CollectiveAccess building steam?

Last year I had the great pleasure of working on a project with Karan Sheldon at Northeast Historic Film, which required migrating an old "flat" film database into CollectiveAccess. Since then I've caught wind of several other moving image archives either using CollectiveAccess or investigating the possibility. If a number of people are intending to use or further develop this tool for moving images, it seems like a perfect opportunity for collaboration. This post is a call for that.

Here's a summary of what CollectiveAccess does from my (somewhat limited) experience with it. I'll refer to it as CA in the rest of this post for short...

  • Provides a very clean cataloging interface through any web browser.
  • Enables the cataloger to define collections, locations, and entities, and relate assets to them.
  • Each asset can be related to other assets and versions, as in FRBR, Dublin Core, PBCore, etc.
  • CA provides picklists for taxonomies, genres, and anything else they might be needed for so you can control vocabularies and other value lists.
  • You can create custom CA "profiles" tailored to your needs, e.g. based on specific metadata schema.
  • Since CA is free and open source software, profiles can be shared with other users. The profile consists of a pretty straightforward config file.
  • You can set up CA to work with ffmpeg which allows uploading of moving image files, automatic transcoding to other file formats, extraction of thumbnail images, and extraction of technical metadata.
  • CA also provides a "front end" web publishing module, so you can make your CA collection accessible online.

So basically CA is a CMS that's been very well designed for cataloging and collections management. It runs on the LAMP stack, and is pretty easy to set up and maintain. The creator of CA is Seth Kaufman who lives in New York. It's fairly well documented and there is an online user forum where Seth provides active support. But I don't see that the user community has connected enough to begin supporting itself without Seth. (If I'm wrong about that, I'd be very happy to learn it.)

The AMIA Open Source committee could play a useful role in starting a dialogue among people using or exploring CA for moving image collections. I know we're planning a special meeting at the 2011 AMIA Conference for CA users and those interested. Kara Van Malssen is taking charge of that, and I'm sure it'll be announced in various ways when scheduled.

Meanwhile, I want to at least mention what we're doing with CA at the University of Illinois. We recently established a thing called the Center for Multimedia Excellence, with representation from many campus units engaged in producing and managing various forms of media. Since nobody has good tools for collections management, we're looking at various systems and tools including CA. We're throwing content against these tools to see how they match our requirements. As part of that, the sub-group focusing on CA has built a PBCore 2.0 profile for CollectiveAccess. It seems pretty great, but needs further development especially as PBCore 2.0 matures. As we are heavily invested in using PBCore, we would love to work with others who are also.

So there's a specific call for collaboration on building and sharing an effective PBCore profile for CA. And I hope we can get a larger conversation on using CollectiveAccess for moving images starting now.



Thanks for the kind words Jack! I'm certainly not the only one providing CA support, but I appear to be the only one naïve enough to not charge for the service :-) Generally I end up handling the open web-forum support posts, which others working on the project provide paid support (which helps to fund our development work). There are also third-party vendors doipaid ng development and support, especially in Europe.

We would welcome more collaboration with the moving image archives community. Let us know how we can help it happen.

– seth