2007/09/26: AHDS Digital Moving Images and Sound Archiving Study.

Today I read the AHDS Digital Moving Images and Sound Archiving Study.

I've been looking into metadata for the preservation of digital video/multimedia. It's an area of intense interest for me but I haven't had a work-related need to review it since I wrote, "OAIS, METS, MPEG-21 and archival values" for the Spring 2002 issue of The Moving Image. That means I haven't taken a good look at it since early 2001, given peer-review time lag.

I've had a request for consultation about archiving digital video, so I'm taking the opportunity to re-familiarize myself with the latest and greatest. The AHDS Digital Moving Images and Sound Archiving Study was released in August 2006 and it's the most current description of the state-of-the-art that I've found so far. Please do send me pointers to more recently available materials if you're in the know. I figure NDIPP probably has some recent information too and I'll be sure to look there next.

The AHDS surveyed 92 individuals and organizations working in digital audiovisual preservation. Most respondents indicated that their focus was on practical work flow management issues like data capture, access/dissemination, metadata, and rights management. They conclude that preservation work is at an early stage of development. Much remains to be done.

The big take-aways I get from the article:

  • there's no sustainable funding model in place

  • organizations and institutions can't do digital preservation alone. There needs to be wider collaboration in order to achieve economies of scale

  • there's no consensus on best practices and standards especially regarding file formats and codecs (the report recommends JPEG 2000 and MFX as the best lossless format and wrapper for digital masters – but they do think there needs to be more research comparing METS and MPEG-21)

  • more tools need to be built for automated metadata extraction

  • technical obsolescence is still the greatest threat to long-term preservation of A/V materials.

It's pretty much the same conclusions that I took away from DigCCurr2007. No surprises.

It's a bit discouraging to see that the same big issues that existed in 2001 still exist today. It is encouraging, however, that big funding agencies like JISC and the NSF are on the case, and that smart folks like Reagan Moore are building the tools to assist with wide scale distributed collaboration on digital preservation. I suspect that digital video/multimedia preservation is going to be all about the Grid. That's the key research area to watch, I think.

Labels: , , ,

posted by Laura @ 5:15 PM

1 Comments:

At 9/10/07 12:49 PM , Blogger Kara said...

Don't know exactly how I stumbled on your blog this afternoon, but definitely took an interest to this post. I work on the NDIIPP Preserving Digital Public Television project, and also have been teaching digital preservation of moving image and sound. You are absolutely right, things haven't come very far at all since 2001. It is indeed frustrating. MXF and J2K are being sold to moving image archivists as the savior for digial video and cinema preservation, but even these are still just not ready. There is not enough vendor adoption for J2K for it to be truly sustainable at this point. And MXF has so many vastly different flavors that files are hardly interoperable between two machines in the same building. Luckily there are a lot of dedicated people working on these issues, but the results are slow coming.

The metadata issue for these digital objects is almost even more frustrating than the essence format issues. We are working hard on PBCore, but at this point it's just not ready. In the meantime, there really isn't a standard out there that adequately captures the necessary metadata for preservation video. Audio is getting closer with the development of AES 098 (due out soon).

 

Post a Comment

Links to this post:

Create a Link

<< Home