interesting decisions, like modeling a Loan and a Borrow as distinct objects that are in effect mirrors of one another (but Loan has much more info than does Borrow). In general, they have several "user definable" fields of various types in the major object types, and then also allow for "Attachment" records that can point to an external document. They have a mechanism for triples associated to attachments, and a more flexible property bag mechanism for objects - pretty much what we have been proposing as a compromise between core relational functionality and a flexible/extensible schema for customization.
This is a specialized system for documentation about architecture and architectural projects. The UI is pretty minimal, and usability does nto seem to have been a high priority. There is some end-user documentation online, but it is not all that helpful to understanding the finer points of the system, how it is intended to be used, etc. I could not find any architecture documents, altough the sourceforge summary describes it as PostgreSQL-based, using Java, JSP and PHP. I checked the forums, and there haas been no activity since Sept. 2007. The CVS has only the PHP, and seems only to archive the demo web app, although the project summary claims it is available under BSD licensing. From what I have seen, this does not look like it is worth pursuing.